UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Connecting natural space exposure to mental health outcomes across Vancouver, Canada Rugel, Emily Jessica 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_may_rugel_emily.pdf [ 1.36MB ]
Metadata
JSON: 24-1.0377727.json
JSON-LD: 24-1.0377727-ld.json
RDF/XML (Pretty): 24-1.0377727-rdf.xml
RDF/JSON: 24-1.0377727-rdf.json
Turtle: 24-1.0377727-turtle.txt
N-Triples: 24-1.0377727-rdf-ntriples.txt
Original Record: 24-1.0377727-source.json
Full Text
24-1.0377727-fulltext.txt
Citation
24-1.0377727.ris

Full Text

CONNECTING NATURAL SPACE EXPOSURE TO MENTAL HEALTH OUTCOMES ACROSS VANCOUVER, CANADA  by  Emily Jessica Rugel BA (magna cum laude), New York University, 1999 MPH, Portland State University, 2008   A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Population and Public Health)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  March 2019  © Emily Jessica Rugel, 2019   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Connecting Natural Space Exposure to Mental Health Outcomes Across Vancouver, Canada  submitted by Emily Jessica Rugel in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Population and Public Health  Examining Committee: Dr. Michael Brauer Co-supervisor Dr. Sarah Henderson Supervisory Committee Member Dr. Mariana Brussoni University Examiner Dr. Cecil Konijnendijk  University Examiner  Additional Supervisory Committee Members: Dr. Richard Carpiano Supervisory Committee Member   iii  Abstract In an increasingly urbanized world, identifying evidence-based strategies to guide the design and maintenance of healthy cities is an essential public health function. Two pressing urban health concerns are high rates of mental disorders and low levels of social connection. Epidemiological studies indicate that access to natural space – either greenspace, such as parks and street trees, or bluespace, such as oceans and lakes – may strengthen social connections and improve mental health. However, gaps remain regarding effects of specific forms of nature, their impacts on objective measures of mental health, and pathways by which any benefits occur. To address these gaps, this dissertation developed and applied a robust model of the presence, form, accessibility, and quality of greenspace and bluespace across the Vancouver, Canada region. This Natural Space Index (NSI) included more than 50 measures at 100-to-1,600-meter buffers for 60,000-plus six-digit postal codes. Analyses based on residential addresses highlighted the extent to which distinct measures result in different assessments, particularly in comparison with standard metrics of surrounding greenness such as the Normalized Difference Vegetation Index (NDVI). Using data from the Canadian Community Health Survey-Mental Health, the percentage of publicly accessible neighborhood nature within 500m had indirect mental health benefits via increased neighborhood social cohesion: each 1% increase was associated with 3-5% increases in reporting higher levels of social cohesion. In turn, individuals with the highest social cohesion had an 86% decrease in the odds of major depressive disorder, a 91% decrease in negative mental health, and a 2.8-point reduction in psychological distress (on a 0-40 scale). When the same question was approached using data on prescriptions related to mental illness, a 0.1-point increase in 250-meter surrounding greenness was linked to a 2% decrease in total psychotropic prescription dispensation and a 3% decrease in antidepressant prescription dispensation. The presence of ten additional street trees within 100m was associated with a 4% reduction in total psychotropic prescriptions. Although many NSI measures showed no association with mental health outcomes, the indirect iv  and direct effects identified by this thesis support calls for expanding equitable access to natural space as part of a broader healthy-city strategy.  v  Lay Summary In cities, nature such as trees and rivers may play an important role in connecting communities and improving mental health. The role of nature is important because one in five Canadians lives with mental illness, costing the national economy at least $50 billion per year. This research aimed to create a better model of the types of nature that residents of Vancouver, Canada can access, and to evaluate how nature impacts their mental health. Although not all forms of nature were linked to benefits, the findings indicate that street trees helped to lower the chance of using antidepressants or related medications. Also, having public parks or beaches in your neighborhood can increase social connectedness, which can then reduce the chance of major depression or negative mental health, as well as lowering levels of distress. These results suggest roles for urban nature to strengthen social ties and improve mental health. vi  Preface The work included here is my own, with guidance from my thesis committee members – Dr. Richard Carpiano and co-supervisors Dr. Michael Brauer and Dr. Sarah Henderson – across all stages of the research process, from initial conception to final submission for publication. Additional research contributions are described in detail below.  Identification, design, and performance of the research program For Chapter 1, fellow students in the University of British Columbia’s interdisciplinary Bridge program – Jason Curran, Rod Knight, and Sarah Partanen – contributed to an early version of the underlying literature review by preparing summaries of some of the cited research articles. The published version received feedback and guidance from Dr. Helen Ward and Dr. Sophie Verhille of the National Collaborating Centre for Environmental Health, as well as from Dr. Michael Brauer. I designed and conducted the final evidence review, interpreted the results, and wrote and edited the manuscript, representing a total contribution to the work as a whole of 90%. The work of Jason Curran, Rod Knight, and Sarah Partanen in preparing initial article summaries represents a contribution of 1% each; Dr. Brauer’s feedback on the wording of later versions of the manuscript represents a contribution of 1%; the work of Dr. Verhille in suggesting improvements to the manuscript outline and search strategy represents a contribution of 2%; and the work of Dr. Ward in suggesting improvements to the manuscript outline and search strategy as well as proposing revisions to various later drafts represents a contribution of 4%.  For Chapter 2, Paul Lesack of the University of British Columbia’s Walter C. Koerner Library offered feedback on working with NASA satellite data within a geographic information system. I developed the research concept; identified, collected, and processed the data, including carrying out all park quality appraisals; conducted all geospatial and statistical analyses; vii  interpreted the results; and wrote and edited the manuscript, representing a total contribution to the work as a whole of 84.5%. Paul Lesack’s feedback represents a contribution of 0.5%; Drs. Brauer, Carpiano, and Henderson provided guidance on the development of the original research concept as well as suggesting revisions to various drafts, representing a contribution of 5% each.  For Chapter 3, Population Data BC served as the research liaison, securing permission and data access from the data stewards at the British Columbia Ministry of Health, BC PharmaNet, and Statistics Canada. I designed the study, conducted all data preparation and analyses, interpreted the results, and wrote and edited the manuscript, comprising a total contribution to the work as a whole of 83%. Population Data BC’s role as the research liaison represents a contribution of 2%. Drs. Brauer and Henderson provided guidance on initial cohort development and case ascertainment as well as on analytic approaches; along with Dr. Carpiano, they also made a number of suggested revisions during the writing and editing of the manuscript; together, these activities represent a contribution of 6% each for Drs. Brauer and Henderson and 3% for Dr. Carpiano. All inferences, opinions, and conclusions drawn in this research are those of the authors, and do not reflect the opinions or policies of the data stewards.    For Chapter 4, the research and analysis are based on data from Statistics Canada and the opinions expressed do not represent the views of Statistics Canada. Data access to the 2012 Canadian Community Health Survey-Mental Health was granted via Statistics Canada’s Research Data Centres (RDC) program, and assistance formatting data and preparing analytic output to be vetted was provided by RDC analysts Cheryl Fu, Lee Grenon, and Wendy Kei. In addition, Dr. Mieke Koehoorn of the University of British Columbia’s School of Population and Public Health provided guidance on preparing data for use at the RDC. I designed the study, conducted all data preparation and analyses, interpreted the results, and wrote and edited the viii  manuscript, representing a total contribution to the work as a whole of 80%. The assistance formatting data and preparing output for vetting purposes provided by Cheryl Fu, Lee Grenon, and Wendy Kei represents a contribution of 0.5% each, as does the guidance on data preparation provided by Dr. Koehoorn. Dr. Carpiano proposed the approach to mediation described in this chapter, identified additional citations to the literature on social ties and health, and suggested revisions during the writing and editing of the manuscript, representing a total contribution of 8%. Drs. Brauer and Henderson offered ongoing feedback on analytic approaches along with providing input on later versions of the manuscript, representing a contribution of 5% each.       For Chapter 5, I wrote and edited the manuscript, which received suggestions for improvement from Drs. Brauer, Carpiano, and Henderson; all in all, my contribution was 85% and they each made a contribution of 5%.  The Walk Score® data used in Chapters 3 and 4 were provided by Redfin Real Estate.   Publications Four research chapters (Chapters 1 through 4) were drafted in part or whole as manuscripts for peer-reviewed publication. I prepared all initial versions of these chapters and then submitted them for two to three rounds of iterative review, either to all thesis committee members (Chapters 2 through 4) or to at least one thesis committee member and a team of external scientific reviewers (Chapter 1). Some content, including sections that appeared in the publications listed below, has been moved among chapters to support readability.  A version of Chapter 1 has been published: Rugel E. Green Space and Mental Health: Pathways, Impacts, and Gaps. National Collaborating Centre for Environmental Health; 2015 ix  Mar. http://www.ncceh.ca/sites/default/files/Full_Review-Greenspace_Mental_Health_Mar_2015.pdf.  A version of Chapter 2 has been published: Rugel EJ, Henderson SB, Carpiano RM, Brauer M. Beyond the Normalized Difference Vegetation Index (NDVI): Developing a Natural Space Index for population-level health research. Environmental Research. 2017 Nov;159:474-83. doi: 10.1016/j.envres.2017.08.033.  A version of Chapter 4 has been published: Rugel EJ, Carpiano RM, Henderson SB, Brauer M. Exposure to natural space, sense of community belonging, and adverse mental health outcomes across an urban region. Environmental Research. 2019 Apr;171:365-77. doi: 10.1016/j.envres.2019.01.034.  Ethics Approvals No ethics approvals were required or sought for Chapters 1 and 2. Initial and ongoing ethics approvals for Chapter 3 (H15-01212) and Chapter 4 (H15-01437) were obtained from the University of British Columbia’s Behavioural Research Ethics Board starting on June 19, 2015.   x  Table of Contents Abstract ....................................................................................................................................... iii Lay Summary ............................................................................................................................... v Preface ......................................................................................................................................... vi Table of Contents ........................................................................................................................ x List of Tables ............................................................................................................................ xiv List of Figures ........................................................................................................................... xvi List of Abbreviations ............................................................................................................... xvii Glossary .................................................................................................................................. xviii Acknowledgements .................................................................................................................. xix Dedication ................................................................................................................................ xxii Chapter 1: Introduction ............................................................................................................... 1 1.1 Defining “natural space” in the urban context ............................................................... 5 1.2 Pathways linking natural space to mental health .......................................................... 7 1.2.1 Direct psychophysiological responses ................................................................ 10 1.2.2 Increased rates and enhanced benefits of physical activity ................................ 13 1.2.3 Facilitated social contact ..................................................................................... 14 1.2.4 Buffered impacts of noxious urban exposures .................................................... 15 1.3 Challenges in natural space exposure assessment at the metropolitan scale ........... 16 1.4 Dissertation rationale and objectives .......................................................................... 18 1.5 Dissertation setting ..................................................................................................... 18 1.6 Dissertation structure .................................................................................................. 20 1.6.1 Developing a model of natural space exposure for the Vancouver CMA ........... 20 1.6.2 Assessing the relationship between natural space exposure and cases of incident psychotropic prescription dispensation ................................................................. 21 1.6.3 Exploring the role of sense of community belonging along the pathway linking natural space exposure to mental health ............................................................................ 21 Chapter 2: Developing a model of natural space exposure for the Vancouver census metropolitan area ...................................................................................................................... 22 2.1 Background ................................................................................................................. 22 2.1.1 Objectives ........................................................................................................... 25 2.2 Materials and methods ............................................................................................... 25 2.2.1 Data collection and measures ............................................................................. 25 2.2.1.1 Natural space presence (“greenness”) ........................................................... 25 xi  2.2.1.2 Natural space form .......................................................................................... 26 2.2.1.3 Natural space access ...................................................................................... 28 2.2.1.4 Natural space quality ...................................................................................... 28 2.2.2 Statistical analysis ............................................................................................... 30 2.3 Results ........................................................................................................................ 32 2.3.1 Natural space presence (“greenness”) ............................................................... 32 2.3.2 Natural space form .............................................................................................. 32 2.3.3 Natural space access .......................................................................................... 34 2.3.4 Natural space quality .......................................................................................... 34 2.3.5 Natural space measure comparisons ................................................................. 35 2.3.6 Principal component analysis ............................................................................. 37 2.4 Discussion .................................................................................................................. 38 2.4.1 Limitations ........................................................................................................... 42 2.5 Conclusions ................................................................................................................ 43 Chapter 3: Assessing the relationship between natural space exposure and cases of incident psychotropic prescription dispensation .................................................................. 45 3.1 Background ................................................................................................................. 45 3.2 Materials and methods ............................................................................................... 47 3.2.1 Measures ............................................................................................................ 47 3.2.1.1 Natural space exposure measures ................................................................. 47 3.2.1.2 Mental health outcomes .................................................................................. 50 3.2.1.3 Covariates ....................................................................................................... 51 3.2.2 Study sample ...................................................................................................... 52 3.2.2.1 Cohort development ........................................................................................ 52 3.2.2.2 Natural space exposure and covariate assignment ........................................ 53 3.2.2.3 Analytic sample development ......................................................................... 54 3.2.3 Statistical analysis ............................................................................................... 55 3.3 Results ........................................................................................................................ 56 3.3.1 Sample characteristics ........................................................................................ 56 3.3.2 Natural space presence, form, accessibility, and quality .................................... 60 3.3.3 Associations between natural space measures and psychotropic prescription dispensation ....................................................................................................................... 63 3.4 Discussion .................................................................................................................. 72 3.4.1 Limitations ........................................................................................................... 77 xii  3.5 Conclusions ................................................................................................................ 80 Chapter 4: Exploring the role of sense of community belonging along the pathway linking natural space exposure to mental health ................................................................................ 81 4.1 Background ................................................................................................................. 81 4.2 Materials and methods ............................................................................................... 84 4.2.1 Mental health data .............................................................................................. 84 4.2.2 Study sample ...................................................................................................... 84 4.2.3 Measures ............................................................................................................ 85 4.2.3.1 Natural space exposure measures ................................................................. 85 4.2.3.2 Mental health outcomes .................................................................................. 87 4.2.3.3 Mediation ........................................................................................................ 88 4.2.3.4 Covariates ....................................................................................................... 89 4.2.4 Statistical analysis ............................................................................................... 90 4.2.4.1 Assessment of mediation ................................................................................ 91 4.3 Results ........................................................................................................................ 91 4.3.1 Sample characteristics ........................................................................................ 91 4.3.2 Natural space presence, form, and accessibility ................................................. 93 4.3.3 Mental health outcomes and sense of community belonging ............................. 94 4.3.4 Associations between natural space measures and mental health outcomes .... 95 4.3.5 Associations between natural space measures and sense of community belonging. ........................................................................................................................... 98 4.3.6 Associations between sense of community belonging and mental health outcomes. ......................................................................................................................... 101 4.3.7 Testing the indirect effect of sense of community on mental health outcomes . 105 4.4 Discussion ................................................................................................................ 106 4.4.1 Limitations ......................................................................................................... 113 4.5 Conclusions .............................................................................................................. 115 Chapter 5: Conclusion ............................................................................................................ 116 5.1 Research summary and contributions ...................................................................... 116 5.1.1 Advances in natural space exposure mapping and modeling ........................... 119 5.1.2 Expanding epidemiological assessments to objective outcomes ..................... 120 5.1.3 Elucidating pathways via mediation analyses ................................................... 122 5.2 Implications for urban design and policy development ............................................. 123 5.2.1 Methodological limitations ................................................................................. 123 xiii  5.2.2 Recommendations for future research and collaboration ................................. 127 Bibliography ............................................................................................................................. 131 Appendix A – Natural Space Index Variable Table .............................................................. 160 Appendix B – World Health Organization Anatomical Therapeutic Chemical (ATC) Classification System Codes and Corresponding BC PharmaNet Drug Identification Number/Product Identification Number ................................................................................ 162 Appendix C – 2012 Canadian Community Health Survey-Mental Health Questionnaire ..... 203  xiv  List of Tables Table 2.1 Public Open Space Desktop Audit Tool (POSDAT) quality appraisal measures ....... 30 Table 2.2 Descriptions and comparisons of satellite-based measures ...................................... 33 Table 2.3 Principal component analysis results ......................................................................... 37 Table 3.1 Summary of sociodemographic factors in the Natural Space Index (NSI) sample ..... 57 Table 3.2 Summary of sociodemographic factors in the POSDAT quality sample ..................... 58 Table 3.3 Summary of sociodemographic factors in the street tree sample ............................... 59 Table 3.4 Summary of natural space exposures in the NSI sample ........................................... 61 Table 3.5 Summary of natural space exposures in the POSDAT sample .................................. 62 Table 3.6 Summary of natural space exposures in the street tree sample ................................ 62 Table 3.7 Comparisons among continuous natural space exposures ........................................ 63 Table 3.8 All-dispensation results for the NSI sample ................................................................ 64 Table 3.9 Antidepressant results for the NSI sample ................................................................. 65 Table 3.10 Anxiolytic results for the NSI sample ........................................................................ 66 Table 3.11 Sensitivity analysis results for the NSI sample ......................................................... 68 Table 3.12 Complete results for the POSDAT sample ............................................................... 69 Table 3.13 Complete results for the street tree sample .............................................................. 70 Table 3.14 Sensitivity analysis results for the complete-cases sample ...................................... 71 Table 4.1 Summary of sociodemographic factors in the Canadian Community Health Survey (CCHS-MH) sample .................................................................................................................... 92 Table 4.2 Summary of natural space exposures in the CCHS-MH sample ................................ 94 Table 4.3 Adjusted results for natural space exposures and mental health outcomes for the CCHS-MH sample ...................................................................................................................... 96 Table 4.4 Crude results for natural space exposures and mental health outcomes for the CCHS-MH sample ................................................................................................................................. 97 Table 4.5 Adjusted results for natural space exposures and sense of community belonging for the CCHS-MH sample ................................................................................................................ 99 Table 4.6 Crude results for natural space exposures and sense of community belonging for the CCHS-MH sample .................................................................................................................... 100 Table 4.7 Adjusted results for sense of community belonging and mental health outcomes for the CCHS-MH sample .............................................................................................................. 102 Table 4.8 Crude results for sense of community belonging and mental health outcomes for the CCHS-MH sample .................................................................................................................... 103 xv  Table 4.9 Multiply adjusted results for sense of community belonging and mental health outcomes for the CCHS-MH sample ........................................................................................ 104 Table 4.10 Results of zMediation test for the CCHS-MH sample ............................................. 106 Table 5.1 Summary of results from multiple analyses .............................................................. 118  xvi  List of Figures Figure 1.1 Vancouver census metropolitan area (CMA) study location ..................................... 20 Figure 2.1 Natural space forms across the Vancouver CMA ..................................................... 34 Figure 2.2 Normalized Difference Vegetation Index versus alternate measures ....................... 36 Figure 2.3 Principal component analysis biplot .......................................................................... 38 Figure 3.1 Analytic sample development for the case-control study .......................................... 55 Figure 4.1 Analytic sample development for the Canadian Community Health Survey-Mental Health (CCHS-MH) study ........................................................................................................... 85 Figure 4.2 Direct and indirect relationships among exposures, mediators, and outcomes for the CCHS-MH study ....................................................................................................................... 108 xvii  List of Abbreviations  ART   Attention Restoration Theory ATC   Anatomical Therapeutic Chemical CCHS-MH  Canadian Community Health Survey-Mental Health  CMA  Census Metropolitan Area DA  Dissemination Area DIN/PIN  Drug Identification Number/Product Identification Number EVI   Enhanced Vegetation Index GHQ-12 12-item General Health Questionnaire GIS   Geographic Information Systems K10   Kessler-10 Psychological Distress Scale LiDAR   Light Detection and Ranging MHC-SF  Mental Health Continuum-Short Form  MDD   Major Depressive Disorder MODIS  Moderate Resolution Imaging Spectroradiometer MSP  Medical Services Plan of British Columbia NDVI   Normalized Difference Vegetation Index NMH   Negative Mental Health NSI   Natural Space Index PCA   Principal Component Analysis PCCF   Postal Code Conversion File  POSDAT  Public Open Space Desktop Audit Tool SES   Socioeconomic Status SRT   Stress Reduction Theory YLD  Years Lived with Disability  xviii  Glossary Bluespace  An area containing water – such as an ocean, lake, river, or decorative fountain – which may fall anywhere on the spectrum from completely wild to fully human-designed. Environmental justice  A social movement and policy doctrine that seeks to ensure equity in the development and implementation of environmental guidelines as well as in the siting and provision of features with environmental impacts. Greenness   A satellite-based measure of greenspace based on surface reflectance. Greenspace   An area containing or comprising vegetation – such as a park, garden, or street tree – that may fall anywhere on the spectrum from completely wild to fully human-designed.   Natural space   A category that comprises both bluespace and greenspace, based on the definitions above.  Public natural space  A natural space, per the definition above, that is fully accessible to the public. Public open space  A publicly accessible area – such as a public plaza or playground – that is generally free of buildings or other large structures, and which may or may not incorporate bluespace or greenspace.  Shinrin-yoku  A Japanese term that may be translated into “forest bathing” in English and which refers to a traditional, multisensory, therapeutic practice that takes place in a forest.  Social capital   Potential or readily available resources that derive from individual social ties or broad networks. Social support  Instrumental, emotional, and informational functions performed by significant others, such as family and friends, for the benefit of an individual. Streetscape greenery  A form of vegetation that is located along a street or sidewalk, including street trees, grasses, and shrubbery, among other types. Walkability  The set of features and characteristics of an area that may promote or impede the movement of pedestrians (including individuals using mobility devices such as wheelchairs).  xix  Acknowledgements I wrote this dissertation on UBC’s Point Grey campus, which sits on the traditional, ancestral, and unceded territory of the xʷməθkʷəy̓əm (Musqueam) First Nation. These Coast Salish Peoples – along with the Skwxwú7mesh (Squamish), Stó:lō, and Səl̓ílwətaʔ/Selilwitulh (Tsleil- Waututh) Nations – have been and continue to be responsible for protecting and preserving the natural spaces at the heart of this work, and I am deeply grateful for their stewardship.  I decided to focus on natural spaces and health due to UBC’s interdisciplinary Bridge Program, so I am thankful for this spark of inspiration, as well as for the program’s funding, training, and mentorship. Getting to work and study alongside fellow Bridgies Ther Aung, Jason Curran, Bojosi Gamontle, Amy Hall, Andrea Jones, Rod Knight, Sarah Partanen, and Jackie Yip was especially rewarding. The program’s success was due in no small part to the organizational prowess and general good cheer of Linda Bonamis, who has my appreciation.         A number of other UBC sources also helped fund my PhD, including the Four-Year Fellowship, the George R.F. Elliot Fellowship in Community Health, and the Faculty of Medicine Graduate Award. In addition to providing financial support, the Public Scholars Initiative created an encouraging community and offered ongoing reminders to consider the ways my research could contribute to the larger world beyond my little office.   Serving as a teaching assistant for the Urban Forests and Well-being course co-taught by Sara Barron and Matilda van den Bosch showed me what interdisciplinarity looks like in practice, while introducing me to the frustrations and fulfillment of post-secondary education. Each was also a critical source of assurance regarding the direction and potential importance of this research, and their perspectives on landscape design and environmental psychology, respectively, greatly strengthened this document. xx  The Bridge Program brought me to UBC, but co-supervisors Michael Brauer and Sarah Henderson allowed me to complete my PhD. They demonstrated how to serve in dual roles as an academic and a practitioner; embodied supportive supervision; and provided the ideal blend of constructive criticism, insightful analysis, and spirited discussion. Many of these discussions took place during lab meetings, and I’m thankful to them for organizing these sessions and to my fellow labmates Raphael Arku, Jesse Cooper, Anders Erickson, Hind Sbihi, Matt Shupler, Matt Wagstaff, Angela Yao, Jessica Yu, and Weiran Yuchi for participating in them. Outside of the lab, classmates Stephanie Harvard, Naz Islam, Alina McKay, Dimi Panagiotoglou, and Mint Ti were vital sources of camaraderie and encouragement.  I began working with committee member Richard Carpiano before my thesis even took shape, and his guidance on social epidemiology and complex modeling provided a solid foundation for these components of this work, allowing me to explore novel pathways with confidence.   I am truly grateful to all three of my supervisory committee members for their unique perspectives and tireless dedication, particularly during the final stages of drafting this dissertation.  I am also very appreciative of the individuals who volunteered to participate in the 2012 Canadian Community Health Survey-Mental Health, and to Statistics Canada staff members Cheryl Fu, Lee Grenon, and Wendy Kei, who supported my work on the data from this survey.  I’m fortunate to have four parents, and they’ve all inspired my career in different ways, while never questioning the many forms it has taken. The knowledge of their unwavering belief in me has been even more important than the practical advice and relaxing interludes away from academia they’ve provided on both coasts of this great continent over the years.  xxi  Finally, to my husband Flail – I wouldn’t have been able to achieve half of what I have without your always-ardent embrace of every new opportunity life brings, especially those that end up taking long and winding paths like this one. Completing the chapters below was an adventure filled with both pitfalls and pleasures, but many of the latter are due to your love and support.  xxii  Dedication To all those who gather  under silent trees, below roaring overpasses, next to flowing rivers  To unite in that oldest of human activities, dancing to bass and delighting in each other’s company      1  Chapter 1: Introduction Mental illnesses are a significant and growing cause of poor health and early death worldwide, with the global burden of such diseases increasing by more than one-third between 1990 and 2010 (1). In 2012, the last year for which nationwide survey data were available, more than nine million Canadians reported having experienced a mental illness at some point during their lifetimes (2). The most commonly reported conditions were mood disorders, including major depressive disorder (MDD) and bipolar disorder (among 5.4% of the population); substance use disorders (4.4%); and generalized anxiety disorder (2.6%) (3). Together, this means that one out of every five Canadians is living with mental illness, as compared with one in 15 Canadians living with type 2 diabetes and one in 25 with heart disease (4). Based on an analysis of current demographic and epidemiological trends, these numbers are only expected to grow, with projections that nearly 4.9 million people in Canada will have a mood or anxiety disorder by 2041 (4).   Mental illnesses often result in significant detrimental effects on physical, emotional, and social well-being, making them a leading cause of years lived with disability (YLDs). In fact, despite a 4.9% decline in related age-standardized YLDs between 2006 and 2016, MDD alone remains the second major cause of YLDs in high-income North America, according to the 2016 Global Burden of Disease study (5). Along with their deleterious impact on individuals and families, mental disorders also place a high burden on the broader economy, due in part to their association with high rates of unemployment and disability. In a nationwide sample of Canadians, individuals with MDD had a combined unemployment-disability rate of nearly one-third; within this sample, residents of British Columbia with MDD had the highest disability rate (14.9%) and the second-highest unemployment rate (19.3%) of those with MDD in any province (6). Integrating this lost productivity along with factors such as the costs of providing medical    2  treatment and social support, the Mental Health Commission of Canada estimates that mental disorders cost the national economy at least $50 billion per year (4).          In light of their prevalence and impact, identifying population-level strategies to prevent or mitigate symptoms of mental disorders is critical. Unfortunately, both individual and population-level strategies to address the burden of mental illness remain less than ideally effective, with more than 500,000 Canadians reporting an unmet need for mental health care in 2012 (3).  Responding to this need, the Mental Health Commission of Canada launched a national mental health strategy in 2012, calling for "governments to take a comprehensive approach to addressing mental health needs" (7). One aspect of such a comprehensive approach requires clarifying the impact of environmental features on the development, progression, and treatment of mental illness. Recognizing the need to focus additional attention on the ways in which the built and natural environments shape public health, the Chief Public Health Officer's Report on the State of Public Health in Canada for 2017 states that “targeted and hypothesis-driven research, standardized data collection and systematic evaluations of the health impact of community design features are needed” (8). Environmental factors may be particularly important in the development of depression (9), with recent studies pointing to the immune and neuroendocrine systems as biological mediators along the pathway between the environment and stress, anxiety, and depression (10,11).  One specific environmental exposure that is particularly relevant to urban policy and planning is natural space, both “greenspace” such as parks and streetscape greenery, and “bluespace” in the form of rivers, oceans, and lakes. Over the past few decades, a large body of research has explored associations between natural space and a wide range of mental health outcomes, both across large urban or national populations and among individuals with pre-existing mental illness. However, among the reviews that have attempted to summarize the evidence base in    3  this area, the majority report fairly weak or inconsistent effects overall, pointing to factors such as the heterogeneity of exposure definitions and metrics (12,13) and the coarseness of many individual measures (14); a lack of attention to potential moderating factors such as socioeconomic status (SES), age, and gender (13,15–17); and the need to identify and embed exposure measures with relevance to specific pathways and policies (17–19).  Among the individual studies that have focused on population-level effects, a nationally representative longitudinal survey of households in the United Kingdom indicated that more publicly accessible greenspace and private gardens in neighborhoods reduced mental distress and increased life satisfaction (20). A nationwide cross-sectional study in Scotland found that individuals who reported the greatest lack of environmental assets (including parks and playgrounds) in their neighborhoods were almost twice as likely to report anxiety and depression as residents who were not so deprived (21). Further highlighting the importance of perceptions of greenspace access, residents of Adelaide, Australia who rated their neighborhoods as highly green in a cross-sectional survey – based on park and path access, streetscape greenery, and other pleasant natural features – had almost twice the odds of being in better overall mental health (22).   Not all of the results are in agreement, however, with a number of studies reporting no associations at all. For example, a cross-sectional study that applied a range of natural space metrics to members of an existing cohort study in Barcelona, Spain found no statistically significant effects of the amount of greenspace within 300 or 500 meters of participants’ homes and any of the four mental health outcomes they studied, as well as no relationship between surrounding greenness and self-reported anxiety and depression (23). A cross-sectional study in England that integrated a similar measure of greenspace percentage also reported no impact on a mental well-being scale after appropriate adjustment for confounding (24). Looking    4  specifically at psychotropic medication prescribing – which comprises any drug prescribed to affect an individual’s mental state – an ecological study in England found no protective effects of either greenspace alone or total natural space on antidepressant prescription rates (25). In some instances, specific natural space measures have even been linked with poorer mental health. A cross-sectional study in Perth, Australia that assessed both park area and quality reported that higher-quality parks were actually associated with increased psychological distress (26). In an exploration of a range of built, natural, and social environment characteristics on both the prevalence and severity of mood disorders, no associations were seen with greenspace percentage, but increasing amounts of bluespace were linked to higher odds of having these conditions (27).  One consequence of these conflicting findings is the fact that although a growing number of municipalities in Canada (28,29) and abroad (30,31) have committed to providing accessible, high-quality natural space to residents, these plans are generally driven by concerns about environmental sustainability or climate change adaptation and mitigation (31,32), rather than their potential public health impacts (15). Additional evidence about the relationship between natural space and mental health is necessary to embed health in these policies, and to further support the development of communities that are both healthy and sustainable. Given that socioeconomically deprived communities may have less and lower-quality natural space, as observed in multiple locations around the globe (33–35), ensuring equitable access is also a component of the broader issue of environmental justice.   Environmental justice – which is both a social movement and a policy principle – has often been narrowly defined as focused on reducing the disproportionate burden of impacts faced by particularly vulnerable communities with respect to the siting of noxious environmental polluters (36). A seminal article by Bullard (1996) highlights the range of compositional factors that can    5  make a community vulnerable, including gender, culture, and occupational status, along with race, ethnicity, and socioeconomic status (36). In addition, the article expands the list of potential impacts to include disparate enforcement of protective regulations, a lack of methodically rigorous risk assessments, discriminatory zoning policies, and exclusion from decision-making processes. It has taken substantially longer for environmental justice to expand to include equity in the provision of beneficial resources such as natural space in its definition, but such an expansion is now well under way (37). The paradoxical potential for natural space improvements to end up exacerbating existing inequities has also been identified, with calls to make cities “just green enough”, meaning green enough to provide health benefits, but not so green as to result in gentrification and associated displacement (38).The amount and types of urban nature that are sufficient to provide health benefits without risking these harms has yet to be identified, however, further highlighting the importance of research linking specific exposure metrics to health outcomes.  1.1 Defining “natural space” in the urban context One issue that complicates any systematic inquiry into the relationship between urban nature and population health is the diversity of approaches to defining and describing “natural space”. The terms “nature” (39–47), “naturalness” (48,49), or “natural space” (21,50) have been employed by some researchers when including both bluespace and greenspace, or to highlight the fact that not all vegetation is green, depending on the season and location (48). Areas that are restricted in some way, such as private gardens or golf courses, may be distinguished from those that are accessible to all, via terms such as “public open space” (51) or “public natural space” (52). Some studies focus narrowly on a specific form of greenspace, such as a playground (21), garden or arboretum (44,53–56), forest (39,42,43,57–64), park (50), or “streetscape greenery”, trees and other plantings along streets (65). In many instances, exactly what is meant by any individual term is unspecified, with a recent systematic review of 125    6  greenspace research studies conducted in 22 countries noting that less than half clearly described what they meant by greenspace (13).  “Surrounding greenness” is often used to denote the simplest objective measure of greenspace, the satellite-based Normalized Difference Vegetation Index (NDVI). “Greenness” is appropriately used with respect to this measure (66–68) because of the way it is calculated, based on the following formula: NDVI = (near-infrared radiation – visible radiation)/(near-infrared radiation + visible radiation) (69). Essentially, NDVI uses the different wavelengths of light absorbed by vegetation in comparison with hard surfaces such as concrete or bare land to produce a value ranging from -1 to 1, with surfaces such as rock and snow having values below 0.1, grassy areas falling in the range of 0.2 to 0.3, and healthy, high-density vegetation such as forests resulting in values over 0.6 (69). A related measure is the Enhanced Vegetation Index (EVI), which covers the same range of values (-1 to 1), but is a more recent measure that was designed to reduce reflection distortions (69) and is less likely to become oversaturated in areas of high-density vegetation (70,71). Exactly how well either NDVI or EVI performs as a proxy for greenspace is unclear, however, because attempts to directly compare NDVI with alternate measures are rare and have produced conflicting results. An Australian study compared participant perceptions of surrounding greenness within 400 meters of their homes with NDVI values in the same buffer, and found negative agreement between the two measures, with participants in areas with lower NDVI actually perceiving them as having higher greenness (72). Conversely, an effort in Seattle, Washington that asked environmental psychologists to rate the overall greenness within a slightly smaller one- to two-block radius around residences found strong correlation between these appraisals and NDVI values (66). In addition, neither NDVI nor EVI can be used to differentiate among all forms of natural space or to identify the accessibility or quality of a specific parcel (73).      7  In this dissertation, nature is defined much more broadly, and more specifically, consistent with recommendations for interdisciplinary research (13). As noted in the glossary, “natural space” is a category that comprises both bluespace and greenspace. “Greenspace” is defined as an area containing or comprising vegetation – including a park, garden, or street tree – that may fall anywhere on the spectrum from completely wild to fully human-designed. “Bluespace”, on the other hand, refers to an area containing water – such as an ocean, lake, river, or decorative fountain – and which may similarly reflect varying degrees of human development or maintenance.  1.2 Pathways linking natural space to mental health The use of relatively broad measures of natural space that encompass multiple possible mechanistic pathways, such as NDVI or EVI, also provides a potential explanation for the inconsistent findings regarding the impacts of nature on mental health. Although very few studies of natural space exposure and mental health outcomes have been designed to address causality, I developed an evidence review on behalf of the National Collaborating Centre for Environmental Health in 2015 in order to describe the principal pathways linking natural space to mental health and to summarize the evidence supporting these suggested mechanisms (74). Searches of CINAHL, Embase, MEDLINE, Psycinfo, the Science Citation Index, and the Social Science Citation Index were conducted in December of 2013 and January of 2014, with no restriction on the original publication date of studies. Search terms and keywords were selected to align with two principal domains: natural space and mental health. To capture the first domain, keywords included: greenery, greenness, green space, natural space, natural view, open space, park, playground, garden, trees, and forest. In order to locate studies related to mental health, terms included: mental health, mental well-being, mental illness, mental disorder, psychological, psychosocial, depression, anxiety, stress, bipolar, schizophrenia, personality disorder, and obsessive-compulsive disorder. Boolean logic was integrated to combine the two    8  constructs and to avoid the inclusion of irrelevant results (e.g., “trees.mp. NOT decision tree.mp.”). Where possible, controlled vocabularies (such as the Medical Subject Headings created for use in MEDLINE) were employed along with keywords. In addition, wildcards were integrated to help account for variability in spellings (such as “green space” versus “greenspace”). In addition, a set of evidence-based search filters developed by the Scottish Intercollegiate Guidelines Network (SIGN) were incorporated into the searches conducted within MEDLINE and Embase in order to improve the specificity of the search.  No limits were applied with respect to the date of publication and articles published in non-English languages were included as part of the abstract-review process, but were not incorporated into the final analysis due to a lack of resources available for translation. In terms of study design, individual and cluster randomized controlled trials, cohort studies, case-control studies, and observational designs were all eligible for inclusion. Studies examining virtual exposure to green space in a laboratory setting were also included. Commentaries, editorials, and studies reported solely as abstracts (such as conference proceedings) were excluded. In addition, after reviewing abstracts, a decision was made to exclude studies on the mental health outcomes of gardening, because this was thought to merit an independent review.  After applying inclusion and exclusion criteria, 32 articles were selected for full-text review, with an additional 13 studies located through forward citation tracking and hand-searching. A total of three systematic reviews or meta-analyses, 19 experimental, one longitudinal, 17 cross-sectional, two mixed-methods, and four qualitative reports were evaluated. This evaluation identified four principal pathways linking natural space exposure to mental health: 1) direct psychophysiological benefits; 2) increased rates and impacts of physical activity; 3) facilitated social contact, with concomitant improvements in social support and social capital; and 4)    9  buffered noxious exposures, such as noise and air pollution. These pathways may be both independent and interdependent, depending on the form of exposure and the outcome.   A number of other articles have summarized the body of evidence – although most often not with respect to mental health outcomes – detailing pathways that are generally in alignment with these four, although sometimes using different terms. For example, a review of reviews published in 2014 focuses specifically on air quality with respect to buffering, while also defining psychophysiological benefits more narrowly as stress reduction, although this construct also includes affective, cognitive, and physiological restoration, making it fairly similar to our description of this pathway (75). By contrast, a literature review of epidemiological studies on green infrastructure separates direct psychophysiological benefits into biophilia, emotional regulation, and stress reduction, while omitting consideration of social contact, and again focusing on air pollution as the primary noxious feature (76). Another review that focused on both physical and mental health cites physical activity as the primary pathway to improvements in physical health, highlights the importance of social ties for mental well-being, and addresses the potential for greenspace to reduce existing health inequities related to SES (15). Two reviews by a single team on the health effects of greenspace – one in 2015 (14) and an updated version in 2018 (17) – also highlight the potential for increased impacts among individuals with lower SES, while focusing on the roles of stress reduction, physical activity, social engagement, and noise and air pollution. In addition, these reviews summarize the findings regarding birth outcomes, highlighting the impacts these can have across the lifecourse. These two reviews also discuss the potential for greenspace to actually harm respiratory health via higher rates of allergies and asthma (14,17). Perhaps the clearest description in this area arose from an interdisciplinary workshop held in Germany in 2016 that explicitly focused on delineating pathways between greenspace and health. The resulting conceptual model describes three broad categories: 1) mitigation, including buffered air pollution, noise, and heat; 2) restoration,    10  which aligns closely with our concept of direct psychophysiological benefits; and 3) instoration, comprising encouraged physical activity and facilitated social cohesion (18). The workshop also called for additional research to examine these pathways via formal mediation testing.   Once again, however, the aim was to explore pathways to a multitude of health outcomes, rather than mental health specifically. The four-pathway model described below omits certain aspects that are primarily relevant to physical health – such as potential buffering via increased shade and a reduced urban heat island effect (77) – while maintaining a distinction between physical activity and social cohesion, which may be critical in light of qualitative research showing that even seemingly small features such as benches can affect perceptions of the suitability of a natural space for interaction and restoration, respectively (78).     1.2.1 Direct psychophysiological responses A substantial amount of research has focused on the underlying psychophysiological responses to natural space, both in controlled laboratory settings and outdoors. This line of inquiry is grounded in two distinct, though complementary, theories: the stress reduction theory (SRT) described by Ulrich (79), and the attention restoration theory (ART), first described by Rachel and Stephen Kaplan (80). In SRT, exposure to nature induces a relaxed psychological state marked by lower levels of stress (79). ART, on the other hand, proposes that natural environments contain elements that help individuals recover from the mental fatigue required to voluntarily direct attention to multiple tasks within their day-to-day lives (44). Both chronic stress (10) and stressful life events (9,11) are known risk factors for anxiety and depression. Difficulties with attention are central to a number of a mental health disorders, including schizophrenia (81) and attention deficit hyperactivity disorder (ADHD) (82). Although well described and commonly cited, these theories are not without their detractors, however. For example, a systematic review that sought to assess evidence for ART collected via experimental and quasi-experimental    11  studies reported meta-analyses showing benefits for only three of 14 included outcome measures, while also noting the lack of clarity regarding the tests best suited to capture attention capacity (83).  Overall, a large number of studies indicate that exposure to natural space improves individuals’ moods (20,41,43,44,47,55,56,58,60–62,84,85). For instance, a study among healthy university students that compared the effects of an hour-long walk in an urban setting to one in an arboretum found improvements, on average, in both mood and directed attention after participants walked in the more-natural setting (44). This finding was replicated in a study that used a similar within-subjects design but focused on adults with major depressive disorder, with the nature walk improving both positive mood and directed attention (43). These studies highlight the complementary nature of SRT and ART as theoretical frameworks, explaining the affective impacts and attentional effects, respectively.  The impact of natural space on levels of the stress hormone cortisol is mostly beneficial, with one study showing no effect (39), but others showing significant decreases in comparison with less natural settings (57,59,61,86,87) or with increased exposure to greenspace (86), and another showing reductions in cortisol, but only at certain times of day (60). Some of these differences may have arisen due to inappropriate accounting for diurnal patterns of cortisol expression (or cortisol slope) (57). A study in Scotland collected multiple cortisol samples over the course of a day and determined that people residing in neighborhoods with parks, forests, and other natural environments not only reported less stress, but also had healthier cortisol slope profiles, marked by levels that peaked shortly after waking and then declined over the course of the day (86). The mix of findings in this pathway may also reflect variations between population subgroups. For example, among studies looking at directed attention and    12  concentration, positive effects of access to greenspace were demonstrated among low-income children (88) and children with ADHD (50,89) but not among pregnant women (46).  Additional insights into this pathway have emerged from experimental research that takes advantage of newly developed technologies. Using the common experimental design of direct comparison between brief walks in natural versus more typically urban settings, Bratman and colleagues complemented self-reported measures of rumination (excessive over-thinking) with neuroimaging to visualize activity in the subgenual prefrontal cortex, which is the area of the brain associated with rumination (90). Following the nature walk alone, they found reduced rumination as measured by self-reports and shown via neuroimaging. Neuroimaging was also carried out with a sample of school-age children participating in a longitudinal study in Barcelona, which found that greater levels of lifelong surrounding greenness were tied to higher brain volume in areas associated with working memory and directed attention (91). Researchers have also used immersive virtual reality to explore the effects of exposure to the sights and sounds of nature on the endocrine system and autonomic stress responses, finding that both visual and audio representations of nature facilitated stress recovery, but had no effect on cortisol response (39).          Despite these advances, one critical methodological issue with the studies examining this pathway is the reliance on brief exposure times, such as a 15-minute viewing session in a lab (39) or an hour-long walk in a park (43,44,58,92). Larger effects over longer timeframes of up to a day have been observed, as demonstrated by a meta-analysis that assessed effects for a range of times spent being physically active in green environments (92). Another limitation of many of the studies focused on direct psychophysiological benefits is the narrow range of their participants, which are often drawn from homogenous populations such as university students (44,47,61) or men (39,60).    13  1.2.2 Increased rates and enhanced benefits of physical activity In the second pathway, natural space is seen as encouraging individuals to engage in physical activity (93), particularly walking (94), and exercise in more-natural settings is hypothesized to augment the mental health benefits of physical activity. Understanding this pathway is especially important because more-active individuals are less likely to experience symptoms of depression (95), and both outdoor and indoor physical activity have been successfully used as a treatment for depression among a wide range of populations (96,97). Physical activity carried out in more-natural environments may also enhance mental health benefits in comparison with exercise indoors or in urban settings (41). The evidence for an increase in physical activity rates in association with natural spaces is somewhat inconsistent, however. A study of 4,950 middle-aged adults in the United Kingdom found no association between access to or quality of greenspace and recreational physical activity such as bicycling, swimming, and tennis (98). A systematic review reported that only 40% of included studies found a positive association between greenspace and physical activity rates (93).   In addition to the type of exercise under study, the form and quality of natural space is an important consideration, with specific features such as trails and wooded areas seen as particularly conducive to physical activity in the general population (99). A systematic review of studies employing a mix of designs reported positive effects on mood for brief, one-time walks held outdoors in comparison with those indoors; two studies in which participants exercised indoors but viewed virtual-reality images of the outdoors were also included, and these found mood benefits as well (41). A stronger association was found with exercise in greenspace (40), or for greener settings when all of the environments were relatively natural (48). A meta-analysis that combined data from ten studies undertaken across the United Kingdom with a total of 1,252 participants found improved self-esteem and mood across a range of natural environments, including forests, urban parks, and wilderness areas; however, all studies were conducted by a    14  single research team that used an opportunistic sampling methodology, which may bias the reported results (92).  1.2.3 Facilitated social contact Natural space may also provide a venue for individuals to come together, strengthening existing networks of social support (73,100) and promoting engagement in socially oriented activities that can increase social capital within communities (101). Social support refers to instrumental, emotional, and informational functions performed by significant others, such as family and friends, for the benefit of an individual (102); social capital, on the other hand, can be defined for individuals or communities, referring to potential or readily available resources that derive from personal social ties or broad networks (103). Both social support and social capital have been found to mitigate stress by providing a sense of security, enhancing self-confidence, reducing the feeling of being alone, and buffering the impacts of stressful situations on an individual (104). As with physical activity, specific aspects of natural settings may be particularly important, with more-structured greenspace such as parks and community gardens providing a unique niche for social contact (73).   The results of studies that examined broad populations have not been wholly consistent, however. A cross-sectional study of more than 2,000 adults in Adelaide, Australia found that those who described their neighborhoods as containing a greater number of natural elements such as parks and streetscape greenery reported higher levels of social cohesion and social interaction (22). However, residents of two separate Washington D.C. suburbs ranked natural features as less important than factors such as the size of housing parcels and the design of street networks when evaluating features of the built environment relating to their sense of community (101). Furthermore, higher overall neighborhood greenness was linked to lower    15  levels of social support in Chicago, although larger total park acreage correlated with higher levels of social support (73).  Natural spaces may also play a unique role in facilitating social interactions for individuals with mental health disorders that may impede their interactions with others, such as anxiety and schizophrenia (42), or among older adults who may be isolated due to mobility limitations or communication difficulties such as those that can accompany hearing loss or age-related cognitive decline. For example, residents of the United Kingdom who felt socially excluded due to issues such as unemployment or economic deprivation reported that a months-long course of environmental volunteering, which generally involved hands-on activities such as habitat maintenance and trail-building, increased their sense of connection to their communities and also improved their interpersonal social skills (105). However, no comparison was made with a program of volunteering indoors or in less natural environments. Elderly individuals in Finland, many of whom had mobility issues that limited their interactions, most often cited “seeing others” as their primary motivation for visits to a garden that featured trees, walking paths, and a pond (55). In a qualitative study in the Vancouver, Canada region, low-income older adults expressed particular appreciation for bluespaces as locations for social engagement, or even simply time spent observing the social interactions of others (106). These opportunities were especially valued by the many participants who lived alone and were concerned by a sense of their own isolation.  1.2.4 Buffered impacts of noxious urban exposures In contrast to the pathways described above, the final construct primarily classifies natural spaces as a buffer against the excess noise, crowding, heat, and air pollution that typify many urban settings, and which may lead to or exacerbate negative symptoms of mental illness (50,107). Access to large natural areas can reduce the detrimental psychosocial symptoms    16  associated with high levels of traffic noise (108), and there is also some evidence that natural spaces can partially mitigate the respiratory harms of exposure to traffic-related air pollution, including asthma hospitalizations (109) and incident cases of asthma among children aged 10 and below (110). As with these latter two studies, the strongest evidence comes from studies that have examined natural space alongside one of these potentially harmful exposures to explain health outcomes. An effort that focused on premature mortality, including deaths due to suicide, among adults aged 15-64 across 50 large English cities reported no significant associations between the proportion of greenspace in each city and mortality rates, and the effects winnowed even further after the inclusion of air pollution in models (111). Looking at potential mediators between long-term exposures to both greenspace and bluespace and mood disorders, a cross-sectional study in Barcelona found that reduced air pollution and noise both mediated the observed associations to some degree, between less than 1% and almost 30% for air pollution and between 2% and 5% for noise (23).   1.3 Challenges in natural space exposure assessment at the metropolitan scale  Although many efforts relating nature to health rely on simplistic NDVI-based measures, it is not clear that these measures adequately represent exposure, whether defined by type, quantity, or quality (51,65,72). One common thread among studies reporting positive associations is the inclusion of subjective exposure measurements, either to distinguish between quantity and quality (51,65) or to replace distance-based measures (22). These results may indicate that perceptions are more relevant when mental health outcomes are under review or, conversely, that such perceptions are a biased measure of exposure that results in overestimating effects.   Such studies also point to the importance of considering natural space quality along with quantity. Residents of neighborhoods in Perth, Australia that contained medium- or high-quality public open space – as defined by participant appraisals of attractiveness, comfort, and safety –    17  had twice the odds of low psychological distress as residents of neighborhoods with low-quality space (51). In a study conducted in four large Dutch cities, increases in quantity and quality of greenspace were linked to better mental health status, but the strongest relationship was found for high-quality streetscape greenery (65). However, integrating quality assessments into large-scale studies is challenging, and no current tools have been validated for application to the full range of urban natural spaces (112), despite numerous calls for appraising quality along with quantity (13,18,113).  Another challenge of assessing population-level impacts at a metropolitan or regional scale is the fact that such large areas may contain multiple demographic subgroups with distinct impacts, which can lead to null findings when effect estimates are averaged for the population as a whole. For instance, a meta-analysis examining exercise in natural settings found the largest effects on mood among men and the middle-aged (92). With respect to SES, greater improvements in mental health were reported for individuals with lower SES, whether defined by income or education (114,115) or by employment status (45,86).   Mental health status may also be a moderating factor: in a number of studies that have included people with good overall mental well-being and those with diagnosed mental disorders, the effects were substantially stronger in the latter group than in the former (43,85,116). In some cases, the benefits for individuals with mental disorders were dramatic: Among children with ADHD, a single, 20-minute walk in a park resulted in improvements in attention roughly equal to the peak effects of the most common pharmaceutical treatment for the condition – unfortunately, the study was not designed to assess whether these effects persisted (50). Similarly, a nine-day therapy program for people with alcohol dependency and depression conducted in a forest led to a remission of their depression symptoms (64). On the other hand, in a study that evaluated garden visits among older adults residing in a nursing home, depressed individuals experienced    18  fewer positive effects on recovery, concentration, and pain from their visits, although this may have been due in part to greater difficulties accessing the garden site (55).  1.4 Dissertation rationale and objectives Although a great deal of research has been conducted since Wilson first articulated the biophilia hypothesis to describe our innate preference for more-natural environments in 1984 (117), these studies have resulted in widely divergent descriptions of the relationship between natural space and mental health, ranging from marked benefits (20–22,50), to no association (23–25), to potentially harmful effects (26,27). One potential cause of these conflicting findings is the diversity of definitions (and terms) used to assign natural space exposure (13). Another important methodological issue is the over-reliance on subjective outcome measures, such as self-reported anxiety and depression (21), general distress (51), or perceived overall mental or emotional health (22,52,65,118). Depending on how they are constructed, such measures may be subject to self-report bias and may also reflect midpoints on the pathway to an eventual mental health outcome, rather than relevant health outcomes themselves. Finally, although recent systematic reviews (14,15,17,18,76) have converged upon potential pathways linking natural spaces to health, formal assessment of these pathways via mediation analyses is still lacking, particularly with respect to facilitated social contact. This PhD dissertation sought to address each of these deficits in the published literature through a set of geospatial and epidemiologic analyses that integrate open-access, administrative, and large-scale survey data, with the aim of fulfilling the specific objectives described below.  1.5 Dissertation setting The Vancouver census metropolitan area (CMA) is located in southwestern British Columbia, Canada (Figure 1.1 inset map), covering 288,255 hectares with a population density of 8.03 persons per hectare (119) and comprising 21 municipalities, one Electoral Area, and one Treaty    19  First Nation (120). Home to 2.5 million residents in 2016, approximately 7% of the Canadian population (121), the region is expected to experience substantial population growth over the next few decades, reaching 3.4 million residents by 2040 (122). The third largest metropolitan area in Canada behind Toronto and Montreal, it is also one of the most racially and ethnically diverse, with 41% of the population made up of immigrants according to the 2016 Census (123). The area’s geography is also diverse, ranging from large mountains along the northern boundary to wetlands and large-scale agricultural areas in the south. The City of Vancouver is separated from the cities of North Vancouver and West Vancouver by the Burrard Inlet, bounded by the Fraser River to the south, and along the Strait of Georgia to the west (124); these bodies of water also serve as shipping routes served by the Port of Vancouver, which is the largest port in Canada (125). Widely known for its urban nature, particularly Stanley Park, Vancouver’s tree canopy has been steadily declining over the past few decades, falling from 22.5% in 1995, to 20% in 2006, and to 18% in 2014 (126). However, long-term planning strategies at the regional (127) and municipal levels (128) include provisions that seek to protect and improve access to natural areas for residents of the region.        20   Figure 1.1 Vancouver census metropolitan area (CMA) study location The main map shows the boundaries and names of the individual municipalities included in the Natural Space Index, along with their relative population sizes in 2015. The complete study area is shown as a red dot on the inset map, which indicates Vancouver’s location within North America.   1.6 Dissertation structure 1.6.1 Developing a model of natural space exposure for the Vancouver CMA The objective of Chapter 2 is to develop and evaluate a comprehensive Natural Space Index (NSI) for the Vancouver CMA that robustly models exposure based on the presence, form, accessibility, and quality of greenspace and bluespace.     21  1.6.2 Assessing the relationship between natural space exposure and cases of incident psychotropic prescription dispensation The objective of Chapter 3 is to explore the relationship between natural space exposure and cases of incident antidepressant and anxiolytic prescription dispensation across the Vancouver CMA.  1.6.3 Exploring the role of sense of community belonging along the pathway linking natural space exposure to mental health The objective of Chapter 4 is to evaluate the relationships between natural space exposure and measures of negative mental health, psychological distress, and major depressive disorder, as well as exploring the potential mediating role of neighborhood social cohesion on these relationships in the Vancouver CMA.     22  Chapter 2: Developing a model of natural space exposure for the Vancouver census metropolitan area 2.1 Background In the three decades since Wilson first popularized the term “biophilia” to describe the innate connection between humans and natural environments (117), the mental health impacts of exposure to nature have been increasingly studied by environmental psychologists, urban planners, and epidemiologists (15,113,129–131). Bridging these disparate disciplines in order to design health-promoting cities is more urgent than ever in light of the mass movement of individuals from rural to urban areas, particularly in low- and middle-income countries (132). One challenge to such collaborations is the diversity of natural environments: in urban areas, nature comes in a vast variety of forms, including both greenspace such as parks, gardens, and street trees, and bluespace such as rivers, oceans, and lakes. Together, these can be conceptualized as “natural space” – standing in stark contrast to the concrete, glass, and brick of the built environment – but precise definitions and their associated impacts are critical to design, planning, and management.   Experimental and observational research indicates that exposure to such natural space has benefits for both physical (22,93,133–135) and mental health (20–22,51,65,86,136), with generally larger benefits for the latter (22,115,118,137). Emerging evidence suggests that natural space may decrease rates (21,115), alleviate symptoms (40,43,138), and provide a unique treatment setting (42,57–59) for mental health conditions. Additional psychological benefits may also accrue to individuals with pre-existing mental illness (43,63,92,116).  As researchers have moved from laboratory-based experiments to population-level studies, however, the impacts of natural space on mental health have become less clear. A growing body of research conducted in settings around the globe has shown weak or non-existent    23  associations with mental health benefits across large populations (52,139), or found positive impacts only within specific subgroups (114,115). These divergent results may be due to the sheer heterogeneity of studies: the wide range of outcomes of interest; variation in data sources and geographic scales; and diverse definitions of “natural space” that represent a multitude of forms and comprise both objective and subjective measures.  For example, when a cross-sectional study in Chicago examined two distinct forms of greenspace, no association was observed between park distance, social support, and stress, but neighborhood vegetation was found to mitigate the impact of stress (73). A separate cross-sectional study in Wellington, New Zealand focused on visibility of both greenspace and bluespace (primarily in the form of ocean views), but only the latter was associated with reductions in psychological distress (140). In contrast, a longitudinal study in the United Kingdom that integrated data from land-cover databases found that higher amounts of neighborhood greenspace reduced mental distress (20) and a study in Scotland linked the perceived presence of “environmental incivilities” (including a lack of greenspaces and playgrounds) to increased rates of anxiety and depression (21). Looking specifically at street trees as a form of urban nature, a study in Toronto, Canada reported that the presence of just ten trees on a single city block was associated with health improvements of equal magnitude to being seven years younger or an increase of $10,000 in annual household income (141). The simplest and most commonly used objective measure of greenspace is the satellite-based Normalized Difference Vegetation Index (NDVI), which is calculated using the different wavelengths of light absorbed by vegetation in comparison with hard surfaces such as concrete or bare land and more accurately represents “greenness” than “greenspace” because of the way it is calculated (66–68). In addition, attempts to compare NDVI with perceived greenness are rare and have produced conflicting results (66,72). More recently, an examination of the correlation between NDVI and land-use types in Barcelona, Spain reported that areas such as    24  parks, gardens, and grassy fields explained only 32% of variability in NDVI scores based on 100-meter buffers, likely due to differences in spatial scales or improper categorization of areas such as private gardens (142). In addition, although classifying NDVI-based measures as representing “greenness” and identifying potential associations with broad land-use types are important measures, they do little to support research translation. Together, these conceptual and measurement issues constitute critical knowledge gaps that prevent health impacts from adequately informing urban sustainability and natural space protection plans, increasingly common worldwide (28,30,31). Such plans affect every urban resident, with the potential to translate small individual benefits into large population-level impacts (143).  Outstanding issues that public health researchers must address to develop effective partnerships with urban planners and policymakers include: 1) working definitions of natural space; 2) reliable methods of estimating exposure at various spatial scales, ranging from a single neighborhood to a large metropolitan area; and 3) research designs that prioritize the information and evaluation needs of these partners. One approach to resolving all of these issues is the development of a robust, comprehensive framework to characterize natural space. Such a framework would identify multiple forms of nature, couple quantity and quality measures, and tailor exposure assignments to the pathways and outcomes of interest. The development of the ParkIndex in Kansas City, Missouri offers a prototype, but is limited by its focus on physical activity and on parks as a single form of natural space (144). Another recent approach to developing a single indicator was similarly limited in the scope of its definition of natural space, basing it on land-use classifications derived from the Urban Atlas database, which is available only within the European Union (145).     25  2.1.1 Objectives Given these knowledge gaps, this study aimed to develop a comprehensive Natural Space Index (NSI) that robustly models exposure based on the presence, form, accessibility, and quality of multiple forms of greenspace and bluespace in the Vancouver, Canada census metropolitan area (CMA). Components of the NSI will be used in the following chapters to evaluate relationships between natural space and a range of mental health outcomes, and to inform ongoing progress in the development of best practices for modeling natural space as a factor in population-level health research.  2.2 Materials and methods To support assignment of residential exposure to natural space, the areal unit of analysis was the centroid of each of the 60,242 unique six-digit postal codes in the region, as derived from the Statistics Canada 2013 Postal Code Conversion File (PCCF) (146). In urban locations, six-digit postal codes represent areas ranging from a single building to a block face (147), leading to average positional discrepancies of 54-109 meters in comparison with the exact location of a specific dwelling (148). Relying on a geographic unit other than residential address is often necessary in population-based research due to privacy concerns and regulations, and may also be recommended in order to support comparisons between cities (145).  2.2.1 Data collection and measures 2.2.1.1 Natural space presence (“greenness”) The presence of vegetation was derived from two separate satellite platforms (Landsat-8 and MODIS) and based on two distinct measures: the NDVI and the Enhanced Vegetation Index (EVI). Sixteen-day MODIS (70) and Landsat-8 (149) data were downloaded from the United States Geological Survey’s EarthExplorer website (150). For MODIS data only, the MODIS Reprojection Tool (151) was used to mosaic, convert, and re-project files for use in geographic information systems (GIS) software. For Landsat-8 data only, images with less than 20% cloud    26  cover were selected to ensure that the presence of water vapor did not unduly influence calculated values (67). In the final step, NDVI and EVI raster values were calculated separately for the dry (May-October) and rainy (January-April and November-December) seasons in the Vancouver region (152), averaged to create annual measures, and then aggregated to circular buffers ranging from 100 to 500 meters in size surrounding all six-digit postal code centroids. Buffer sizes in this range have been applied in numerous previous epidemiological studies that integrate NDVI as an exposure measure (12,66,72,153), and the extremes of the range represent two of the four pathways described in Section 1.2, with 100 meters designed to capture the visual elements of nature connected to direct psychophysiological benefits and 500 meters selected to represent the neighborhood scale most relevant to facilitated social cohesion.     Because bluespace generates low NDVI and EVI values similar to those of hard surfaces such as bare land or concrete paving (69), three approaches were tested for handling the presence of water: 1) relying on the original NDVI/EVI values (66,67,73); 2) masking water bodies identified in land-use databases (72,137,154) and then assigning these areas a value equal to the highest seasonal NDVI average for the study area, under the assumption that bluespace may provide similar benefits as areas of high greenness; and 3) following this same masking procedure but setting bluespace to the highest possible NDVI/EVI value of +1.  2.2.1.2 Natural space form These satellite-based measures of “greenness” were expanded to incorporate specific types of nature by integrating data on three distinct forms: greenspaces, bluespaces, and street trees. Greenspaces were defined as largely green parcels, including athletic facilities, botanical gardens, cemeteries, golf courses, greenways, zoos, nature preserves, and local, regional, and provincial parks. To ensure that this measure was comprehensive, data were drawn from both    27  private and public databases. Private databases included the DMTI Spatial 2014 CanMap Parks database (155) and the ESRI 2014 North American parks database (156). Five of the 23 administrative units in the area had publicly available databases that were accessed online (157–161), including the two largest municipalities, Vancouver and Surrey. This procedure was designed to capture as many potential natural sites as possible, but also resulted in a large number of duplicates. When duplicates existed between the public and private databases, the public database details were included. All unique parcels were compared with orthophoto imagery within ArcMap 10.3.1 and their boundaries were manually edited to remove non-green areas, such as large parking lots.   The presence of bluespace was based on polygon data indicating the locations and sizes of permanent water features such as oceans, lakes, and rivers, as well as intermittent sources such as sloughs and bogs (162). Such intermittent sources comprise a much smaller percentage of total bluespace than large bodies of water such as the Fraser River, but often play vital roles in local First Nations’ history and present-day ecology, with the Camosun Bog (163) and Finn Slough (164) serving as two prime examples. No minimum sizes were placed on either green or blue parcels for inclusion in the index.   Multiple studies have linked street trees to direct mental health benefits (165) or to factors known to improve mental health, such as increased social support and cohesion (166,167). Because street trees may account for a large proportion of the natural environment experienced by individuals through window views (168,169) or during their daily commutes (170), street tree density was included as a stand-alone measure. Geocoded street tree locations were available for five municipalities (New Westminster, North Vancouver, Surrey, Vancouver, and White Rock) – retrieved from open-access data websites (171–173) and received from municipal government employees (174,175) – and aggregated to create a single database.    28  2.2.1.3 Natural space access One concern about relying on vegetation indices or land-cover databases is that areas with limited access, such as golf courses, may not be demarcated (73). Because restricted access may reduce the amount of exposure for nearby residents, previous studies have limited their natural space measures to publicly accessible greenspace by removing features such as private gardens (118,145,176) or by focusing solely on municipal parks (26,73,144,177). An alternative approach is delineating natural areas with restricted access but continuing to include them as a source of potential exposure via specific pathways. For example, golf courses that are only physically open to members may not provide others with opportunities for physical activity or social interaction, but they may still provide direct psychological benefits via natural views (178,179).  In this effort, all greenspace parcels in the region were categorized as either public-access or restricted. The access-categorization procedure entailed two steps. First, the initial coding of greenspace parcels was used to identify those that might have restricted access, including golf courses, cemeteries, athletic facilities, botanical gardens, exhibition grounds, campgrounds, amusement parks, and zoos. Next, an online search was performed to determine whether these parcels should be properly categorized as private or fee-based, with any mention of an entrance or reservation fee, requirement for special permits, or need to call ahead to obtain entry used as determining criteria. In the absence of readily available data that might indicate access restrictions for bluespace, all bluespace was considered to be publicly accessible.  2.2.1.4 Natural space quality The quality of natural space was assessed using the desktop version of the Public Open Space Tool (POST), known as POSDAT (180). POST has been applied in multiple studies examining park quality and outcomes such as recreational walking (94,181) and psychological distress    29  (51). POSDAT was designed to replicate the findings of POST while reducing the travel time associated with direct-observation audits and has shown both high reliability and validity in comparison with POST (180). Due to the large number of natural space parcels in the study region, a stratified selection procedure was used to identify a set of 200 parks (approximately 10% of all eligible sites) for appraisal via POSDAT. Parks were stratified by neighborhood-level household median income to support future analyses (beyond this thesis) on the relationship between area-level socioeconomic status (SES) and park quality. Neighborhood income was based on area-level median household values from the 2006 Canadian Census data due to concerns about data quality with the 2011 Census (182). Using an online random number generator (183), 100 municipal park sites were selected from each list of high- and low-income areas.  Appraisal procedures were based on the POSDAT training manual (184), with minor refinements made to account for: 1) the lack of pre-populated data on amenities; 2) the use of a single auditor; 3) translation from the Australian to the Vancouver context, including removing items that referred to activities such as netball and the presence of playground shade cover (in line with a recent recommendation from the POSDAT team) (185); and 4) the integration of a few items in the original POST (94) due to data availability (Table 2.1). These modifications resulted in a total summary score across the five POSDAT domains ranging from the lowest-quality ranking of 0 to a maximum of 45. In brief, the modified appraisal process was conducted via: 1) an initial scan of overhead satellite imagery from Google Maps; 2) a virtual tour of the exterior of each parcel via Google Maps Street View (using images collected in 2015 or the closest possible previous date); and 3) a final check against online data on amenities and features provided by each municipality, when available. If any items remained unclear after completing this three-step process, the first two steps were conducted again in order to ensure completeness and accuracy.    30  Table 2.1 Public Open Space Desktop Audit Tool (POSDAT) quality appraisal measures Adapted and refined from the original POSDAT tool  Domain  Items  Activities Does the space support the following activities?       tennis, soccer, football, baseball, fitness, basketball, hockey,      skateboarding/BMX, children’s playground, othera  Environmental Quality Is the parcel located on a beach or river foreshore? Are the following water features present?      lake/pond, fountain, stream/creek, wetlands Are the following environmental features present?      birds, other wildlife, gardens Estimate of approximate number of trees       none, 1-50, 50-100, 100 or more Where are the trees placed?       along all sides, along some sides, along walking paths,       randomly throughout Are there walking paths within or around the space? What is the shade like along paths?      poor, moderate, good Is the children’s playground fenced? Does the grass show evidence of watering?  Dog Regulations  Are dogs allowed off-leash or is there a separate area for dogs?b  Amenities Are the following amenities available?      BBQs, seating, picnic tables, café/kioskc, toilets, public art,       off-street parking    Safety What areas offer lighting?      amenities, paths, all sides, some sides, randomly        throughout, none Are the surrounding roads minor? Do at least half of the surrounding houses face the parcel?c  a Netball, cricket, rugby, and “athletics” removed from original item b Adapted from original POST measure (94) and included here due to data availability c Included in original POST measure (94) and included here due to data availability   2.2.2 Statistical analysis All spatial analyses were conducted in ArcMap 10.3. The multiple data sources described above were added as individual layers and clipped to the extent of the Vancouver CMA based on the    31  Metro Vancouver boundary file (186), which extends into surrounding maritime property around the region’s land masses (Figure 1.1 main map).  All individual natural space measures were constructed using circular buffers ranging from 100 to 1,600 meters around the six-digit postal code centroids (see Appendix A). These sizes were selected in alignment with previous research (187), such as the one-kilometer proximity to bluespace used in multiple studies (114,188) and the 1,600 meters often described as the maximum distance individuals are likely to travel to a park for regular use (26,144). As with NDVI/EVI, selection was driven by the hypothesis that smaller buffers would be more relevant to direct psychological benefits, while larger buffers would be more applicable to socially mediated benefits. In addition, the presence of public greenspace within 400 meters was assessed to align with Vancouver’s Greenest City 2020 Action Plan, which defines this distance as the maximum for publicly accessible natural space (28).  To evaluate relationships among these multiple measures of natural space presence, form, accessibility, and quality, postal-code level data were analyzed using R version 3.2.4 (189). Descriptive statistics were used to summarize the data, with Pearson’s correlations calculated for associations between continuous variables (such as NDVI and EVI values). Because many population-level studies characterize NDVI categorically to ease interpretation (67,68,137,190), variables were also transformed into quintiles for use in subsequent analyses. Finally, to support the development of a single index value for each postal code, principal component analysis (PCA) was conducted on centered and scaled variables using the prcomp function in the stats package in R (189). Due to the large number of potential factors, an initial PCA was performed on a broad set of variables comprising multiple indicators for each of the four main domains (presence, form, accessibility, and quality), and the resulting rotated loadings and biplot were used to identify potential variables for pruning.     32  2.3 Results 2.3.1 Natural space presence (“greenness”) Satellite source, index type, buffer size, and the method used to account for bluespace all impacted the greenness value at the postal-code level (Table 2.2). Based on 250-meter buffers, the smallest scale available for MODIS, mean NDVI values were significantly higher than mean EVI values for both seasons and from both MODIS and Landsat, ranging from 0.19 for MODIS EVI during the rainy season to 0.49 for MODIS NDVI during the dry season. However, seasonal MODIS NDVI and EVI values were still highly correlated, at 0.85 in the dry season and 0.95 in the rainy season. Mean annual Landsat EVI was significantly lower (0.23) than annual NDVI (0.45), and this difference remained when annual Landsat EVI/NDVI values were calculated using the two alternate approaches for handling bluespace described earlier. 2.3.2 Natural space form Variation also existed when examining specific forms of natural space. Using the City of Vancouver’s definition of greenspace access (28), 84.9% of postal codes were within 400 meters of public greenspace. However, other municipalities recommend distances of 250-300 meters (145), and these proximities have been associated with reduced reporting of depressive symptoms (177). Changing the buffer size to 250 meters reduced the percentage of postal codes with access to 62.9%. The addition of bluespace increased this figure slightly, to 65.3%. However, proximity to bluespace is generally evaluated within a 1,000-meter buffer (114,188), and 56.9% of postal codes had access based on this criterion. Looking at access to public greenspace within 250 meters or bluespace within 1,000 meters, 83.9% of postal codes met one of these two criteria. Conversely, setting a more-stringent definition of accessibility as the presence of both public greenspace within 250 meters and bluespace within 1,000 meters, only 35.9% of postal codes had such access. Street tree data were available for close to 60% of all postal codes (34,677) and counts in those locations ranged [median] widely: 1) from 0-203 [31]    33  trees within a 100-meter buffer; 2) from 0-596 [182] trees within 250 meters; and 3) from 0-1,757 [653] trees within 500 meters.  Table 2.2 Descriptions and comparisons of satellite-based measures  All comparisons are based on 60,242 six-digit postal codes   Greenness measure, season/bluespace correctiona     Min     Median     Mean    25%ile, 75%ile     SD     Max     r  Seasonal        Landsat NDVI, dry -0.29  0.45  0.45  0.37, 0.52  0.13  0.88  ref Landsat NDVI, rainy -0.55  0.46  0.45  0.36, 0.56  0.15 0.89  0.80 Landsat EVI, dry  -0.01 0.27 0.27 0.23, 0.32 0.08 0.64 0.88 Landsat EVI, rainy  0.00 0.19 0.19 0.14, 0.23 0.06 0.57 0.71 MODIS NDVI, dry -0.30 0.49  0.49  0.42, 0.57  0.14 0.89  0.62 MODIS NDVI, rainy -0.30  0.44  0.42  0.34, 0.51  0.14 0.86  0.69 MODIS EVI, dry -0.30  0.25  0.25  0.21, 0.30  0.09 0.60  0.68 MODIS EVI, rainy -0.30 0.20 0.19 0.15, 0.24 0.08 0.52 0.60  Annual        Landsat NDVI -0.26 0.46 0.45 0.37, 0.54  0.12 0.83  ref Landsat NDVI, bluespace-adjusted to highest average value  -0.21  0.47  0.46  0.38, 0.54 0.12 0.88  0.93 Landsat NDVI, bluespace-adjusted to +1  -0.20  0.47  0.46  0.38, 0.54  0.12 1.00 0.92 Landsat EVI -0.01 0.23  0.23 0.19, 0.27  0.07 0.56 ref Landsat EVI, bluespace-adjusted to highest average value  0.00  0.23  0.23  0.19, 0.27 0.07 0.62 0.91 Landsat EVI, bluespace-adjusted to +1 0.00  0.23  0.23  0.19, 0.27 0.07 1.00 0.82 a All values calculated based on a 250-meter buffer SD = standard deviation; r = correlation coefficient All values are rounded       34  2.3.3 Natural space access Most greenspaces were publicly accessible (Figure 2.1 inset map). Only 90 of 2,205 largely green parcels (4.1%) had some restriction on access, and the percentage of total greenspace area with restrictions was similar (3.8%). The majority of restricted spaces (52 of 90) were golf courses, followed by cemeteries (20), and botanical gardens (8). Integrating private greenspace had little impact on access measures, increasing the number of postal codes with access within 100 meters by a single percent, and those with access within 250 meters from 62.9% to 64.4%.   Figure 2.1 Natural space forms across the Vancouver CMA  The main map shows a zoomed-in section of the study area, containing a range of natural space forms. Bluespaces are marked in blue, public greenspaces in light green, and private greenspaces in lilac. The area of this zoomed-in region is marked in red on the inset map, which shows all natural spaces across the Vancouver, Canada census metropolitan area.   2.3.4 Natural space quality As measured by POSDAT, quality was fairly low across the 200 parks, with a mean summary score of 15 and a maximum score of 29 out of 45. The lowest scores were in the activities    35  domain (2 out of 9 points), which comprises dedicated areas for sports such as tennis or basketball as well as playgrounds. Other low scores were in the amenities domain (2 out of 7 points), which includes features such as picnic tables and public bathrooms. No statistically significant difference was seen in overall quality by neighborhood-level household income, and many domain scores were slightly higher in the low-income group. The one exception was environmental quality, with an average of 7.8 out of 19 in high-income areas compared with 6.8 in low-income areas. Although the appraised sites represented only 19.7% of all eligible greenspaces, 81.9% of postal codes were within 1,600 meters of one of these sites, representing robust geographic coverage despite the limited number of appraised parcels.  2.3.5 Natural space measure comparisons Recoding continuous natural space variables into categories is common among population-level health studies, and this approach was followed as a first step in comparing the various methods of assessing natural space exposure. Landsat NDVI was selected as the standard for comparison due to its use in multiple previous studies (67,68,137,190), and consistency with a recent effort to correlate NDVI with land-use and land-cover assessments (142). Annual values were used because our results indicated little seasonal variation, which agrees with other studies of greenness in the region (67). Finally, because some measures were not available for the complete study region, the results presented comprise the 31,921 postal codes (53%) for which all relevant data were available.  Following this approach, quintile rankings of postal codes based on annual Landsat NDVI values showed considerable variation in comparison with alternate measurements of natural space at the same scale. For example, comparing NDVI with the percentage of total greenspace within 250 meters resulted in an upward or downward shift of at least one quintile for 73% of postal codes (Figure 2.2). A similar shift (74%) was observed for the percentage of    36  all natural space (public and private bluespace and greenspace). The magnitude of the shifts was even more extreme for POSDAT quality scores and street tree density, with just over half and close to two-thirds shifting by two or more quintiles, respectively. Finally, setting bluespace areas to the highest average seasonal NDVI value resulted in greater agreement with unadjusted NDVI values (r = 0.93), but the change still resulted in one-fifth of postal codes being categorized differently.    Figure 2.2 Normalized Difference Vegetation Index versus alternate measures  This figure provides additional detail on the direction and size of quintile ranking changes for 31,921 six-digit postal codes when comparing alternate approaches to assessing natural space with annual Landsat NDVI based on a 250-m buffer. The height of the bars indicates the percentage of postal codes with a shift, with each grid mark representing a change of 10%. Grey bars indicate no change, orange bars indicate a shift to a lower quintile, and green bars indicate an upward shift; darker colors show more-extreme changes.     37  2.3.6 Principal component analysis After following the selection procedures detailed, a total of five natural space measures were entered into the final principal component analysis: 1) 250-meter Landsat EVI, with bluespace polygons set to the maximum seasonal EVI value; 2) percentage of public greenspace within 250 meters; 3) percentage of all natural space (greenspace and bluespace) within 250 meters; 4) access to bluespace within one kilometer; and 5) POSDAT quality appraisal summary score for the nearest park within 1,600 meters. Based on rotated loadings, the first two principal components explained 68% of the total variance, with the first principal component explaining 42.1% and the second principal component explaining 25.9% (Table 2.3). These five variables were grouped into three main domains, with overall greenness loading in the opposite direction of bluespace access and park quality, and public greenspace and total natural space percentage loading as a set (Figure 2.3).  Table 2.3 Principal component analysis results Proportion of variance and individual factor loadings (rotated) of selected natural space measures for 31,921 six-digit postal codes     Principal component 1  Principal component 2  Total proportion of variance explained   0.4206  0.2591     Natural space measure loadings (rotated)             Bluespace-adjusted Landsat EVI (250m)        0.4084        0.3964            Public greenspace (250m)      0.6437       -0.1010            Public and private natural space (250m)      0.6458       -0.1546             Accessible bluespace (1km)      0.0249       -0.6843            POSDAT summary score (1600m)     -0.0344      -0.5836  All values are rounded       38   Figure 2.3 Principal component analysis biplot  Biplot showing the first (PC1) and second (PC2) principal components, based on data representing 31,921 six-digit postal codes. Variable names shown in the plot are as follows: X15ELAS250B = Bluespace-adjusted (maximum seasonal average) Landsat EVI (250m) PG_250_PER = Percentage of public greenspace (250m) NS_250_PER = Percentage of public and private natural space (250m) BS_1KM = Presence of accessible bluespace (1km) SUMMARY_1600 = POSDAT quality appraisal summary score (1600m)    2.4 Discussion Our results illustrate that a wide range of factors can influence the estimation of exposure to natural space across a metropolitan area. Even the difference in NDVI between satellite platforms has the potential to shift classification at the postal-code level. The statistically significant differences between NDVI and measures reflecting natural space form, access, and    39  quality indicate that NDVI cannot simply be correlated with the percentage of greenspace, or with any other individual form of natural space. Although earlier studies have identified this concern (66–68), the present study adds additional weight and specificity. Our multiple comparisons also help fill the gaps identified in a recent effort to equate NDVI with land-use and land-cover types, which was limited by the need to collapse land-use and land-cover categories due to inadequate sample size (142).  Looking at the subset of studies purporting to examine solely “greenness”, a number of decisions have the potential to influence results and should therefore be made with care. In light of the variation seen among NDVI and EVI values, the selection of satellite platform and index type should be driven by factors such as the spatial precision of health outcome data, the types of vegetation that characterize the study area, and the temporal range of the study. Landsat data offer the finest spatial resolution at 30-by-30 meters, but the May 31, 2003 Scan Line Corrector error on Landsat-7 resulted in bands of missing values until Landsat-8 came fully online on April 12, 2013 (191). Although numerous approaches have been developed to overcome this error (192), relying on MODIS vegetation indices as an alternate approach may be preferable due to the time and complexity of these methods.   Our findings support existing research indicating that the Enhanced Vegetation Index is preferable when measuring high-density vegetation due to artificially high NDVI values (70,71). However, such vegetation is not present in all areas and researchers may also select NDVI to ease comparisons over time or with other studies. We also found that different methods of handling bluespace can change exposure for 20% of postal codes, suggesting that researchers should clip out areas of bluespace for studies focusing solely on residential greenness. In light of research showing that bluespace may be even more psychologically restorative than greenspace (52,92,140), alternate approaches are recommended for studies looking at natural    40  space more broadly, such as setting bluespace polygons at the end of the NDVI scale (+1) or to the highest recorded NDVI for the study area, as done for the PCA component of this study.  Paying attention to the form and size of natural space is also important for translating research findings into landscape designs and urban policies. Outlining the development of an urban greenspace indicator for the European Union, a recent study posits that one hectare is likely the minimum size for spaces designed to support physical activity and stress reduction (145). However, our measure of public greenspace, which largely comprised such areas, had weak correlation with NDVI. This suggests that smaller areas such as traffic circle gardens or grassy areas alongside buildings may produce at least some of the health benefits currently attributed to “greenness”, a supposition also put forward in a recent review of natural space accessibility metrics (187). As such, communities interested in the direct psychological benefits provided by natural space might prioritize small patches that are dispersed throughout an urban landscape within their plans and budgets. Planting additional street trees is another approach to achieving these benefits, and a growing body of research (141,165,193) supports this strategy. Once again, our findings indicate that this form of nature cannot be assessed simply by relying on NDVI data, making it critical to include as a distinct type in epidemiological studies.  In addition, although the distinction between public and private greenspace was minimal within the Vancouver CMA, this aspect of natural space provision may still be relevant in other settings or contexts. For example, a study in Accra, Ghana reported that access restrictions resulted in a dearth of usable public spaces and the authors posited that the situation was likely similar in many other regions of sub-Saharan Africa (194). With this in mind, factors such as private ownership and admission fees should always be assessed for studies hoping to identify the impact of natural resources specifically within the public domain. This process can be time-consuming because the relevant data are not generally collected in a single place and additional    41  challenges are likely to occur at larger geographic scales, however, so researchers should balance these costs against their translational aims. Factoring in access may be particularly important when physical activity or obesity are the outcomes of interest, because a physically inaccessible greenspace will not provide opportunities for recreation even though it may provide psychological benefits in the form of natural views or reduced traffic noise.  Selecting buffer sizes a priori based on the theorized pathway and outcome of interest, rather than simply testing multiple measures post hoc, is also important to advancing the field. For example, smaller buffer sizes such as 30 or 100 meters may be most relevant when it comes to direct psychophysiological benefits, in alignment with research that has demonstrated improvements to stress, mood, and attention that arise from passive views of nature (195), or even virtual immersion (47). Other studies have also demonstrated stress reductions based on proximity rather than use (196). Projects that aim to inform the development or refinement of planning initiatives may also wish to select buffers that align with existing policies, such as the City of Vancouver’s plan for providing public greenspace within 400 meters of all residential addresses (28). Further refinements could be possible depending on the availability of health outcome data, where access to precise locations could allow for the use of network-based distances that better reflect the natural spaces individuals encounter through windows or while traveling through their neighborhoods.  Finally, the results of the principal component analysis support the use of multiple measures of natural space, or the creation of a multicomponent measure, depending on study objectives and the availability and reliability of data at an appropriate scale. The final set of factors in the PCA was determined by following all of the best practices outlined above: 1) relying on EVI rather than NDVI to account for high-density vegetation; 2) adjusting for the presence of bluespace; 3) selecting buffer sizes based on hypothesized pathways to mental health; and 4) including    42  measures that align with relevant policy concerns. Even in this refined set of only five variables the loadings indicated three distinct sets, highlighting the independent contributions of these specific methods of measuring exposure to natural space.  Identifying the health effects of distinct forms of natural space may be preferable when it comes to knowledge translation, because it provides essential guidance for policymakers seeking to create evidence-based targets for natural space provision. As the field advances, such a granular understanding could also support the design of natural environments that are tailored to prevent or ameliorate the most-pressing health concerns of a specific population. At the same time, combining multiple metrics into a single index may be preferable from a methodological perspective due to the complexity of the spatial relationships among various forms of nature in urban environments and the interest of reducing measurement error in assessing latent constructs. Similar to their use in neighborhood health-effects research (197), PCA and other factor analytic or data-reduction methods could also be used to assess the relationship between varying neighborhood designs and health outcomes by enabling a closer examination of how neighborhood features may correlate with and capture specific latent constructs. Such an ecometric approach offers the potential to reduce measurement error and to capture the effects of natural elements as part of a broader environmental and social context (198).  2.4.1 Limitations Although the model described here was designed to build on less comprehensive measures of natural space, it has a number of limitations. First, not all data were available at the same level of specificity (or at all) across the study region, due to its large size and multiple administrative areas. This is especially true for the street tree analysis, which was limited to five municipalities. Even within this small set, there was heterogeneity in what constituted “street trees”, with some municipalities including screening trees along park borders (173) as well as those planted along    43  residential streets. Additionally, most databases excluded trees planted on private property, which may also provide green views for nearby residents. Second, there was temporal variation across data sources, with our initial search finding databases that had not been recently updated and those that were updated weekly. Some databases also included parcels that were not yet fully converted to usable greenspace. Third, there was low variability in the POSDAT quality appraisal scores, which may have arisen due to the reliance on a subset of park sites despite their selection via a stratified random sampling process. The use of a validated tool and a systematic approach to site selection help offset this limitation. Fourth, the NDVI and EVI measures selected as comparators and included in the PCA are widely used, but there are other, more-novel methods that we could not assess. These include LiDAR data, which have been used in other Vancouver-based studies, but are expensive to obtain and complex to interpret (199). Another approach relies on the visibility of natural spaces (140,200), but it requires sophisticated computer modeling and precise addresses that may not be available in many population-level studies. Fifth, a lack of address details also limits the range of potential buffer sizes that could be explored as part of the current effort: because postal code locations are based on a 100-meter centroid, smaller buffers – which may be particularly relevant in terms of the direct psychological benefits provided by natural views – were precluded from inclusion.   2.5 Conclusions This study demonstrates that many current approaches to modeling exposure to natural space likely misclassify exposures and have limited specificity. Nature in urban environments comes in many shapes and sizes, and each may have distinct health effects. Studies that rely solely on NDVI reduce the complexity of nature to mere “greenness”, which restricts the refinement of existing study designs and precludes clarification of the underlying pathways that link specific forms of nature to individual outcomes. In addition, such simplicity is a major barrier to    44  developing partnerships between public health professionals and the individuals and governments with the power to create change.      45  Chapter 3: Assessing the relationship between natural space exposure and cases of incident psychotropic prescription dispensation 3.1 Background Humans have an innate connection with nature: it soothes us, reminds us of our role in a broader ecological system, and links us back to our ancestral beginnings as forest-dwelling primates (117). Exploring this innate biophilic connection, a large body of research has linked exposure to natural space to human health benefits (14,201), and to mental health benefits in particular (12,19). The potential for natural spaces to benefit mental health is particularly critical because almost one-third of all Canadians experience mental illness during their lifetimes (2), and the estimated national economic impact of these conditions is at least $50 billion per year (4). The most recent major nationwide survey of mental well-being found that more than 1.6 million Canadians had some level of unmet need for mental health care (202), highlighting the importance of identifying upstream interventions to prevent the development of mental illness or ameliorate the associated symptoms.      Among the many elements of the urban environment that may present an opportunity to intervene, natural space has increasing evidence for its potential. For instance, in a Wisconsin study that examined the relationship between satellite-based surrounding greenness (based on the Normalized Difference Vegetation Index), tree canopy coverage, and scores on a measure of negative emotional states, increases in both measures of greenspace were associated with reductions in symptoms of anxiety, stress, and depression (203). Street trees as a specific form of nature were also associated with significantly reduced depressive symptoms among both adolescents and adults residing in deprived neighborhoods across the Netherlands (204). Data on more-objective mental health outcomes has produced more mixed results, however. A Dutch study that examined a range of both mental and physical illnesses diagnosed by general    46  practitioners linked additional greenspace within one kilometer of patient postal codes to reductions in 15 of the 24 disease clusters selected, with the largest improvements seen with depressive and anxiety disorders (115). A nationwide study in England that evaluated scores on a screening tool designed to support clinical diagnoses of disorders such as depression and anxiety (the GHQ-12) connected residence in areas with the highest amounts of greenspace to significant score improvements (20). However, an effort in Catalonia, Spain found improvements on this same GHQ-12 scale as a result of increasing surrounding greenness, but not due to the presence of either greenspace or bluespace within 300 meters of home (137). A study in Wellington, New Zealand that integrated another tool used to screen for serious mental illness (the Kessler-10 Psychological Distress Scale) saw improvements only with the visibility of bluespace and not greenspace (140).    Psychotropic prescriptions are a relatively understudied outcome that could be associated with natural space exposure. We identified only five studies that explored the relationship between natural spaces and psychotropic prescriptions: three ecological studies (25,165,205), one cross-sectional study (137), and one cohort study (23). This makes it challenging to evaluate what is already known about the impact of natural space on prescriptions related to mental health conditions. However, four of these five studies (23,137,165,205) reported some reduction in rates or self-reported medication use linked to higher amounts of natural space, with both street trees (165) and greenspace percentage (205) linked to lower rates of antidepressant prescriptions, and increasing surrounding greenness associated with less self-reported use of benzodiazepines (23), and less self-reported use of both antidepressants and sedatives (137). Even among these studies, however, not all forms of natural space exposure were associated with prescriptions, and the final study in this set of five found no significant protective effects of area-level greenspace and bluespace or greenspace alone on antidepressant rates (25).    47  Within Canada, the use of psychotropic medications as a treatment for mental illness is well defined. A combination of antidepressant medication and psychotherapy is the standard first-line therapy for major depressive disorder (MDD) (206,207). Psychotropic medications also account for a large percentage of all prescriptions written across the nation. The latest edition of the Canadian Prescription Drug Atlas lists antidepressants as the second-most-common drug category behind antihypertensives, and benzodiazepines (which fall under the broader class of anxiolytics) are ranked seventh (208).   This nested case-control study aims to advance upon this limited body of evidence by applying individual components of a robust model of natural space exposure across the Vancouver census metropolitan area (CMA) to data on incident antidepressant and anxiolytic prescription dispensations from a population-based cohort across the region. We hypothesize that individuals identified as cases will have lower levels of natural space exposure than their matched controls.  3.2 Materials and methods 3.2.1 Measures 3.2.1.1 Natural space exposure measures Natural space data were drawn from the Natural Space Index (NSI), developed with the intent of exploring the impacts of natural space exposure on mental health. A complete description of the NSI appears in Chapter 2, but a brief description of the way these measures were adapted or calculated for use in this specific study is provided here. For these analyses, we sought to select elements of the NSI to represent each of its four principal domains: 1) presence; 2) form; 3) accessibility; and 4) quality. Measures were also selected with attention to two of the four pathways described in Chapter 2: 1) direct psychophysiological responses and 2) facilitated social interaction. In line with these pathways, many measures were assigned based on 100-   48  meter buffers to represent the types of visible nature shown to reduce distress (140), facilitate stress recovery (39), and improve moods (79), while others were calculated using 500-meter buffers to represent the neighborhood scale (209) at which many important social interactions take place (22). In addition, surrounding greenness presence was assigned at an intermediate 250-meter buffer to capture elements of both visible and neighborhood nature and to support comparison with earlier research efforts that have used this metric at this scale (153,210).  As detailed in Chapter 2, the majority of the underlying data for these measures was collected between 2014 and 2017. For surrounding greenness alone, values of the Enhanced Vegetation Index (EVI) were extracted from NASA’s MODIS instrument dataset for each biweekly period over the full study period of January 1, 2008 through December 31, 2012 (70). The EVI was selected over the more commonly used Normalized Difference Vegetation Index (NDVI) due to its superior performance in areas that are marked by extensive areas of high-density vegetation (70,71) or with high levels of water vapor in the atmosphere (69), such as the Vancouver CMA. In light of earlier research indicating relative stability in these measures across seasons – both in the context of the NSI development (211) and within other epidemiological studies conducted in Vancouver (67) – annual values were calculated for each year between 2008 and 2012.   In addition, two different approaches were used to account for the presence of bluespace such as rivers, oceans, and lakes. In the first approach, water bodies included in a nationwide spatial database from 2010 (212) were masked before EVI values were calculated; in the second, these masked areas were assigned the maximum EVI value of +1 as a means of accounting for the possibility that bluespace may confer greater mental health benefits than bluespace, as described in multiple studies both in Canada (52) and around the globe (92,140).      49  The final set of exposure measures comprised the 12 items detailed below. All variables were calculated for the 60,242 six-digit postal codes included in the Statistics Canada 2013 Postal Code Conversion File (PCCF) (146), unless otherwise indicated.   Presence The presence of surrounding greenness was assessed in two ways. The 1) Residential Greenness 250m variable was defined as mean annual EVI in a 250-meter buffer, with bluespace polygons clipped before averaging to properly account for the negative EVI values created by water sources. The 2) Bluespace-Adjusted Residential Greenness 250m variable was based on the same methodology, but with bluespace polygons set to the end of the EVI scale.  Form In alignment with research showing the mental health benefits of natural views (169,195), two forms of visible nature were included: 3) Visible Greenspace 100m was defined as greenspace percentage in a 100-meter buffer and 4) Visible Natural Space 100m was defined as greenspace and bluespace percentage within a 100-meter buffer. Because previous efforts have associated street trees with both overall mental health improvements (141,203,204,213) and reductions in area-level antidepressant prescription rates (165), street tree density was also included as a distinct measure and assigned at three different buffers, selected based on their visibility (100m), alignment with EVI (250m), and neighborhood scale (500m): 5) Street Tree Density 100m, 6) Street Tree Density 250m, and 7) Street Tree Density 500m, respectively. Street tree data were only available for five of the 23 municipalities in the study area, resulting in coverage for just over half of all postal codes.     50  Accessibility To assess the impact of nature that could be accessed by all residents of a neighborhood, 8) Accessible Greenspace 500m was defined as publicly accessible greenspace percentage within a 500-meter buffer and 9) Accessible Natural Space 500m was defined as publicly accessible greenspace and bluespace percentage within a 500-meter buffer. These measures were complemented by two metrics in which access was defined on a binary basis. The 10) Binary Bluespace 1000m variable indicated the presence of bluespace within a 1,000-meter buffer, based on prior work showing health benefits of access to bluespace within this proximity (114,188). The 11) Binary Greenspace 400m variable was defined as an indicator of the presence of publicly accessible greenspace within a 400-meter buffer, based on the definition of access included in the City of Vancouver’s Greenest City 2020 Action Plan (28).   Quality Finally, 12) Public Park Quality was incorporated by examining the overall Public Open Space Desktop Audit Tool (POSDAT) score for the nearest public park within 1,600 meters, a distance used by other studies that have incorporated quality measures, including the ParkIndex developed in Kansas City, Missouri (144) and an application of POSDAT to psychological distress in Perth, Australia (26). Quality appraisals were only conducted for a stratified random sample of 200 parks (100 each in high-income and low-income areas), resulting in data for just under 80% of all postal codes at this buffer size. This variable was recoded into terciles to represent low-, moderate-, and high-quality parks, based on the average scores within the Vancouver CMA region.  3.2.1.2 Mental health outcomes Outcome data were drawn from the British Columbia-wide PharmaNet database (214), which captures details on all prescription medications dispensed from community pharmacies and    51  outpatient pharmacies located within hospitals, as well as on some medications dispensed by providers during visits to clinics or emergency departments (215). The World Health Organization’s Anatomical Therapeutic Chemical (ATC) classification system (216) was used to identify relevant prescriptions, based on two categories: N05B (anxiolytics) and N06A (antidepressants). Prescriptions falling within these two categories and eligible for dispensation by Health Canada (217) were then linked to their associated Drug Identification Number/Product Identification Number (DIN/PIN), the unique identification code listed within the PharmaNet database to identify a dispensed medication (218). Finally, a table containing this ATC-to-DIN/PIN crosswalk (Appendix B) was submitted to the data liaison for case ascertainment.  3.2.1.3 Covariates Due to use of linked administrative data, only two individual-level covariates were available to assess potential confounding: Age (in years, calculated as of January 31 for each year of data) and Sex (male, female, or unknown), which were controlled for via the matching process described below. Both of these characteristics were drawn from the provincial Medical Services Plan (MSP) Registration and Premium billing consolidation file (219), with Sex reflecting an individual’s current documented sex for each calendar year, which may change over time. In addition, three neighborhood-level covariates were included. Neighborhood Income Decile was used as a relative measure of area-level socioeconomic status (SES). Based on household income and adjusted for household size, it was derived from the 2006 Canadian Census (220) and assigned at the level of census dissemination areas (DAs), which contain between 400 and 700 residents on average (221). Both Population Density and Walkability were also assigned at the DA level. Population Density was based on data from the 2006 Census for the years 2008 through 2010 and on the 2011 Census for the years 2011 and 2012 (222). Walkability was based on 2016 data provided by Walk Score® (223), which is based on intersection density and density of amenities such as schools, shopping outlets, public transit stops, restaurants, and    52  theaters (224). Scores are based on the distance of these amenities, using a decay function to weight closer amenities more heavily. Such a weighting scheme is common in researcher-based approaches to assessing walkability, including the Walkability Index developed specifically for Vancouver and applied in previous studies examining the health effects of residential surrounding greenness (67), as is the score’s incorporation of intersection density. Walk Score® has been validated against individual objective indicators of neighborhood walkability in four large metropolitan areas across the United States and was found to correlate most strongly with the density of retail, service, and cultural/educational walking destinations (225). In addition, this validation effort reported the strongest correlations in metropolitan areas with higher average population density, as is the case with the current study location. Within Walk Score®, parks are included as one component of the broader recreational domain, but the study reported a Spearman correlation of only 0.24 with the number of parks per square kilometer (225).   3.2.2 Study sample 3.2.2.1 Cohort development Study participants were drawn from a cohort defined by members of the research team and identified by staff at Population Data BC (PopData BC), an organization that supports researchers across the province of British Columbia by serving as the data liaison for a range of governmental data sources (226). To be considered for inclusion in the cohort, individuals must have reached the age of 14 at baseline (January 1, 2003) and must not have filled either an antidepressant or anxiolytic prescription at any point between January 1, 2003 and December 31, 2007. Cases were subsequently defined as any member of this cohort who then filled at least one prescription for an antidepressant or anxiolytic at any point between January 1, 2008 and December 31, 2012 while residing in the Vancouver CMA at the time of the first dispensation and during the year prior to the first dispensation date, based on billing data from the British Columbia Medical Services Plan (MSP). The most common dispensation type was a    53  single 30-day supply of medication, with 63.4% of cases having just one dispensation, 20.1% having two, 9.1% having three, and 7.4% four or more. Taking multiple dispensations into account, the average number of total days supplied was 48.9 per individual. Regardless of the total number of dispensations, the initial service date (date dispensed) was subsequently used as the index date for the selection and assignment of controls. Controls were selected from the remaining members of the cohort and matched on age, sex, and residence in the Vancouver CMA on the index date and during the year prior to the index date. Controls were initially matched at a ratio of 4:1 and were allowed to be matched to multiple cases. This process resulted in the initial selection of 197,528 cases and 553,607 unique matched controls, with 236,505 controls being matched to more than one case. In summary, this study used a nested case-control design, in which cases and controls were nested in a cohort developed specifically to be free of any prior antidepressant or anxiolytic prescription dispensation.  3.2.2.2 Natural space exposure and covariate assignment To protect individual privacy, PopData BC replaced the six-digit postal codes contained in the NSI dataset and MSP and PharmaNet databases with study-specific geographic identification numbers that were used by the research team to assign exposure variables. Individual-level covariate data were assigned using study-specific unique participant identifiers (UIDs), which were generated by PopData BC as replacements for each unique, lifelong Personal Health Number used within the MSP and PharmaNet databases.  In order to ensure that natural space exposures were assigned to the correct six-digit postal code locations, the relevant Statistics Canada Postal Code Conversion file (PCCF) for 2008 through 2011 was used to identify the active postal codes in each year (227–230); the 2013 PCCF (146) was used for 2012 due to the unavailability of a file for that year. Because the PCCF files contain latitude and longitude of 100-meter population-weighted centroids for each    54  six-digit postal code, the exact location associated with an individual postal code may change from year to year, even though the actual sites served remain the same. In addition, new postal codes may be initiated, old postal codes may be retired, and some postal codes may be moved to a new location entirely. In light of this, only postal codes with geocoded locations within 100 meters of that used for initial NSI development (based on the 2013 PCCF) were included in final analyses, resulting in losses in the range of 4.5% to 5.3%, depending on the year.        3.2.2.3 Analytic sample development Due to missing exposure data, three distinct primary analytic samples were created: 1) one for analyses examining the majority of NSI measures; 2) one for POSDAT quality alone; and 3) one for street tree data only. Individuals were also removed due to missing data for any of the three covariates included in regression models: Neighborhood Income Decile, Population Density, and Walkability (Figure 3.1). The final primary analytic samples included cases matched to non-unique controls ranging from a ratio of 1:1 to 1:4 – depending on data availability among controls – by Age (by year), Sex (female, male, and unknown), and residence in the Vancouver CMA on the case index date and the year prior to this date. As a result, the final figures for each analytic sample were as follows: 1) main NSI sample: 177,351 cases and 639,390 controls; 2) POSDAT quality sample: 148,779 cases and 447,975 controls; 3) street tree sample: 96,788 cases and 207,667 controls. In addition, two analytic samples were created for use in sensitivity analyses: 1) complete-cases sample, in which data were available for all exposure variables to support direct comparison across these measures: 91,249 cases and 189,434 controls; and 2) main NSI sample with exact 1:1 matching: 177,351 cases and 177,351 controls. Each of these analytic samples is a subset of the total cohort of cases and matched controls, but they are not subsets of each other. For example, an individual did not need to have data available for all of the NSI variables in order to be included in the POSDAT sample, and vice versa.    55   Figure 3.1 Analytic sample development for the case-control study This figure illustrates the criteria used at each stage of restricting the study sample, as well as the number of potential cases and unique controls retained (in green) and omitted (in orange) as a result of each decision.   3.2.3 Statistical analysis The data-cleaning steps outlined above were conducted using R, version 3.5.0 (189), and analyses were carried out using SAS, version 9.4. The analysis proceeded in stages, starting with the calculation of descriptive statistics and bivariate models. Next, conditional logistic regression models were developed in each analytic sample to independently describe the relationship between each of the included natural space variables and the odds of receiving a psychotropic prescription dispensation, based on the strata defined by each case and its matched control(s). Finally, a set of models was developed to examine antidepressant and anxiolytic prescriptions separately as a means of identifying differential effects depending on the prescription type. All models comprised a single explanatory variable due to considerable correlations among them – particularly among those variables selected to reflect a single NSI    56  domain, as detailed in Section 3.3.2 below – and unadjusted effects were estimated as well as effects adjusted for the potential confounders described in the section on covariates above.  3.3 Results 3.3.1 Sample characteristics The main NSI sample totaled 816,741 cases and controls, consisting of 56.4% female, 43.6% male, and 0.01% unknown-gender members, with an average age just over 50 years and 51.4% from lower neighborhood income deciles (Table 3.1). These gender and age profiles were nearly identical in the POSDAT quality (Table 3.2) and street tree (Table 3.3) samples, but both contained a higher proportion of individuals from lower neighborhood income deciles, at 53% and 55.4%, respectively, likely due to the fact that the initial POSDAT quality appraisals were carried out on a stratified random sample of parks in only low- and high-income areas, while the street tree dataset was limited it its geographic scope to the five municipalities with available data. Members of all three samples also came from neighborhoods with higher-than-average population density, which was 5,249 persons per square kilometer in 2011 across the Vancouver CMA, according to Census figures for that year (231). The mean for the main NSI sample was 6,126 persons per km2, the mean for the POSDAT quality sample was 6,420 per km2, and the mean for the street tree sample was 7,281 per km2. The samples were marked by residence in “somewhat walkable” areas, on average, as categorized by the Walk Score® methodology (224), with all three means (59.6, 62.6, and 67.9, respectively) falling into this category.      57  Table 3.1 Summary of sociodemographic factors in the Natural Space Index (NSI) sample  Results presented separately for the full sample, cases, and controls   Full Sample (n = 816,741)  Cases (n = 177,351)  Controls (n = 639,390)    Percentage  Sex       Female 56.4 56.3 56.4      Male 43.6 43.7 43.6      Unknown 0.01  0.02 0.01 Neighborhood income decile  1st decile (lowest income) 10.3 10.8 10.2 2nd decile 10.5 10.7 10.4 3rd decile 10.5 11.0 10.4 4th decile 10.2 10.4 10.1 5th decile 10.3 10.3 10.3 6th decile 10.1 9.9 10.1 7th decile 9.6 9.6 9.6 8th decile 9.7 9.5 9.8 9th decile 9.7 9.4 9.8 10th decile (highest income) 9.1 8.6 9.3   Mean (SD)  Age 50.4 (17.9)  50.2 (17.9) 50.4 (17.9) Population density (persons per km2) 6,126 (6,791) 6,037 (6,634)  6,150 (6,832) Neighborhood walkability (Walk Score® summary)   59.6 (25.8) 59.2 (25.9) 59.6 (25.8) SD = standard deviation All means, percentages, and SDs are rounded       58  Table 3.2 Summary of sociodemographic factors in the POSDAT quality sample  Results presented separately for the full sample, cases, and controls   Full Sample (n = 596,754)  Cases (n = 148,779)  Controls (n = 447,975)    Percentage Sex       Female 56.4 56.3 56.4      Male 43.6 43.7 43.6      Unknown 0.01 0.02 0.01 Neighborhood income decile  1st decile (lowest income) 10.2 11.0 9.9 2nd decile 10.2 10.6 10.1 3rd decile 11.4 11.9 11.2 4th decile 10.7 10.9 10.7 5th decile 10.5 10.4 10.6 6th decile 10.1 9.9 10.2 7th decile 9.2 9.0 9.3 8th decile 9.2 8.9 9.3 9th decile 9.2 8.8 9.3 10th decile (highest income) 9.2 8.6 9.4   Mean (SD)  Age 50.3 (18.0)  50.2 (18.0) 50.3 (18.1) Population density (persons per km2) 6,420 (7,064) 6,358 (6,898)  6,440 (7,119) Neighborhood walkability (Walk Score® summary)   62.6 (24.5) 62.2 (24.5) 62.7 (24.5) SD = standard deviation  All means, percentages, and SDs are rounded       59  Table 3.3 Summary of sociodemographic factors in the street tree sample  Results presented separately for the full sample, cases, and controls   Full Sample (n = 304,455)  Cases (n = 96,788)  Controls (n = 207,667)    Percentage  Sex       Female 56.0 55.9 56.0      Male 44.0 44.1 44.0      Unknown 0.01  0.01 0.01 Neighborhood income decile  1st decile (lowest income) 11.5 12.5 11.0 2nd decile 11.0 11.5 10.8 3rd decile 12.3 12.9 12.0 4th decile 11.0 11.0 11.0 5th decile 9.6 9.4 9.6 6th decile 9.0 8.7 9.1 7th decile 8.2 8.0 8.4 8th decile 8.2 7.9 8.4 9th decile 9.4 9.1 9.5 10th decile (highest income) 9.9 9.0 10.4   Mean (SD) Age 49.8 (18.1)  49.8 (18.0) 49.8 (18.1) Population density (persons per km2) 7,281 (7822) 7,177 (7588)  7,330 (7928) Neighborhood walkability (Walk Score® summary)   67.9 (23.6) 67.0 (23.9) 68.3 (23.4) SD = standard deviation All means, percentages, and SDs are rounded        60  3.3.2 Natural space presence, form, accessibility, and quality Just over half of the members of the main NSI sample were within 1,000 meters of bluespace, and almost 90% were within 400 meters of public greenspace (Table 3.4). With respect to visible nature, on average 3.7% of the area within a 100-meter buffer of each postal code consisted of greenspace, while 3.8% was natural space. When these buffers were expanded to 500 meters to represent the neighborhood scale and natural space was limited to only publicly accessible areas, the averages increased to 8.2% and 9.8%, respectively. In addition, the two methods of handling bluespace to calculate surrounding greenness within 250 meters produced nearly identical results: simply clipping bluespace polygons before calculating values led to a mean greenness value of 0.22, while assigning these clipped areas the maximum EVI value led to a mean of 0.23. With respect to park quality, 43% of sample members had access to a low-quality park within 1,600 meters, while 29% had access to a moderate-quality park, and 28% had access to a high-quality park (Table 3.5). Finally, among those sample members for whom street tree data were available (Table 3.6), there was an average of 32 trees within a 100-meter radius, 198 trees within 250 meters, and 748 trees within 500 meters. None of these figures were markedly different when comparing cases with controls, with the exception of street tree density, for which controls had consistently higher averages across all three buffers. In addition, these figures are similar to those found for the region as a whole, as reported in Chapter 2. Pearson’s correlations calculated for the continuous explanatory variables demonstrate both the extent to which individual measures of the four main NSI domains – presence, form, accessibility, and quality – are related, and the degree to which each domain captures a distinct aspect of the natural environment. For example, looking at data from 2012, Residential Greenness 250m and Bluespace-Adjusted Residential Greenness 250m have an r of 0.89, and the three Street Tree Density measures have correlations ranging from 0.81 to 0.95 depending on the buffer size. Conversely, comparing Residential Greenness 250m to Street Tree Density 250m results in a correlation of -0.33 (Table 3.7).     61  Table 3.4 Summary of natural space exposures in the NSI sample  Results presented separately for the full sample, cases, and controls   Full Sample (n = 816,741)  Cases (n = 177,351)  Controls (n = 639,390)    Percentage Binary access to nature  (% yes)       Binary Bluespace 1000m    53.3 53.0 53.4       Binary Greenspace 400m       87.0 87.7 86.8    Mean (IQR)  Surrounding greenness (250m)        Residential Greenness  0.22 (0.09) 0.22 (0.09) 0.22 (0.09)      Bluespace-Adjusted       Residential Greenness  0.23 (0.09) 0.23 (0.09) 0.23 (0.09)  Visible nature (% in 100m)         Visible Greenspace   3.70 (0.33) 3.75 (0.56) 3.68 (0.27)     Visible Natural Space   3.81 (0.46) 3.88 (0.73) 3.80 (0.38) Accessible neighborhood nature (% in 500m)         Accessible Greenspace 8.19 (7.95) 8.20 (7.94)  8.18 (7.95)      Accessible Natural Space 9.76 (9.96) 9.84 (10.06) 9.74 (9.94) IQR = interquartile range All means, percentages, and IQRs are rounded        62  Table 3.5 Summary of natural space exposures in the POSDAT sample Results presented separately for the full sample, cases, and controls   Full Sample (n = 596,754)  Cases (n = 148,779)  Controls (n = 447,975)    Percentage Public Park Qualitya          Low Quality 42.9 43.0 42.9      Moderate Quality 29.4 29.1 29.5      High Quality 27.7 27.9 27.6 a Categories based on terciles among all 200 randomly sampled parks within Vancouver CMA All percentages are rounded       Table 3.6 Summary of natural space exposures in the street tree sample Results presented separately for the full sample, cases, and controls   Full Sample (n = 304,455)  Cases (n = 96,788)  Controls (n = 207,667)    Mean (IQR)      Street Tree Density 100m 32 (38) 30 (38) 32 (39)      Street Tree Density 250m  198 (225) 189 (222) 202 (225)      Street Tree Density 500m 748 (802) 715 (788) 764 (804) IQR = interquartile range All means and IQRs are rounded       63  Table 3.7 Comparisons among continuous natural space exposures All comparisons are calculated using 2012 values assigned at the six-digit postal code  Natural space exposure measure   Pearson’s correlation coefficient (r)    A  B  C  D  E  F  G  H  I  Residential Greenness 250m (A)  ref 0.89 0.11 0.09 0.20 -0.06 -0.31 -0.33 -0.31 Bluespace-Adjusted Residential Greenness 250m (B) 0.89 ref 0.13 0.16 0.19 0.08 -0.31 -0.35 -0.33 Visible Greenspace 100m (C) 0.11 0.13 ref 0.98 0.42 0.37 -0.09 -0.14 -0.14 Visible Natural Space 100m (D)   0.09 0.16 0.98 ref 0.41 0.41 -0.10 -0.15 -0.15 Accessible Greenspace 500m (E)   0.20 0.19 0.42 0.41 ref 0.81 -0.21 -0.27 -0.29 Accessible Natural Space 500m (F)   -0.06 0.08 0.37 0.41 0.81 ref -0.20 -0.27 -0.31 Street Tree Density 100m (G)   -0.31 -0.31 -0.09 -0.10 -0.21 -0.20 ref 0.87 0.81 Street Tree Density 250m (H)    -0.33 -0.35 -0.14 -0.15 -0.27 -0.27 0.87 ref 0.95 Street Tree Density 500m (I)    -0.31 -0.33 -0.14 -0.15 -0.29 -0.31 0.81 0.95 ref All values are rounded   3.3.3 Associations between natural space measures and psychotropic prescription dispensation In the main NSI model, there were no associations between Visible Greenspace 100m, Visible Natural Space 100m, Accessible Greenspace 500m, or Accessible Natural Space 500m and the odds of psychotropic prescription dispensation (Table 3.8). These results remained the same when examining only cases of antidepressant (Table 3.9) or anxiolytic dispensation (Table 3.10).     64    Table 3.8 All-dispensation results for the NSI sample  Results of unadjusted and adjusted models examining all cases and matched controls   OR (95% CI)  aOR (95% CI)+ Surrounding greenness (250m)        Residential Greennessa 1.005 (0.999, 1.011) 0.984 (0.976, 0.991)      Bluespace-Adjusted       Residential Greennessa  1.009 (1.003, 1.015) 0.991 (0.984, 0.999) Visible nature (% in 100m)         Visible Greenspaceb 1.001 (1.000, 1.001) 1.001 (1.000, 1.001)       Visible Natural Spaceb 1.001 (1.000, 1.001) 1.001 (1.000, 1.001)  Publicly accessible neighborhood nature (% in 500m)          Accessible Greenspaceb 1.000 (1.000, 1.001) 1.000 (1.000, 1.001)      Accessible Natural Spaceb  1.001 (1.000, 1.001) 1.002 (1.001, 1.002) Binary access to nature (% yes)          Binary Bluespace 1000mc  0.984 (0.974, 0.995) 1.008 (0.997, 1.019)     Binary Greenspace 400mc 1.086 (1.069, 1.103) 1.098 (1.080, 1.115)  + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = unadjusted odds ratio; aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    65     Table 3.9 Antidepressant results for the NSI sample Results of unadjusted and adjusted models examining antidepressant cases and matched controls   OR (95% CI)  aOR (95% CI)+  Surrounding greenness (250m)        Residential Greennessa 1.001 (0.992, 1.010) 0.971 (0.960, 0.983)      Bluespace-Adjusted       Residential Greennessa  1.004 (0.995, 1.013) 0.979 (0.968, 0.991)  Visible nature (100m)         Visible Greenspaceb 1.001 (1.001, 1.002) 1.002 (1.001, 1.002)       Visible Natural Spaceb 1.002 (1.001, 1.002) 1.002 (1.001, 1.002)   Publicly accessible neighborhood nature (500m)          Accessible Greenspaceb 1.001 (1.000, 1.002) 1.001 (1.000, 1.002)      Accessible Natural Spaceb  1.001 (1.001, 1.002) 1.002 (1.002, 1.003)  Binary access to nature (% yes)          Binary Bluespace 1000mc  0.991 (0.975, 1.007) 1.019 (1.003, 1.036)      Binary Greenspace 400mc 1.110 (1.083, 1.137) 1.127 (1.099, 1.155)  + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = unadjusted odds ratio; aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    66    In the hypothesized direction, both Residential Greenness 250m (OR = 0.98; 95% CI = 0.98, 0.99) and Bluespace-Adjusted Residential Greenness 250m (OR = 0.991; 95% CI = 0.984, 0.999) were associated with reduced odds of any psychotropic prescription dispensation after adjustment for confounding, at 2% and 1% respectively (Table 3.8). These effect sizes were larger when examining antidepressants alone – with an adjusted OR of 0.97 (95% CI = 0.96, 0.98) for Residential Greenness 250m and 0.98 (95% CI = 0.97, 0.99) for Bluespace-Adjusted Residential Greenness 250m (Table 3.9). For anxiolytics alone, both measures had ORs with confidence intervals including 1.0 in both the unadjusted and adjusted models (Table 3.10).  Table 3.10 Anxiolytic results for the NSI sample Results of unadjusted and adjusted models examining anxiolytic cases and matched controls   OR (95% CI)  aOR (95% CI)+  Surrounding greenness (250m)        Residential Greennessa 1.009 (1.001, 1.017) 0.994 (0.984, 1.004)     Bluespace-Adjusted       Residential Greennessa  1.012 (1.004, 1.021) 1.001 (0.991, 1.011) Visible nature (100m)         Visible Greenspaceb 1.000 (1.000, 1.001) 1.000 (1.000, 1.001)      Visible Natural Spaceb 1.000 (1.000, 1.001) 1.000 (1.000, 1.001)  Publicly accessible neighborhood nature (500m)          Accessible Greenspaceb 1.000 (0.999, 1.001) 1.000 (0.999, 1.001)      Accessible Natural Spaceb  1.000 (1.000, 1.001) 1.001 (1.000, 1.002) Binary access to nature (% yes)          Binary Bluespace 1000mc 0.980 (0.966, 0.994) 0.999 (0.984, 1.013)      Binary Greenspace 400mc 1.068 (1.046, 1.091) 1.076 (1.053, 1.099)  + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = unadjusted odds ratio; aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    67  The results for Binary Bluespace 1000m were mixed: unadjusted models for all prescriptions (OR = 0.984; 95% CI = 0.974, 0.995), antidepressants alone (OR = 0.99; 95% CI = 0.98, 1.01), and anxiolytics alone (OR = 0.98; 95% CI = 0.97, 0.99) showed reduced odds in the range of 1% to 2%, but the results for antidepressants alone were not significant. In adjusted models, beneficial effects were attenuated across the board and the model for antidepressants alone indicated a slightly increased risk (aOR = 1.019; 95% CI = 1.003, 1.036). By contrast, and counter to our hypotheses, both unadjusted and adjusted models of Binary Greenspace 400m were associated with increased odds of any psychotropic dispensation, with a 10% increase in the adjusted model (aOR = 1.10; 95% CI = 1.08, 1.12). The increased odds of antidepressant dispensation were higher at 13% (aOR = 1.13; 95% CI = 1.10, 1.16), while those for anxiolytic dispensation were lower at 8% (aOR = 1.08; 95% CI = 1.05, 1.10). Due to concerns about a potential statistical artifact in these results arising from the low contrast between cases and controls and the original 1:4 matching approach, an additional analytic sample was developed following a 1:1 matching approach. The results within this sample (Table 3.11) were identical in terms of the overall pattern, although the associated confidence intervals for both binary access variables were wider.         68   Looking at Public Park Quality, results were also mixed (Table 3.12). The presence of a moderate-quality park within 1,600 meters was associated with reduced odds in unadjusted models for all prescriptions (OR = 0.984; 95% CI = 0.970, 0.998), antidepressants alone (OR = 0.99; 95% CI = 0.97, 1.01), and anxiolytics alone (OR = 0.979; 95% CI = 0.961, 0.998), but the results for antidepressants alone were not significant: a pattern identical to that seen by prescription type for Binary Bluespace 1000m. Adjusted models showed similar effects, though smaller in magnitude. Conversely, high-quality parks were linked to increased odds of approximately 3% in adjusted models for all prescriptions (aOR = 1.03; 95% CI = 1.02, 1.05), antidepressants alone (aOR = 1.03; 95% CI = 1.01, 1.05), and anxiolytics alone (aOR = 1.03; 95% CI = 1.01, 1.05).  Table 3.11 Sensitivity analysis results for the NSI sample  Results of unadjusted and adjusted models examining binary access to nature measures and all cases and matched controls in 1:1 matched Natural Space Index (NSI) sample   OR (95% CI)  aOR (95% CI)+  All prescriptions (n = 354,702)         Binary Bluespace 1000ma 0.978 (0.965, 0.991) 1.001 (0.987, 1.015)       Binary Greenspace 400ma 1.084 (1.063, 1.106) 1.093 (1.072, 1.115)   Antidepressants (n = 154,524)         Binary Bluespace 1000ma 0.997 (0.978, 1.017) 1.026 (1.005, 1.047)      Binary Greenspace 400ma  1.114 (1.081, 1.148) 1.129 (1.095, 1.163) Anxiolytics (n = 200,178)         Binary Bluespace 1000ma  0.986 (0.969, 1.003) 1.006 (0.988, 1.024)     Binary Greenspace 400ma 1.077 (1.049, 1.105) 1.084 (1.056, 1.113)  + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on having access to nature OR = unadjusted odds ratio; aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    69   In the hypothesized direction, the presence of ten additional trees in models for Street Tree Density 100m was linked to adjusted odds reductions of 3% for all prescriptions (aOR = 0.97; 95% CI = 0.97, 0.98), 4% for antidepressants (aOR = 0.96; 95% CI = 0.96, 0.97), and 2% for anxiolytics (aOR = 0.98; 95% CI = 0.98, 0.99) (Table 3.13). These effects weakened when looking at street trees within larger buffers, with reductions of approximately 1% for Street Tree Density 250m and Street Tree Density 500m across all prescription types.  Table 3.12 Complete results for the POSDAT sample Results of unadjusted and adjusted models examining public park quality level and all cases and matched controls in POSDAT sample          OR (95% CI)                          aOR (95% CI)+   All prescriptions        Moderate Qualitya 0.984 (0.970, 0.998) 0.984 (0.970, 0.998)      High Qualitya 1.007 (0.992, 1.021) 1.030 (1.015, 1.045)  Antidepressants (n = 259,926)        Moderate Qualitya 0.990 (0.969, 1.012) 0.989 (0.968, 1.010)      High Qualitya 1.002 (0.981, 1.025) 1.031 (1.008, 1.054)  Anxiolytics (n = 336,828)        Moderate Qualitya 0.979 (0.961, 0.998) 0.980 (0.961, 0.999)      High Qualitya 1.010 (0.991, 1.030) 1.029 (1.009, 1.049) + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Based on comparisons to reference category of low quality  OR = unadjusted odds ratio; aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    70   Finally, a sensitivity analysis conducted in the complete-cases sample (Table 3.14) indicated similar results across the majority of exposure measures, with no or only slight shifts in the point estimates. The one major exception was Public Park Quality, in which moderate-quality parks showed an association with increased odds across all three prescription types, versus the decreased odds seen in the POSDAT analytic sample. However, the odds ratios included 1.0 for both all prescriptions and anxiolytics alone; for antidepressants alone, the aOR was 1.04 (95% CI = 1.01, 1.08).       Table 3.13 Complete results for the street tree sample Results of unadjusted and adjusted models examining street tree density and all cases and matched controls in street tree sample          OR (95% CI)                          aOR (95% CI)+  All prescriptions (n = 304,455)        Street Tree Density 100ma 0.967 (0.964, 0.970) 0.973 (0.969, 0.977)      Street Tree Density 250ma 0.992 (0.992, 0.993) 0.993 (0.992, 0.994)      Street Tree Density 500ma  0.998 (0.998, 0.998) 0.998 (0.998, 0.998) Antidepressants (n = 133,072)        Street Tree Density 100ma 0.956 (0.951, 0.960) 0.960 (0.955, 0.966)      Street Tree Density 250ma 0.990 (0.989, 0.991) 0.990 (0.989, 0.991)      Street Tree Density 500ma  0.997 (0.997, 0.997) 0.997 (0.997, 0.997) Anxiolytics (n = 171,383)        Street Tree Density 100ma 0.976 (0.972, 0.980) 0.983 (0.978, 0.988)      Street Tree Density 250ma 0.994 (0.994, 0.995) 0.995 (0.994, 0.996)      Street Tree Density 500ma 0.998 (0.998, 0.999) 0.999 (0.998, 0.999)  + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on the presence of 10 additional trees within the buffer area aOR = adjusted odds ratio; CI = confidence interval All odds ratios (OR and aOR) and 95% confidence intervals (95% CI) are rounded    71    Table 3.14 Sensitivity analysis results for the complete-cases sample Results of adjusted+ models examining all explanatory variables and all cases and matched controls in complete-cases sample   All  Antidepressants  Anxiolytics   aOR (95% CI)  Visible nature (100m)         Visible Greenspacea 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00)      Visible Natural Spacea 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00)      Street Tree Densityb  0.97 (0.97, 0.98) 0.96 (0.96, 0.97) 0.98 (0.98, 0.99) Surrounding greenness (250m)         Residential Greennessc 0.95 (0.94, 0.96) 0.92 (0.91, 0.94) 0.97 (0.96, 0.99)      Bluespace-Adjusted        Residential Greennessc 0.96 (0.95, 0.97) 0.94 (0.92, 0.96) 0.98 (0.96, 0.99)      Street Tree Densityb  1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) Publicly accessible nature (500m)          Accessible Greenspacea 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00)      Accessible Natural Spacea 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00)      Street Tree Densityb 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) 1.00 (1.00, 1.00) Binary access to nature          Binary Bluespace 1000mc 0.98 (0.97, 0.99) 1.00 (0.97, 1.03) 0.97 (0.95, 0.99)      Binary Greenspace 400mc 1.08 (1.04, 1.11) 1.11 (1.06, 1.17) 1.05 (1.01, 1.10) Public Park Quality         Moderate Qualitye 1.02 (0.99, 1.04) 1.04 (1.01, 1.08) 1.00 (0.97, 1.03)      High Qualitye 1.03 (1.01, 1.05) 1.02 (0.99, 1.05) 1.03 (1.01, 1.06) + Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability a Models based on a 1% increase in area within buffer b Models based on the presence of 10 additional trees within the buffer area c Models based on a 0.1-unit increase in EVI d Models based on having access to nature within the buffer area e Models based on comparison to lowest-quality park category (based on terciles) aOR = adjusted odds ratio; CI = confidence interval All odds ratios (aOR) and 95% confidence intervals (95% CI) are rounded     72  3.4 Discussion This study examined the effects of a diverse set of natural space exposure measures and the odds of incident antidepressant or anxiolytic prescription dispensation via a nested case-control design drawn from a population-based cohort, with mixed results. In line with initial hypotheses, the presence of ten additional street trees within buffers ranging from 100 to 500 meters was associated with a 1-4% odds reduction depending on the prescription type and buffer size, with larger reductions seen for smaller buffers and for antidepressant dispensations. This variation in effect may be due to the fact that such an addition represents an increase in the number of trees of 30% above the mean within a 100-meter buffer, but only 5% for 250 meters, and 1.3% for 500 meters. It could also be due to the fact that each of these buffers aligns with a separate pathway: street trees within 100 meters are likely to represent a visible form of nature and therefore provide direct psychophysiological benefits, while trees within 500 meters relate more directly to the neighborhood scale, and may therefore be more relevant to the pathway via facilitated social contact. Finally, these street tree measures represent a distinct, discrete, and easily measurable form of nature, which stands in contrast to many of the other measures included in this study, particularly surrounding greenness, which has shown little correlation with specific land uses (142) or perceived greenness (72).   That said, small reductions in the odds of dispensation were seen with surrounding greenness, with a 0.1-point increase in EVI linked to a 3% decrease for antidepressants and a 2% decrease for any dispensation. These benefits align with earlier studies that have demonstrated roles for surrounding greenness (12,23,203) and street trees in promoting mental health (141,203,204,213), while expanding the limited set of individual-level epidemiological studies that have examined prescriptions related to mental health conditions as an outcome (23,25,137,165,205). However, no associations were observed with either of the two measures of visible nature or the two measures of accessible neighborhood nature.     73  With respect to park quality, in comparison to low-quality parks, the presence of moderate-quality parks within 1,600 meters was associated with reduced odds of a prescription in the range of 1-2% across all prescription types. By contrast, high-quality parks were actually associated with a 3% increase in the odds, a somewhat counterintuitive finding that has also been reported in an Australian study that used POSDAT to assess park quality, although modeled using different methods. This Australian effort found increased odds of reporting high levels of psychological distress, as measured via the Kessler-10 Psychological Distress Scale (K10), in which the presence of a higher-quality park within 800 meters was associated with 19% higher odds of reporting high psychological distress (95% CI = 1.00, 1.38); for 1,200-meter buffers, the odds were 18% higher (95% CI = 1.00, 1.38) (26).    Initially, we explored two distinct methods of calculating the presence of surrounding greenness, including one that set bluespace areas to the highest possible EVI value, following the hypothesis that this approach would more properly reflect natural space exposure as a whole. However, this approach actually reduced the benefits associated with surrounding greenness, including slight attenuation of the relationship with anxiolytic dispensation to null. Considered together with our finding that having access to bluespace within one kilometer was tied to increased odds of dispensation, this suggests that bluespace may behave differently in the Vancouver CMA study setting than in some other environments. For example, a study in Wellington, New Zealand linked increases in bluespace visibility to significant reductions in psychological distress (140) and a panel study in England linked coastal proximity to better mental health (232). Other findings regarding bluespace and mental health have been more mixed, however: examining positive emotional well-being among 17,249 youth across Canada, Huynh et al. found indications of a protective effect of bluespace via a significant linear trend, particularly among young people residing in small cities, but the overall effect was weak (52). An analysis of national health survey data in the Netherlands, on the other hand, linked greater    74  bluespace exposure to poorer mental health: higher percentages of lakes, seas, and other inland water types within a neighborhood were associated with a 13% increase in the odds of having either an anxiety or depressive disorder and a 16% increase in the odds of anxiety disorders alone (27). Capturing this mixed picture, a systematic review that sought to summarize the mental health benefits of long-term exposure to natural space found that only one of three included studies that integrated a bluespace component reported benefits (12).    Even so, qualitative reports by older adult residents of the Vancouver region highlight the potential for psychological restoration offered by bluespace, with participants identifying bluespaces as particularly tranquil locations for relaxation and contemplation (106). However, these locations were selected by participants as places they enjoyed visiting, and they may not reflect the full range of bluespace across the region. The Fraser River is a busy shipping channel and home to the Port of Vancouver, through which 142 million tons of cargo were moved in 2017 (125). There are numerous other industrial uses of bluespace in the region as well, including a large cruise ship terminal in the heart of downtown Vancouver (125). Such sites may be associated with high levels of noise and air pollution (233), and therefore not tranquil. Unfortunately, although the NSI does contain details on specific types of bluespace, it does not allow for the identification of related land uses, so we are unable to explore the potential impact of this factor in the present analyses. More-rigorous approaches to bluespace exposure assessment are an emerging area of research – with efforts to examine both visibility (140) and quality recently described (112) – but these methodologies have yet to be validated or widely implemented.   The larger effect seen with respect to antidepressants as compared with anxiolytics is consistent with some studies that have included measures related to both disorders, but not others. For example, a large cross-sectional survey in Wisconsin found that surrounding    75  greenness and tree canopy had much larger benefits for measures of depressive symptoms than for measures of anxiety symptoms (203). Another effort that looked at surrounding greenness and self-reported use of tranquilizers or sedatives, antidepressants, and sleeping medication also reported larger reductions for antidepressants (20%) than for tranquilizers or sedatives (12%) (137). In addition, it reported no associations between greenspace or bluespace access (based on a 300-meter residential buffer) and any psychotropic medication use. Conversely, a Spanish cohort study found large reductions in self-reported benzodiazapene use (an anxiolytic) with higher levels of NDVI, but null findings when examining self-reported antidepressant use (23).      Unfortunately, the majority of studies that have examined psychotropic prescriptions have only looked at antidepressant use, limiting the ability to draw distinctions between antidepressants and anxiolytics. Even among these studies, however, findings diverge. Based on ecological analyses, the presence of one additional street tree per kilometer was linked to 1.4 fewer antidepressant prescriptions for every 1,000 residents of London boroughs (165). However, a separate study that examined England as a whole found no effects of greenspace, bluespace, or natural space percentage on antidepressant rates (25). Conversely, an ecological study in the Netherlands that used Bayesian geoadditive quantile regression to model exposure found high correlations between the presence of more greenspace and lower antidepressant prescription rates (205). These effects were not linear, however, and among areas in the lowest quantile of greenspace, increasing amounts were actually linked to higher rates.   The fact that only some forms of nature were associated with positive effects in our study is not a novel finding, and one that has grown increasingly common as this body of research has developed. This is particularly true when more objective mental health outcomes are selected, rather than self-reported mood or overall mental well-being. Other studies that have modeled    76  exposure to natural space based on percentages of greenspace or bluespace within a particular buffer have similarly resulted in null findings, including the area-level examination of antidepressant use across England described above (25). One potential explanation for these findings may be that treating this form of exposure as a continuous variable, as done in this study, obscures effects that only occur after a particular minimum dose has been reached (153,205). Exactly what constitutes the minimum and maximum dose of any particular form of natural space remains unclear, however. One study designed to define the minimum amount of nature for reducing rates of depression reported reductions occurring when as little as 15% of a neighborhood was vegetated, with increasing benefits seen until a maximum of 35% was reached (234). An urban greenspace indicator developed for the European Union also supports the notion of a minimum exposure to attain health benefits, setting one hectare as the smallest size for greenspaces to contribute toward achieving population-level goals (145), and the World Health Organization recommends at least eight square meters of public greenspace per capita within their City Prosperity Index (235), but neither guideline is based on epidemiological analyses.  Finally, we found that access to public greenspace within 400 meters was linked to a 13% increase for antidepressants, 10% for any dispensation, and 8% for anxiolytics, which was contrary to our initial hypotheses. In our study, only a small subset of both cases (12.3%) and controls (13.2%) did not meet criteria for this exposure metric, however, which may mean that these individuals differ with respect to an unmeasured spatially covarying factor that is the true source of the association with increased odds. For example, they could reside in suburban areas in which public greenspaces are generally reached by car, and therefore remain accessible, although at a larger buffer than 400 meters. This does not imply that the measure itself should be adjusted, however, because it was selected in alignment with local policy and planning objectives (28). Another explanation would be that basing analyses on such a small    77  proportion of the overall sample led to underpowered analyses. In a study that examined roles for both greenspace and bluespace in recovery from psychotic disorders in Utrecht, Boers et al. similarly reported an association contrary to hypotheses, but cited this statistical issue as a factor (236). The fact that confidence intervals for these models were significantly larger than those calculated for the majority of the other exposure metrics could lend support to this theory, but although larger, they were still relatively narrow.  3.4.1 Limitations Psychotropic medication use is an important outcome, and one that has direct relevance to health service planning and clinical care. However, incident prescription dispensation is not a direct proxy for incident disease, with estimates indicating that only 19% of Canadians with a lifetime history of mental illness also report psychotropic prescription use (237). Looking solely at individuals with MDD in a nationwide survey, Patten et al. reported that just one-third of the sample used antidepressant medications (238). In addition, this measure cannot be used to identify specific diagnoses among individuals who receive dispensations: in the same analysis by Patten et al., a sizable proportion of respondents reported atypical antidepressant use (6.4%) or sedative-hypnotic use (7.9%), neither of which were included in our case ascertainment criteria (see Appendix B for a complete list of included medications). The fact that medications are classified as antidepressants or anxiolytics also cannot be used to differentiate between prescriptions for depressive versus anxiety disorders, which can occur concurrently with or independently from depression and may be treated with similar medications (85). Unlike many other aspects of mental health service provision across Canada, however, access to necessary psychotropic prescriptions is quite high: an unmet need for medication was expressed by only 9% of respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH), far lower than unmet needs for counseling (36%) or information (31%), for instance (202).      78  The potential for residual confounding is also high due to the limited covariates contained in the available dataset, but this is a common issue with studies conducted using administrative data, both in British Columbia (67) and beyond (236,239). At the individual level, we were able to control for age and sex (via matching); at the neighborhood level, we adjusted for population density, walkability, and relative household income. Descriptive epidemiological surveys that have investigated the prevalence of demographic factors linked to differing rates of MDD across Canada identified all of these as predictors of disease, along with marital status and education (240). Systematic reviews that have looked more broadly at relationships between urban residence and mental health have identified a wide range of other factors that may be linked both to the environment and to the development or exacerbation of common mental disorders, including race/ethnicity, employment, and marital status at the individual level (241), and housing type, industrial land-use, and neighborhood quality (242) at the neighborhood level.   Despite the limitations of administrative data with respect to controls for confounding, this limitation is offset by the fact that the PharmaNet database reflects the population of the province and is comprehensive with respect to prescriptions, capturing details on all dispensations from community pharmacies as well as to outpatients in hospital and clinic settings. As a result, we were able to follow a semi-longitudinal design, in which all cohort members had a five-year run-up period without a history of antidepressant or anxiolytic prescription dispensation before case ascertainment. Although we did not have complete residential address histories for sample members, this criterion should reduce the potential for reverse causality. However, this could occur if individuals with good mental health moved to areas with higher levels of natural space exposure as a means of supporting their health, or if poor mental health led to underemployment (6) that pushed individuals to live in lower-SES areas with less natural space (38). Finally, relying on this dataset resulted in good power for our analyses, as indicated by the generally narrow confidence intervals.     79  One particular concern that may remain is the inability to directly adjust for individual SES. In the absence of a universal prescription access program in Canada, individuals with lower incomes may find this limits their ability to cover the costs of a written prescription, which would then go unfilled. However, the province of British Columbia does administer the Fair PharmaCare plan, which provides income-based prescription coverage on a sliding scale (243), reducing the potential impact of this factor. In addition, previous studies that have used our same measure of neighborhood income level as a proxy for individual SES found high correlations with both individual annual household income and educational level (244).  Another limitation of our study was the assignment of a single address based on the six-digit postal code, rather than a more-complex activity space calculation. For example, an individual may spend many hours of psychologically restorative time in a private backyard garden, but this would not be captured by either of our visible nature measures. Our measures of surrounding greenness would capture the presence of these areas, but not the extent to which they are accessible to or used by an individual. The reliance on residential postal codes also omits natural space exposures that may occur during commutes or time spent in school, work, or recreation, but which may represent important sources of restoration. Six-digit postal codes do offer a fairly precise estimate of residential location, however, generally reflecting mail delivery to a specific building or block-face in urban areas such as the study setting (147). This is also the most commonly available spatial scale for population-level health research, and efforts to generate additional urban exposure data by the Canadian Urban Environmental Health Research Consortium (CANUE) are based on the utility of this spatial scale for exposure assignment (245).         Finally, the majority of the exposure measures were calculated using data from 2014 to 2016, resulting in a potential temporal mismatch with prescriptions, particularly those that were    80  dispensed toward the beginning of the study period in 2008. Surrounding greenness was calculated on an annual basis, however, which may somewhat offset this concern.   3.5 Conclusions In this population-based case-control study, we reported evidence for reductions in antidepressant and anxiolytic prescription initiation associated with the presence of street trees and overall surrounding greenness. Many of the ten other included measures of natural space had no significant association with prescription dispensation, however, and access to public greenspace within 400 meters was associated with a 10% increase in the odds. Trees immediately surrounding residential locations showed the strongest beneficial effect, with the addition of ten trees within 100 meters linked to a 4% reduction in the odds of being dispensed an antidepressant and a 2% reduction for anxiolytics. Many cities around the world – including Vancouver, Madrid, and New York City – already have policies designed to expand their urban forest, and our findings add weight to calls to expand these policies to other locales. Additional research to elucidate the best method of integrating street trees into particularly dense environments could help ensure that efforts to increase densification as a means of increasing long-term sustainability do not come at the cost of such a valuable resource. Identifying neighborhoods and communities that are particularly vulnerable to poor mental health or have significantly lower-than-average street tree coverage may also be a useful target for future interventions, as a means of reducing inequity in the provision of this form of natural space and as a first step in allocating this resource to provide the greatest population health improvements.     81  Chapter 4: Exploring the role of sense of community belonging along the pathway linking natural space exposure to mental health 4.1 Background In an increasingly urban world (132), identifying evidence-based strategies to guide the design and maintenance of healthy cities is an essential function of public health (8). Urban living has been linked to certain health advantages, including better self-reported health, reduced mortality rates, and a narrowing of health inequities (246). By reducing the energy requirements for transportation and residential heating and cooling, a shift to denser urban environments may also lower greenhouse gas emissions and facilitate climate change mitigation (247). Conversely, dense urban living has also been tied to a number of environmental stressors, such as increased noise, crime, air pollution, and overcrowding (154,248,249). These stressors are, in turn, linked to increased odds of depression and anxiety (242). Furthermore, architectural designs that support density, such as high-rise towers, have been associated with higher levels of psychological distress and lower levels of social connection (250).  Mental disorders such as anxiety and depression are among the most common non-communicable diseases, and their prevalence is growing steadily, particularly in lower-income nations and among aging populations (166). Although standard approaches to estimating the burden of these disorders rank them among the top ten causes of years lived with disability (YLDs) worldwide (167), Vigo et al. assert that properly accounting for all associated morbidity, including non-lethal self-harm and chronic disease exacerbation, would increase this burden substantially, from 21% to 32% of global YLDs in 2013 (168).   Social isolation is another pressing concern, particularly in countries such as Canada that have large and growing numbers of older and elderly adults (121). A meta-analysis of 70 studies linked social isolation to a 29% increase in the likelihood of premature mortality (251). A    82  separate review identified cardiovascular disease as a pathway to such premature mortality, linking loneliness and social isolation to a 29% increase in the risk of developing coronary heart disease and a 32% increase in stroke incidence (252). These studies show the potential harms of the absence of social ties. Conversely, the presence of social ties has a critically important role in human well-being, particularly among individuals under stress (103). Social connections are especially important with respect to mental health because they offer expressive (e.g., sense of purpose), informational, and instrumental forms of support that can help to reduce anxiety, increase self-esteem, and provide positive influences regarding health-promoting behaviors, which may be more difficult to maintain during periods of psychological distress (102). Because social connections affect such a multitude of health outcomes and their impact may be cumulative across the lifespan, there have been calls to identify and enact population-level interventions and policies that work to strengthen these connections, particularly among groups with increased vulnerability to isolation and illness (253).  Natural space is increasingly recognized as one component of the urban fabric that can reduce the harms of exposure to stressors while facilitating social connections. Numerous studies have linked various forms of nature to mental health benefits across urban populations, including: reductions in chronic stress (86) and psychological distress (254); improvements in self-reported mental health (12,139); lower prevalence of depressive symptoms (153) and clinician-diagnosed anxiety and depressive disorders (255); and fewer antidepressant prescriptions (23,137,165,256). Not all studies have found such benefits, however: greenspace prevalence had no association with positive mental well-being among a nationally representative sample in England (24); access to bluespace had no association with multiple mental health outcomes in Barcelona, Spain (23); and greater park attractiveness was linked to increased psychological distress among residents of Perth, Australia (26).     83  Much like the evidence directly linking natural space to mental health improvements, the evidence regarding indirect influences via social connections is mixed. A recent study in the Catalonia region of Spain found no association between natural spaces and higher levels of social support, precluding potential mediation via this pathway (137). Yet, Fan et al. (2011) found divergent effects in Chicago, Illinois depending on the type of nature. Decreasing park proximity was positively associated with social support, but the association with higher surrounding greenness was negative, reducing the overall positive effect of nature on stress reduction, a component of psychological well-being (73). A Dutch study gives further indication that the definition of natural space matters with respect to mediation via social connections. It found that both the quality and quantity of streetscape greenery were independently associated with increased social cohesion and improved mental health status, but quantity had a larger indirect effect than quality (257). Together, these reports highlight how important the form of natural space may be when examining specific pathways. Higher levels of overall greenness may provide direct psychophysiological benefits, but may also reflect neighborhood designs that impede social connections; the accessibility of natural space may be crucial for enhancing social ties.   This analysis explores the relationship between exposure to multiple measures of natural space and three distinct mental health outcomes across the Vancouver, Canada census metropolitan area (CMA) via the application of components of the Natural Space Index (NSI) described in Chapter 2 (211). Our first objective was to assess whether there was an association between natural space and three mental health outcomes: 1) major depressive disorder; 2) negative mental health; and 3) psychological distress, as measured by a nationally representative survey. Our second objective was to assess whether any associations were mediated by sense of community belonging, which has been previously evaluated as an indicator of neighborhood-based social capital (258,259).    84  4.2 Materials and methods 4.2.1 Mental health data The study was conducted using data from the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH), a population-based, cross-sectional survey periodically undertaken as a complement to the annual Canadian Community Health Survey. Detailed information about the purpose and methods for the survey is available through Statistics Canada (3), but a brief summary is provided here. The CCHS-MH was designed specifically to collect data on both positive and negative mental health outcomes among Canadians. To ensure that estimates were reliable by province, sex, and four large age categories, the 2012 CCHS-MH used a multi-stage sampling strategy and aimed to enroll 27,500 participants across the nation. The target population comprised all Canadian residents over the age of 14, with the exception of individuals residing on First Nations reserves and other Aboriginal lands, full-time members of the Canadian Forces, and those institutionalized at the time of the survey. Together, these excluded groups comprised approximately 3% of the population (3). Data collection took place between January and December of 2012 using computer-assisted personal interviews conducted by field interviewers who were trained to question participants either in-person or over the phone (3). Almost all interviews (87%) were carried out in-person, with the remainder taking place by phone. No proxy interviews were permitted (3).  4.2.2 Study sample There were 25,113 interviews completed during the study collection period, for a combined household and individual response rate of 69% (3). Our natural space variables were only available for the 60,242 six-digit postal codes that were: 1) within the Vancouver CMA – which was home to 2,313,328 residents representing 6.91% of Canada’s total population in 2011 (260) – and 2) included in the Statistics Canada 2013 Postal Code Conversion File (146). As a result, the study sample was limited to respondents residing within the Vancouver CMA whose postal    85  codes were represented (stage 2 in Figure 4.1). As an estimate, based on the total number of completed interviews and the proportion of Canada’s residents living within the Vancouver CMA in 2011, approximately 1,700 survey respondents would have been expected to reside in the study area. Subsequent exclusions were made due to missing outcome (stage 3 in Figure 4.1) and mediator data (stage 4 in Figure 4.1). Finally, participants with missing data or unknown responses for any of the study variables were excluded (5%) from the study sample, resulting in a final weighted sample size of 1,930,048 (stage 5 in in Figure 4.1). In order to comply with privacy regulations, all analyses were conducted within the Statistics Canada Research Data Centre (RDC) and only weighted values were permitted for reporting.     Figure 4.1 Analytic sample development for the Canadian Community Health Survey-Mental Health (CCHS-MH) study This figure illustrates the criteria used at each stage of restricting the study sample, as well as the loss of respondents associated with each decision. Per Statistics Canada requirements, all figures are weighted.  4.2.3 Measures 4.2.3.1 Natural space exposure measures The NSI, which was developed specifically to assess the relationship between natural space and mental health across the Vancouver CMA, is described fully in Chapter 2. For this analysis,    86  the presence of greenspace was based on satellite measurements of the Enhanced Vegetation Index (EVI) from the MODIS instrument that were collected biweekly in 2012 at a 250-meter resolution (70,149). All other natural space exposure measures were developed using the methodology detailed in Chapter 2, Section 2.2.1.   Because the original NSI dataset contained 53 distinct measures at buffers ranging from 100 meters to 1,600 meters, hypotheses regarding the relevance of each individual measure and its most appropriate buffer size were explored in advance of data analysis. Due to the selection of six-digit postal code centroids as the primary unit of exposure assignment, these buffers represent radial distances from the postal code centroid. Based on initial exploration, seven measures were selected for inclusion in four categories, all described in greater detail in Chapter 3, Section 3.2.1.1. This is lower than the 12 measures included in Chapter 3 because a number of exposure variables were not available for all CCHS-MH participants, resulting in analyses that failed to meet the cell-size restrictions mandated by Statistics Canada.   Surrounding residential greenness 1) Residential Greenness 250m was defined as mean EVI in a 250-meter buffer, with bluespace polygons clipped before averaging to properly account for the negative EVI values created by water sources.   Visible nature 2) Visible Greenspace 100m was defined as greenspace percentage in a 100-meter buffer, and 3) Visible Natural Space 100m was defined as greenspace and bluespace percentage within a 100-meter buffer.     87  Accessible neighborhood nature 4) Accessible Greenspace 500m was defined as publicly accessible greenspace percentage within a 500-meter buffer and 5) Accessible Natural Space 500m was defined as public greenspace and bluespace percentage within a 500-meter buffer.  Binary access to nature  6) Binary Bluespace 1000m was based on the presence of bluespace within a 1,000-meter buffer, based on prior work showing health benefits of access to bluespace within this proximity (114,188), while 7) Binary Greenspace 400m was defined as an indicator of publicly accessible greenspace within a 400-meter buffer, based on the definition included in the City of Vancouver’s Greenest City 2020 Action Plan (28).  4.2.3.2 Mental health outcomes Three primary outcome variables were examined: major depressive disorder (MDD), negative mental health (NMH), and psychological distress. The variable Major Depressive Disorder was based on the CCHS-MH version of the World Health Organization’s Composite International Diagnostic Interview (WHO-CIDI), a standardized tool designed to be given by trained lay interviewers to assess mental disorders as part of epidemiologic studies (261) (see Appendix C for the complete wording of all CCHS-MH questionnaire items included in this study). The WHO-CIDI instrument aligns with the criteria and definitions of both the International Statistical Classification of Diseases and Related Health Problems, 10th Edition (ICD-10), which is used in medical coding and billing, and the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV), used by clinicians and other mental health service providers (3). In addition, the WHO-CIDI tool has shown high concordance with structured clinical interviews used for diagnosis in treatment settings (262). For the purposes of this study, we examined the derived variable for experiencing a major depressive episode in the 12 months prior to the interview (3).     88  The variable Negative Mental Health was derived from responses to the Mental Health Continuum Short Form (MHC-SF) developed by Keyes, a scale that includes 14 items related to emotional well-being and positive functioning (263). In Keyes’ conceptualization, emotional well-being comprises a constellation of measures related to positive feelings about life, including perceived life satisfaction and positive and negative affect. By contrast, positive functioning relates both to aspects of psychological well-being such as self-acceptance and environmental mastery and to social well-being, consisting of dimensions such as social integration and social contribution. A psychometric evaluation confirmed the MHC-SF’s reliability and validity within a population-level survey of residents of the Netherlands aged 18 to 87 (264), and a recent confirmatory factor analysis based on data from the 2012 CCHS-MH found support for the underlying model (265). For our study, the original three-level classification scheme (3) was recoded to either negative mental health (based on the categories of “moderate” or “languishing”) or positive mental health (based on the category of “flourishing”), to represent the two poles of Keyes’ continuum: incomplete versus complete mental health.   Lastly, the variable Psychological Distress was derived from the Kessler-10 Psychological Distress Scale (K10) and calculated as a continuous measure ranging from 0-40, with higher scores indicating increasing levels of distress. The K10 was originally developed to screen for serious mental illness in large health surveys in lieu of lengthier interviews (Kessler et al., 2003) and multiple validation studies have demonstrated high concordance with clinical appraisals in both general populations and groups with identified mental illness (266).  4.2.3.3 Mediation In order to assess potential mediation via social connections, the study integrated the sole relevant item included within the 2012 CCHS-MH: sense of community belonging. The variable Sense of Community Belonging (SoC) was derived from a single question asking, “How would    89  you describe your sense of belonging to your local community?” with response options of “very strong”, “somewhat strong”, “somewhat weak”, “very weak”, or “don’t know”. We excluded participants who did not answer the question or responded “don’t know” from the analytic sample. The SoC variable is widely used in Canadian health surveys, and prior research has demonstrated that it is most strongly associated with the domain of neighborhood-based social capital (258). A recent validation study compiled data from three separate Canadian health surveys with a total of more than 60,000 respondents. It corroborated the correlation between SoC and network-based social capital, while also noting its relationships to neighborhood satisfaction and place attachment (259). We hypothesized that increased values of the Accessible Greenspace 500m and Accessible Natural Space 500m variables would be associated with higher SoC, leading to an indirect effect of these natural space exposure measures on mental health improvements, because their accessibility at the neighborhood scale makes them the forms of nature most likely to contribute to neighborhood social cohesion.  4.2.3.4 Covariates All potential confounders were selected a priori based on relevant literature. If two potentially confounding variables were thought to be collinear, the percent of missing data for each variable was taken into consideration in determining which to include: for example, the World Health Organization’s Disability Assessment Schedule was originally assessed as a metric of functional limitations that might both limit use of natural spaces and increase the risk of poor mental health, but pain health status was selected instead due to its higher response rate among eligible participants. The final set of covariates comprised 11 variables: three demographic, two socioeconomic, two household, one physical health, and three urban design. Demographic variables were Sex, Age, and Race-Ethnicity. Socioeconomic status was reflected by Provincial Household Income Level, which compares the participant with others residing in the same province, and Highest Household Education Level. Household type included both Dwelling    90  Type, such as single-family home or apartment building, and Household Living Arrangement, indicating the presence of and relationship to others in the household. General health was represented by an ordinal Pain Health Status variable that reflects both the degree of pain generally felt by participants as well as the extent to which this pain interferes with their daily activities. Finally, urban design was represented by Urbanicity, Population Density, and Walkability. The latter two variables were assigned at the level of census dissemination areas (DAs), containing 400 to 700 residents (221), to represent the neighborhood scale. All covariates with the exception of Population Density (222) and Walkability (223) were drawn from the CCHS-MH questionnaire; these final two covariates were calculated using the methods described in Chapter 3, Section 3.2.1.3.   4.2.4 Statistical analysis All analyses were conducted in the Statistics Canada RDC at the University of British Columbia using SAS, version 9.4. First, descriptive statistics were computed for all variables. Second, conditional logistic regression models were used to explore the relationship between each of the natural space variables and the odds of Major Depressive Disorder and Negative Mental Health. Generalized logit regression models were then estimated for this same set of explanatory variables and SoC, the hypothesized mediator. Third, linear regression models were fitted to the Psychological Distress scores, which were treated as linear in alignment with previous studies that have examined the relationship between natural space and this measure of mental health (140). Each model was regressed on a single explanatory variable due to substantial correlations among explanatory variables (see Chapter 3, Section 3.3.2 for more detail) and in order to support research translation.   All models used sample weights to compute point estimates and 500 baseline replication weights to calculate confidence intervals, per recommendations by Statistics Canada. The    91  domain statement was also used within SAS to properly account for the overall stratified sampling design of the survey and the study restriction to individuals residing in the Vancouver CMA. We estimated unadjusted effects and estimates fully adjusted for the 11 covariates.  4.2.4.1 Assessment of mediation To test for mediation, we used current best practices to properly account for the complexity of relationships among explanatory, mediator, and outcome variables (267,268). This approach comprises assessing the relationships between: 1) explanatory variables and outcomes; 2) explanatory variables and mediators; and 3) mediators and outcomes. The indirect effect (ab) of an explanatory variable associated with a mental health outcome via the SoC mediating variable is computed from the relationships (slope coefficients) described in 2) and 3) above. Specifically, ab is the product of two slope coefficients, respectively estimating paths a and b: path a is the adjusted slope for the natural space variable of interest when regressed against the mediator (SoC), while path b is the adjusted slope for SoC when the mental health outcome of interest is regressed against the mediator (SoC) (as per 3). The statistical significance of ab was then determined using the zMediation test, which enables computing standardized indirect effects using estimates obtained from regression models with ordinal, categorical, and linear outcome variables (269).  4.3 Results 4.3.1 Sample characteristics The weighted sample was representative of 1,930,048 individuals, and evenly balanced across sexes and age categories (Table 4.1). The vast majority of analytic sample members resided in the urban center core (89%) because our sample only included residents of the Vancouver CMA. Slightly more than half lived in single, detached homes (52%) and almost two-thirds had received a certificate, diploma, or degree beyond the high-school level (64%). Half described themselves as White (50%) while almost one-third were East Asian (29%).    92  Table 4.1 Summary of sociodemographic factors in the Canadian Community Health Survey (CCHS-MH) sample Results presented for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)    Weighted %  Sex       Female 52.4      Male 47.6  Age   15-24 years 17.4 25-34 years 15.3 35-44 years 18.7 45-54 years   19.2 55-64 years   12.8 65 and older 16.5  Ethnicity       White 49.9      Aboriginal   2.1      East Asian 29.1      South or West Asian 13.9      Black, Latin American, Multiracial/Multiethnic, or Other   5.0  Household income level, relative to other British Columbia respondents  1st quintile (lowest income) 22.9 2nd quintile 23.5 3rd quintile 20.2 4th quintile 18.2 5th quintile (highest income) 15.1  Highest household education level  Less than secondary education 13.5 Secondary graduate or some post-secondary education 22.6 Post-secondary graduate 63.9  Dwelling type       Single, detached 52.4      Apartment building, mobile home, or other 19.3      Duplex, double, row, or terrace 28.3         93   Household living arrangement  Unattached, living alone 17.1 Unattached, living with others; or Other 19.4 Couple 20.6 Two-parent family 33.9 Single-parent family   9.0  Pain health status  No pain 79.0 Doesn’t prevent activities   6.7 Prevents a few activities   5.5 Prevents some activities   5.9 Prevents most activities   2.9  Urbanicity       Urban center core 89.3      Non-urban center core 10.7   Weighted mean (SE)  Population density (persons per km2) 7,466.26 (317.58)  Neighborhood walkability (Walk Score® summary)  60.18 (0.92)  SE = standard error All means, percentages, and SEs are rounded  4.3.2 Natural space presence, form, and accessibility Access to both greenspace and bluespace was fairly high: 44% of sample members had bluespace within a 1,000-meter buffer, while 88% met the City of Vancouver’s criterion for access to public greenspace within a 400-meter buffer (Table 4.2). On average, within a 100-meter buffer, 4.3% of the area consisted of greenspace (95% CI = 3.6%, 4.9%), which increased to 4.5% (95% CI = 3.8%, 5.2%) when both greenspace and bluespace were included. Within the 500-meter buffers, these values increased to 9.4% (95% CI = 8.6%, 10.1%) and 11.0% (95% CI = 10.2%, 11.9%), respectively. After clipping bluespace polygons, the mean annual EVI value within a 250-meter buffer was 0.23 (95% CI = 0.23, 0.24).      94  Table 4.2 Summary of natural space exposures in the CCHS-MH sample  Results presented for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)    Weighted mean (SE) Surrounding greenness (250m)       Residential Greenness 0.23 (0.0) Visible nature (% in 100m)  Visible Greenspace 4.3 (0.3) Visible Natural Space 4.5 (0.3) Accessible neighborhood nature (% in 400m)       Accessible Greenspace 9.4 (0.4)      Accessible Natural Space 11.0 (0.4)    Weighted % (SE) Binary access to nature (% yes)  Binary Bluespace 1000m  43.9 (2.0) Binary Greenspace 400m 88.3 (1.2)  SE = standard error All means, percentages, and SEs are rounded   4.3.3 Mental health outcomes and sense of community belonging The criteria for Major Depressive Disorder were met by 4.2% of the sample, and 25.5% were characterized as having Negative Mental Health. Rates of both conditions were higher among women (4.9% and 27.0%, respectively) than men (3.5% and 23.9%). Higher-than-average rates of both outcomes were also seen among those living alone (6.3% and 28.1%) and those reporting the most functional interference due to pain (18.4% and 50.8%). The mean Psychological Distress scale score was 4.9 (95% CI = 4.5, 5.3), well below the cut-off of 13 for serious mental illness (262). Most respondents reported very strong (17.3%) or somewhat strong (48.9%) SoC, with only 6.8% falling into the lowest category of very weak SoC.     95  4.3.4 Associations between natural space measures and mental health outcomes There were no consistent associations between any of the seven natural space measures and the three mental health outcomes in the adjusted (Table 4.3) or unadjusted models (Table 4.4). For Major Depressive Disorder, the point estimate for the Binary Bluespace 1000m variable was elevated and in the opposite of the hypothesized direction (aOR = 1.72; 95% CI = 0.85, 3.49), and the same was true for Binary Greenspace 400m (aOR = 1.20; 95% CI = 0.47, 3.07); however, the confidence intervals for both measures were quite wide and included 1.0. In the anticipated direction, the point estimate for the odds of Major Depressive Disorder associated with Residential Greenness 250m (based on each 0.1-point increase in EVI) was 0.80 (95% CI = 0.45, 1.42); however, the confidence interval included 1.0. For Negative Mental Health, all of the natural space measures had adjusted odds ratios around 1, with the exception of Binary Bluespace 1000m and Binary Greenspace 400m, which suggested increased odds of having the condition, with aORs of 1.20 (95% CI = 0.81, 1.76) and 1.34 (95% CI = 0.75, 2.42), respectively, although the related confidence intervals were wide. A similar picture emerged for Psychological Distress, with coefficients close to zero for all measures except those related to the binary access to nature variables. There were slight increases for both Binary Bluespace 1000m (b = 0.68, 95% CI = -2.04, 3.40) and Binary Greenspace 400m (b = 0.57, 95%CI = -1.79, 2.93), but the confidence intervals spanned zero.     96     Table 4.3 Adjusted results for natural space exposures and mental health outcomes for the CCHS-MH sample Results of adjusted+ models examining natural space exposures and major depressive disorder, negative mental health status, and psychological distress for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)  Major  Depressive Disorder  Negative  Mental  Health Psychological Distress  (Kessler-10 scale)   OR (95% CI)  OR (95% CI)  b (95% CI) Surrounding greenness (250m)         Residential Greennessa 0.80 (0.45, 1.42) 1.07 (0.80, 1.44) 0.05  (-0.23, 0.83)  Visible nature (% in 100m)         Visible Greenspaceb 1.027 (0.999, 1.055) 1.013 (0.999, 1.026) 0.02  (-0.02, 0.06)       Visible Natural Spaceb 1.026 (0.998, 1.055) 1.012 (0.999, 1.025) 0.02  (-0.03, 0.06) Publicly accessible neighborhood nature (% in 500m)          Accessible Greenspaceb 0.984 (0.950, 1.020) 1.008 (0.991, 1.025) -0.01 (-0.05, 0.02)       Accessible Natural Spaceb  0.999  (0.968, 1.030) 1.006 (0.991, 1.020) -0.02 (-0.05, 0.01)  Binary access to nature (% yes)          Binary Bluespace 1000mc 1.72  (0.85, 3.49) 1.20  (0.81, 1.76) 0.68 (-2.05, 3.40)       Binary Greenspace 400mc 1.20 (0.47, 3.07) 1.34  (0.75, 2.42) 0.57  (-1.79, 2.93)  + Each model includes a single explanatory variable and has been adjusted for sex, age, race-ethnicity, household income level, household education level, dwelling type, living arrangement, pain health status, urbanicity, population density, and walkability a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope coefficient All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded    97    Table 4.4 Crude results for natural space exposures and mental health outcomes for the CCHS-MH sample Results of unadjusted+ models examining natural space exposures and major depressive disorder, negative mental health status, and psychological distress for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)  Major  Depressive Disorder  Negative  Mental  Health Psychological Distress  (Kessler-10 scale)   OR (95% CI)  OR (95% CI)  b (95% CI) Surrounding greenness (250m)         Residential Greennessa 0.71 (0.52, 0.98) 1.08 (0.90, 1.30) -0.01  (-0.47, 0.45)  Visible nature (% in 100m)         Visible Greenspaceb 1.025 (0.995, 1.055) 1.012 (0.999, 1.025) 0.03 (-0.02, 0.07)       Visible Natural Spaceb 1.024 (0.993, 1.055) 1.012 (0.999, 1.025) 0.02  (-0.02, 0.07) Publicly accessible neighborhood nature (% in 500m)          Accessible Greenspaceb 0.982 (0.954, 1.011) 1.006 (0.990, 1.023) -0.02 (-0.05, 0.02)       Accessible Natural Spaceb  0.995  (0.971, 1.021) 1.002 (0.988, 1.015) -0.03 (-0.05, 0.00)  Binary access to nature (% yes)          Binary Bluespace 1000mc 2.40  (1.26, 4.59) 1.15 (0.81, 1.62) 1.00 (-2.97, 4.97)       Binary Greenspace 400mc 1.43 (0.57, 3.62) 1.16 (0.68, 2.00) 0.71  (-2.28, 3.71)  + Each model includes a single explanatory variable  a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope coefficient All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded    98  4.3.5 Associations between natural space measures and sense of community belonging  Consistent with our original hypotheses, we found associations between both measures of accessible neighborhood nature and sense of community belonging as well as a trend in effect sizes, in which higher values of the accessible neighborhood nature variables were associated with increasing odds of reporting higher levels of SoC, although many of the confidence intervals for lower levels of SoC included 1.0 (Table 4.5). In adjusted regression models, for each 1% increase in Accessible Natural Space 500m there was a 5% increase in the odds of reporting a “very strong” SoC (OR = 1.05; 95% CI = 1.01, 1.10) and a 4% increase in the odds of a “somewhat strong” SoC (OR = 1.04; 95% CI = 1.01, 1.08) in comparison with the reference category of “very weak”, with an OR of 1.029 (95% CI = 0.991, 1.067) for “somewhat weak” SoC. Similar results were seen for Accessible Greenspace 500m, and effect sizes for both measures were largely the same in unadjusted models (Table 4.6). Based on the point estimates, there was also some evidence of association between higher levels of SoC and the Binary Bluespace 1000m and Binary Greenspace 400m variables: the adjusted odds ratios were greater than 1 for all three levels in comparison with the reference group of “very weak” but with wide confidence intervals that still included 1.0. Conversely, no associations were evident with the Residential Greenness 250m, Visible Greenspace 100m, or Visible Natural Space 100m measures, with all confidence intervals spanning 1.0 in both adjusted and unadjusted models.     99   Table 4.5 Adjusted results for natural space exposures and sense of community belonging for the CCHS-MH sample   Results of adjusted+ models examining natural space exposures and sense of community belonging level for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)     Sense of community belonging level (reference = very weak)  Somewhat weak  Somewhat strong  Very  strong   OR (95% CI)  OR (95% CI)  OR (95% CI) Surrounding greenness (250m)         Residential Greennessa 1.13  (0.74, 1.74) 1.03  (0.68, 1.56) 0.86  (0.50, 1.48)  Visible nature (% in 100m)          Visible Greenspaceb 0.999  (0.971, 1.028) 1.009  (0.982, 1.036) 1.008  (0.978, 1.038)        Visible Natural Space 100mb 0.999  (0.972, 1.028) 1.010  (0.984, 1.038) 1.009 (0.980, 1.039)  Publicly accessible neighborhood nature          Accessible Greenspace 500mb 1.033 (0.996, 1.082) 1.038 (0.990, 1.089) 1.051 (1.002, 1.103)       Accessible Natural Space 500mb  1.029 (0.991, 1.067) 1.042 (1.005, 1.081) 1.052 (1.011, 1.095)  Binary access to nature          Binary Bluespace 1000mc 1.60 (0.84, 3.07) 1.42 (0.76, 2.67) 1.69 (0.83, 3.42)       Binary Greenspace 400mc 1.18  (0.48, 2.91) 1.36  (0.59, 3.13) 1.07  (0.39, 2.96)  + Each model includes a single explanatory variable and has been adjusted for sex, age, race-ethnicity, household income level, household education level, dwelling type, living arrangement, pain health status, urbanicity, population density, and walkability a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = adjusted odds ratio; CI = confidence interval  All odds ratios (OR) and 95% confidence intervals (95% CI) are rounded    100     Table 4.6 Crude results for natural space exposures and sense of community belonging for the CCHS-MH sample   Results of unadjusted+ models examining natural space exposures and sense of community belonging level for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)     Sense of community belonging level (reference = very weak)  Somewhat weak  Somewhat strong  Very  strong   OR (95% CI)  OR (95% CI)  OR (95% CI) Surrounding greenness (250m)         Residential Greennessa 1.05  (0.71, 1.56) 1.07  (0.73, 1.57) 0.93  (0.60, 1.43)  Visible nature (% in 100m)          Visible Greenspaceb 1.000  (0.973, 1.028) 1.009  (0.982, 1.036) 1.003  (0.975, 1.032)        Visible Natural Space 100mb 1.001  (0.974, 1.029) 1.010  (0.984, 1.038) 1.005 (0.977, 1.034)  Publicly accessible neighborhood nature          Accessible Greenspace 500mb 1.034 (0.988, 1.081) 1.040 (0.996, 1.087) 1.047 (1.002, 1.095)       Accessible Natural Space 500mb  1.032 (0.995, 1.069) 1.042 (1.007, 1.078) 1.047 (1.010, 1.085)  Binary access to nature          Binary Bluespace 1000mc 1.41 (0.82, 2.45) 1.18 (0.71, 1.94) 1.30 (0.72, 2.34)       Binary Greenspace 400mc 0.999  (0.44, 2.26) 1.28  (0.62, 2.65) 0.94  (0.38, 2.32)  + Each model includes a single explanatory variable a Models based on a 0.1-unit increase in EVI b Models based on a 1% increase in area within buffer c Models based on having access to nature OR = adjusted odds ratio; CI = confidence interval  All odds ratios (OR) and 95% confidence intervals (95% CI) are rounded    101  4.3.6 Associations between sense of community belonging and mental health outcomes  The SoC variable was associated with improvements across all three mental health outcomes in adjusted regression models (Table 4.7). In fully adjusted models, there was an 86% decrease in the odds of Major Depressive Disorder among individuals reporting a “very strong” SoC (aOR = 0.14; 95% CI = 0.03, 0.66) and a 78% decrease among individuals with a “somewhat strong” SoC (OR = 0.22; 95% CI = 0.08, 0.63) in comparison with the reference category of “very weak”; the OR among individuals with a “somewhat weak” SoC was 0.41 (95% CI = 0.15, 1.10). For Negative Mental Health, the decreases were 91% (OR = 0.09; 95% CI = 0.03, 0.22) and 78% (OR = 0.22; 95% CI = 0.12, 0.39) for “very strong” and “somewhat strong” SoC, respectively, with an OR of 0.69 (95% CI = 0.37, 1.28) for “somewhat weak”. Finally, for Psychological Distress, the two highest levels of SoC were associated with reductions in K10 scores in comparison with the reference category of “very weak”: a 2.8-point reduction [-4.4, -1.2] for those in the “very strong” category and a 2.0-point reduction [-3.5, -0.6] for those in the “somewhat strong” category, with a beta of -1.1-point [-2.5, 0.4] for those in the “somewhat weak” category. These general relationships were also seen in unadjusted models (Table 4.8) – although many effect sizes were smaller – as well as in models that were adjusted for the natural space exposure measures associated with sense of community belonging along with all confounding variables (Table 4.9).      102   Table 4.7 Adjusted results for sense of community belonging and mental health outcomes for the CCHS-MH sample  Results of adjusted+ models examining sense of community belonging level and major depressive disorder, negative mental health status, and psychological distress for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)    Major  Depressive Disorder  Negative  Mental  Health Psychological Distress  (Kessler-10 scale)   OR (95% CI)  OR (95% CI)  b (95% CI) Sense of community belonging level         Very weak Ref Ref Ref      Somewhat weak 0.41  (0.15, 1.10) 0.69 (0.37, 1.28) -1.06  (-2.54, 0.42)      Somewhat strong 0.22 (0.08, 0.63) 0.22  (0.12, 0.39) -2.04  (-3.49, -0.59)      Very strong 0.14 (0.03, 0.66) 0.09 (0.03, 0.22) -2.79  (-4.39, -1.19)  + Each model has been adjusted for sex, age, race-ethnicity, household income level, household education level, dwelling type, living arrangement, pain health status, urbanicity, population density, and walkability OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope coefficient All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded    103   Table 4.8 Crude results for sense of community belonging and mental health outcomes for the CCHS-MH sample  Results of unadjusted+ models examining natural space exposures and sense of community belonging level for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)    Major  Depressive Disorder Negative  Mental  Health Psychological Distress  (Kessler-10 scale)   OR (95% CI)  OR (95% CI)  b (95% CI) Sense of community belonging level         Very weak Ref Ref Ref      Somewhat weak 0.43  (0.18, 1.05) 0.64 (0.34, 1.19) -1.52  (-3.20, 0.17)      Somewhat strong 0.25 (0.10, 0.61) 0.24  (0.13, 0.43) -2.60  (-4.18, -1.07)      Very strong 0.15 (0.02, 1.18) 0.10 (0.04, 0.23) -3.72  (-5.44, -1.20)  + Each model includes a single explanatory variable  OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded    104   Table 4.9 Multiply adjusted results for sense of community belonging and mental health outcomes for the CCHS-MH sample  Results of multiply adjusted+ models examining sense of community belonging level and major depressive disorder, negative mental health status, and psychological distress for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)     Major  Depressive Disorder  Negative  Mental  Health Psychological Distress  (Kessler-10 scale)   OR (95% CI)  OR (95% CI)  b (95% CI) Sense of community belonging level (adjusted for Accessible Greenspace 500m)         Very weak Referent Referent Referent      Somewhat weak 0.42  (0.16, 1.13) 0.67 (0.35, 1.26) -1.05  (-2.53, 0.43)      Somewhat strong 0.23 (0.08, 0.65) 0.21  (0.11, 0.38) -2.02 (-3.47, -0.58)      Very strong 0.14 (0.03, 0.70) 0.08 (0.03, 0.21) -2.77  (-4.37, -1.17)    Sense of community belonging level (adjusted for Accessible Natural Space 500m)         Very weak Referent Referent Referent      Somewhat weak 0.40  (0.15, 1.09) 0.66 (0.35, 1.26) -1.03  (-2.52, 0.45)      Somewhat strong 0.22 (0.08, 0.61) 0.20  (0.11, 0.38) -1.99 (-3.44, -0.55)      Very strong 0.13 (0.03, 0.66) 0.08 (0.03, 0.21) -2.74  (-4.34, -1.13)   + Each model has been adjusted for sex, age, race-ethnicity, household income level, household education level, dwelling type, living arrangement, pain health status, urbanicity, population density, and walkability, as well as for the specified natural space exposure OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded    105  4.3.7 Testing the indirect effect of sense of community on mental health outcomes The two criteria for proceeding to a zMediation test of the significance of an indirect effect were met for both the Accessible Greenspace 500m and the Accessible Natural Space 500m variables and at least one higher level of SoC in comparison with the “very weak” reference group: 1) each of these explanatory variables was associated with the hypothesized mediator (Table 4.5); and 2) the hypothesized mediator was associated with at least one mental health outcome (after controlling for the explanatory variable and all other covariates; reported in Table 4.9). Based on these results (Table 4.10), at an alpha of 0.05, there was a statistically significant indirect effect of Accessible Greenspace 500m on lowered Psychological Distress for individuals reporting a “very strong” SoC and those reporting a “somewhat strong” SoC; this was also true for Accessible Natural Space 500m. Looking at Negative Mental Health, the test was significant for both Accessible Greenspace 500m and Accessible Natural Space 500m, but only among those with a “somewhat weak” SoC. There were no significant indirect effects of either natural space exposure measure on the odds of Major Depressive Disorder.      106   4.4 Discussion This study explored associations between seven measures of natural space exposure and three mental health outcomes, and then tested potential mediation via social cohesion using population-level health survey data. We found no evidence for a direct association in the expected direction between any of seven measures of natural space with Major Depressive Disorder, Negative Mental Health, or Psychological Distress. However, higher percentages of Accessible Greenspace 500m and Accessible Natural Space 500m were linked to higher levels of SoC, which were, in turn, associated with improvements across all three mental health measures (Figure 4.2). Using the zMediation test, we found significant indirect effects of Table 4.10 Results of zMediation test for the CCHS-MH sample  Results of zMediation test+ of the indirect effect of sense of community belonging level on associations between natural space exposures and mental health outcomes for the weighted n of 1,930,048 Vancouver CMA respondents to the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH)  Major Depressive Disorder Negative  Mental  Health Psychological Distress  (Kessler-10 scale)  ZMediation ZMediation ZMediation  Accessible Greenspace 500ma Sense of community belonging         Very weak Referent Referent Referent      Somewhat weak 0.82 2.05* -1.39      Somewhat strong 0.43 0.67 -2.75*      Very strong 0.17  0.17 -3.40* Accessible Natural Space 500ma Sense of community belonging         Very weak Referent Referent Referent      Somewhat weak 0.79 2.03* -1.37      Somewhat strong 0.41 0.65 -2.70*      Very strong 0.16  0.16 -3.35* + Tests were based on the results of regression models reported in Table 4.5 and Table 4.9 using the formula detailed in Iacobucci (2012). All zMediation test scores are rounded. a Models based on a 1% increase in area within buffer * Statistically significant at an α = 0.05 All values are rounded    107  Accessible Greenspace 500m and Accessible Natural Space 500m on both reduced Psychological Distress and lower odds of Negative Mental Health via higher levels of SoC. This finding highlights the importance of social ties for mental well-being and supports the results of earlier studies that demonstrated a role for neighborhood natural space in various measures of social connections (73,166,167), including others that have used 500-meter buffers to define neighborhoods (270,271). In addition, it indicates the potential importance of publicly accessible neighborhood nature in addressing issues of both social isolation and poor mental health in dense urban environments.      108  Direct Effects    OR of SoC (95% CI) Very Strong: 1.05 (1.00,1.10) Somewhat Strong: 1.04 (0.99,1.09) Somewhat Weak: 1.03 (1.00,1.08) OR of SoC (95% CI) Very Strong: 1.05 (1.01,1.10) Somewhat Strong: 1.04 (1.01,1.08) Somewhat Weak: 1.03 (0.99,1.07)   Indirect Effects    b of Psychological Distress (95% CI) Very Strong: -2.79 (-4.39,-1.19) Somewhat Strong: -2.04 (-3.49,-0.59) Somewhat Weak: -1.06 (-2.54,0.42)  OR of NMH (95% CI) Very Strong: 0.09 (0.03,0.22) Somewhat Strong: 0.22 (0.12,0.39) Somewhat Weak: 0.69 (0.37,1.28)  Figure 4.2 Direct and indirect relationships among exposures, mediators, and outcomes for the CCHS-MH study This figure summarizes the direct and indirect associations found among publicly accessible neighborhood nature, sense of community belonging, and all mental health outcomes. Arrows depict statistically significant relationships, while dashes indicate a lack of statistically significant effects. Exposure variables appear in green, mediating variables in orange, and outcome variables in blue.   The lack of direct associations between urban nature and improved mental health adds to the mixed results arising from population-level studies. The extent to which existing evidence is Accessible Greenspace 500mIncreased Sense of Community BelongingAccessible Natural Space500mIncreased Sense of Community BelongingSense of Community BelongingReduced Psychological DistressAccessible Greenspace & Natural Space500mSense of Community BelongingReduced Odds of Negative Mental HealthAccessible Greenspace & Natural Space500m   109  unclear is well described in Gascon et al.’s (2015) systematic review. Gascon’s work points to the following factors as potential causes of variability in the relationships between urban nature and mental health: 1) lack of consensus regarding how best to define and capture exposure to nature; 2) a lack of clarity with respect to the relationship between pathways and forms; and 3) the failure to explore how other elements of the built environment may influence exposure to natural space (12). Our results highlight some of these potential factors, as well as pointing to the necessity of exploring indirect pathways, such as the measure of social cohesion examined here.  The potential impact of other elements of urban design is supported by our findings that unadjusted models indicated a role for Residential Greenness 250m in reduced odds of Major Depressive Disorder (Table 4.4), but that models adjusted for both Population Density and Walkability showed no such effect (Table 4.3). This may indicate that some of the health benefits previously ascribed to surrounding greenness are actually due to other aspects of the urban form, such as the design of street networks or the presence of large open areas on private property. Although the extent of such unmeasured confounding is challenging to quantify, a recent study of more than one million Canadians linked a one interquartile range increase in NDVI to reductions of 9-10% in the hazard of premature mortality (210). The same study found that controlling for population density reduced hazard ratios by 4%, as much as controlling for levels of the air pollutant nitrogen dioxide. Walkability may be particularly important to consider when social interaction is the pathway of interest. Residents of highly walkable neighborhoods in Galway, Ireland had 80-95% higher odds of high social capital than those in less walkable neighborhoods (272), and individual components of walkability were linked to a greater sense of community belonging among adult residents of Atlanta, Georgia (273). Conversely, adjusting for these same variables but with a minimal set of confounders otherwise actually strengthened the association between Residential Greenness 250m and the    110  odds of antidepressant dispensation, as described in Chapter 2, shifting from an OR of 1.001 with a confidence interval that included 1.0 to an aOR of 0.97 (95% CI = 0.96, 0.98), indicating a 3% reduction.  Another potential explanation for our results is our use of objective outcomes. One of the few efforts to examine clinical diagnoses of anxiety and depression found a relatively weak association between these conditions and the percentage of large natural spaces within a one-kilometer buffer, with a 10% increase over the average associated with a 4% reduction in the odds of depression and a 5% reduction in the odds of anxiety (115). In addition, a number of studies that have found positive associations with objective mental health outcomes have relied on much coarser geographic scales than the six-digit postal code. Boroughs were used as the unit of analysis in an ecological assessment of antidepressant prescriptions in London, England (165). Another English study used Lower Layer Super Output Areas (with up to 3,000 residents) (274) to link increased greenspace to improvements on the GHQ-12, a screening tool for mood disorders (20).  Yet another relevant factor is the heterogeneity of exposure measures used in earlier studies examining natural space. For instance, multiple reports have pointed to greater mental health improvements from bluespace as compared with greenspace (52). These include a qualitative study conducted with older adults from the Vancouver region (106) and a more-recent study examining psychological distress in Wellington, New Zealand (140). Although we also observed differences between bluespace and greenspace, our measure of Binary Bluespace 1000m was linked to increased, rather than decreased, odds of Major Depressive Disorder and adding bluespace to the measure of Accessible Greenspace 500m (i.e., the Accessible Natural Space 500m variable) had little impact on its association with SoC.  Examining psychotropic prescription dispensations in Chapter 3, we also found little impact of adding bluespace to either    111  Accessible Greenspace or Visible Greenspace; setting bluespace areas to the end of the EVI scale also had only a small effect on estimates for Surrounding Greenness in that study, although it was in a negative direction. The results for Binary Bluespace 1000m differ to a greater degree (Table 3.8, Table 3.9, Table 3.10): this measure of access was linked to reduced odds of prescription dispensation in unadjusted models, but associated with increased odds of antidepressant prescriptions after adjustment for a limited set of confounders (aOR = 1.019; 95% CI = 1.003, 1.036).   Our results regarding the relationship between publicly accessible neighborhood nature and sense of community belonging also support calls for increased attention to the importance of selecting natural forms in light of relevant pathways (12,257). The differential effects of specific natural forms on measures of social connections may also explain the null findings seen in other studies that have explored this pathway, including one that looked at both surrounding greenness and access to greenspace and bluespace within a 300-meter buffer, but did not account for private versus public access (137). The importance of accessibility has also been identified with respect to mental health benefits, including a study that found 42% lower odds of psychological distress among individuals who reported that greenspaces were accessible versus inaccessible (254).  Our study is not the first to find evidence for an indirect effect of natural space on mental health improvements via social connections in the absence of a direct effect. Using residential tree canopy cover within a 250-meter buffer as its definition of nature, a cross-sectional study based on data from the California Health Interview Survey found no associations with psychological distress. However, it did report an association with the odds of reporting higher levels of perceived neighborhood social cohesion, derived from a three-question index regarding the neighborhood social environment, although a relatively small one, with an OR of 1.014 (95% CI    112  = 1.004, 1.023) associated with a 10% increase in tree cover. Even at this effect size, however, neighborhood social cohesion mediated 11% of the overall relationship between tree canopy and general health (166). Another study investigated multiple mediators of the relationship between three distinct measures of natural space exposure and subjective general health. It found evidence for mediation via perceived social support for both surrounding greenness (based on NDVI within 100-meter, 250-meter, and 500-meter buffers) and subjective proximity to greenspace (based on participants’ reports of having a park within a ten-minute walk of home) – with increased odds of 15-22% associated with the former and 31% with the latter – as well as indicating that 500-meter buffers were most relevant for social connections (270). Our study adds specificity to these findings by focusing on objective, clinically relevant mental health outcome measures rather than general health. Although it is somewhat challenging to compare results across studies due to the varying exposure types and measures of social connections, our effect sizes are close to those found by Ulmer et al. (166), but substantially lower than those reported in Dadvand et al.’s study (270), which may reflect the tighter alignment between the constructs of sense of community belonging and neighborhood social cohesion (259).  In addition to the use of objective mental health outcomes that align with clinical diagnoses, our study has a number of strengths. Principal among these is the use of data from the CCHS-MH, a health survey that was specifically designed to examine mental health across a representative sample of the population. These data also allowed us to incorporate a large number of potential confounders, adding to the interpretability of adjusted models. The selection of natural space measures and buffer sizes based on distinct pathways is another strength, as is the inclusion of measures that align with local sustainability plans. Finally, our use of best practices for assessing mediation (267,268) via the zMediation test (269) represents an advancement over the many other studies that have examined potential pathways linking nature to mental health using Baron and Kenny’s classical four-step approach (154,204,270,275).    113  4.4.1 Limitations There are several limitations to this study. Most importantly for a study examining mediation, we are unable to address causality due to the cross-sectional nature of the CCHS-MH. In addition, even though we controlled for walkability and population density, the potential for unmeasured confounding via built-environment factors remains. This is particularly true because the reliance on six-digit postal codes to assign exposure did not allow us to assess the role of street networks in accessibility to natural spaces. Overall, however, using six-digit postal codes as a proxy for residential location results in little spatial error in urban regions within Canada (147).  Our ability to broadly address potential mediation via social connections is also restricted by the reliance on the sole relevant variable within the CCHS-MH, sense of community belonging, which is based on responses to a single Likert-scale questionnaire item. Although the measure is widely used in Canadian population health studies and has been corroborated using data on more-discrete measures of social capital embedded in other population-based studies (259), it is still reflective of just one component of the complex array of social ties that are critical to a range of health domains (102). Ideally, future efforts will be able to integrate measures reflecting specific aspects of social connectedness, allowing for additional insights into this causal pathway.  In addition, although interviews for the CCHS-MH survey were carried out between January and December of 2012, the natural space measures were based on data collected between 2014 and 2016, leading to a temporal mismatch. This may have resulted in misclassification error for a measure such as Binary Greenspace 400m, which could potentially include public park space unavailable at the time of the survey. However, a measure such as Binary Bluespace 1000m that primarily reflects large bodies of water is unlikely to be impacted, and studies that have    114  examined residential greenness in cohort studies have found that this measure of natural space also remains stable over time (276).  More greenspace may make neighborhoods more attractive, which may lead to escalating housing costs. This can cause upward shifts in area-level economic status that result in displacement via gentrification (38), making this another potential source of unmeasured confounding via area-level socioeconomic status. Unfortunately, the 2012 CCHS-MH omits any relevant area-level measures, and data quality from the 2011 Canadian Census long form was compromised (182), which prevented us from assigning such measures from an external source. The CCHS-MH survey does include a robust set of household SES measures, and we adjusted for these in our models. Previous studies have also pointed to potential moderation via urbanicity, but the low proportion of individuals residing outside the urban core precluded full exploration of this issue.   This limitation primarily affects the generalizability of our findings, as does the fact that Vancouver is a region with relatively high levels of naturalness, as demonstrated by the high average EVI and large percentages of participants with access to both greenspace and bluespace. Our mean value for residential surrounding greenness within a 250-meter buffer was 0.23, for instance, while an effort in Barcelona that also explored associations between this exposure metric and social interaction reported a mean almost 20% lower, at 0.19 (270). However, our rate of access to public greenspace within 400 meters conforms closely with those found in a multi-city effort to develop an urban greenspace indicator for Europe, at 88% in comparison to their range of 86% to 97% depending on the locale in question (145). These relatively high levels of natural space exposure also resulted in low contrast for the binary access measures, which may have led to a lack of power to robustly detect their effects, as demonstrated by the large effect estimates and wide confidence intervals. Because a wide    115  range of sociodemographic factors are known to affect self-reported sense of community belonging – including gender, age, individual and area-level income, and housing design (259) – caution is also warranted in applying the mediation results to cities with markedly different demographic profiles. In general, because urban nature is always embedded in complex physical and social environments, we would argue that the body of literature regarding the impacts of urban natural spaces on health is inherently tied to place.   4.5 Conclusions Our study did not find evidence for a direct effect of exposure to urban natural space on mental health outcomes in the expected direction. However, it did provide evidence for the role of natural space in facilitating social connections and for providing indirect mental health benefits via increased sense of community belonging. Positive social connections may be particularly vital for urban residents in comparison with their suburban or rural counterparts due to the increased spatial proximity of social networks (77). Identifying environmental supports for social interaction is crucially important in Vancouver, where residents consistently rank social isolation as one of their biggest concerns (277). Future research should explore whether these impacts can be ascribed to all members of an urban population equally – in particular, the extent to which members of groups with additional vulnerability to isolation may benefit – and further elucidate how best to integrate nature into the urban fabric in order to maximize its positive effects for all urban residents.    116  Chapter 5: Conclusion 5.1 Research summary and contributions This dissertation applied interdisciplinary methods drawn from the fields of geography, sociology, and epidemiology to develop and apply a robust model of natural space exposure for the Vancouver census metropolitan area (CMA) to evaluate impacts on mental health. By collecting and utilizing data on the presence, form, accessibility, and quality of both greenspace and bluespace, the Natural Space Index (NSI) represents a novel approach to exposure assessment. It presents a detailed picture across an urban Canadian region, contributes to the ongoing debate about best practices for exposure assignment, and creates a robust set of metrics at the six-digit postal-code level with direct utility for analyses based on administrative and survey data. The application of this model to psychotropic prescription dispensation produced a unique analysis based on individual-level exposures and outcome data, advancing prior efforts that employed ecological designs or relied upon self-reported medication use. Finally, the integration of results from a nationally representative survey with measures of both mental health and social connections illuminated the complex relationships among distinct measures, objective outcomes, and potential pathways, while taking advantage of modern approaches to mediation analysis.   Both of these latter two analyses indicated a role for higher surrounding greenness in mental health improvements, with each 0.1-point increase in the Enhanced Vegetation Index (EVI) reducing the odds of psychotropic prescription dispensation by 2% and antidepressant prescription dispensation by 3%, and being linked to a 20% decrease in the odds of meeting criteria for major depressive disorder. Two forms of publicly accessible neighborhood nature were connected to substantial increases in neighborhood social cohesion. Each 1% increase in public greenspace or public natural space within a 500-meter buffer was associated with 3%, 4%, and 5% increases in reporting successively higher categories of sense of community    117  belonging (SoC) in comparison with the reference category of “very weak”. In turn, these higher levels of SoC were associated with a range of mental health improvements, with individuals in the highest SoC category of “very strong” having an 86% decrease in the odds of major depressive disorder, a 91% decrease in the odds of negative mental health, and a 2.8-point reduction on the K10 psychological distress scale. Perhaps the clearest benefits were seen with street trees, with just ten additional street trees within 100- to 500-meter buffers associated with a 1-4% reduction in the odds of psychotropic prescription dispensation, with the greatest reductions seen for antidepressant dispensation based on a 100-meter buffer. That said, many of the measures included in the NSI showed no association with mental health outcomes, or even indications of a negative effect.   Table 5.1 summarizes the findings from adjusted models for these measures and each of the others common to both Chapters 3 and 4, highlighting the connections and discrepancies identified through the epidemiological analyses as a whole and the extent to which the majority of exposures resulted in null associations. At the same time, this table demonstrates the particular strengths of each of the individual studies: by being sufficiently powered to examine the impact of street tree density despite the geographic gaps in coverage, Chapter 3 was able to describe a role for this specific form of nature; by integrating data from the CCHS-MH that included details on both outcomes and potential mediators, Chapter 4 (278) was able to uncover indirect associations that would have otherwise remained hidden.     118   Table 5.1 Summary of results from multiple analyses Summary of direct associations between natural space exposures and all mental health outcomes across epidemiological studies   Prescription Dispensation*  Major  Depressive Disorder   Negative  Mental  Health  Psychological Distress  (Kessler-10 scale)  OR (95% CI)+  OR (95% CI)+  OR (95% CI)+  b (95% CI) Visible nature     Greenspace 100ma 1.00 (1.00, 1.00) 1.03 (1.00, 1.06) (0.999, 1.055) 1.01 (1.00, 1.03) 0.02 (-0.02, 0.06) Natural Space 100ma 1.00 (1.00, 1.00) 1.03 (1.00, 1.06) 1.01 (1.00, 1.03) 0.02 (-0.03, 0.06)  Surrounding greenness      Residential Greenness 250mb 0.98 (0.98, 0.99) 0.80 (0.45, 1.42) 1.07 (0.80, 1.44) 0.05 (-0.23, 0.83)  Accessible neighborhood nature     Accessible Greenspace 500mc 1.00 (1.00, 1.00) 0.98 (0.95, 1.02) 1.01 (0.99, 1.03) -0.01 (-0.05, 0.02) Accessible Natural Space 500mc 1.00 (1.00, 1.00) 1.00 (0.97, 1.03) 1.01 (0.99, 1.02) -0.02 (-0.05, 0.02)  Binary access to nature      Binary Bluespace 1000md 1.01 (1.00, 1.02) 1.72 (0.85, 3.49) 1.20 (0.81, 1.76) 0.68 (-2.05, 3.40) Binary Greenspace 400md 1.10 (1.08, 1.12) 1.20 (0.47, 3.07) 1.34 (0.75, 2.42) 0.57 (-1.79, 2.93) * Each model includes a single explanatory variable and has been adjusted for neighborhood income decile, population density, and walkability + Each model includes a single explanatory variable and has been adjusted for sex, age, race-ethnicity, household income level, household education level, dwelling type, living arrangement, pain health status, urbanicity, population density, and walkability a Models based on a 1% increase in area b Models based on a 0.1-unit increase in EVI c Models based on a 1% increase in area d Models based on having access to nature  OR = adjusted odds ratio; CI = confidence interval; b = unstandardized slope coefficient All odds ratios (OR), 95% confidence intervals (95% CI), and b values are rounded     119  5.1.1 Advances in natural space exposure mapping and modeling The shift from experimental and clinical studies on the direct psychophysiological impacts of natural space exposure to population-level studies of the relationship with a broader range of mental health outcomes has produced a wide range of results. This is due in part to the heterogeneity of exposure methods and to the over-reliance on a single relatively simplistic metric of “greenness”, the Normalized Difference Vegetation Index (NDVI), which may capture unrelated elements of the urban form along with the natural environment. Expanding upon this approach to examine a range of natural space domains previously associated with mental health benefits in either laboratory or real-world settings, the study described in Chapter 2 formed the foundation for all subsequent epidemiological analyses, while comparing distinct measures with each other and with those most commonly employed within the literature. Previous attempts to achieve these aims have been more limited, including comparisons of NDVI with subjective appraisals (66,72) or land-cover classifications (142); or systematic reviews that provide a narrative rather than quantitative appraisal (187). The development of the Urban Green Space Indicator for Europe represents an approach to developing a multi-measure exposure model on an even-broader geographic scale, but it relies on land-cover data that are only available for the region, omits any consideration of the potential role of bluespace, and includes only quantitative components without any assessment of quality (145).   By contrast, the approach described in Chapter 2 is both broader and more detailed, modeling exposure based on the presence, form, accessibility, and quality of multiple forms of greenspace and bluespace. The range of data sources was similarly diverse: greenness presence was derived from remote sensing (NDVI/EVI), forms were extracted from municipal and private databases, accessibility was based on restrictions such as private ownership, and quality was assessed via the application of the previously validated POSDAT appraisal tool. This resulted in the creation of more than 50 exposure metrics at buffers ranging in size from    120  100 to 1,600 meters for all 60,242 postal codes across the study region in 2013, a scale that represents anywhere from a single building to a block-face in urban areas such as the study setting. Comparing NDVI with alternate approaches for assessing natural space resulted in widely divergent results, with quintile rankings shifting for 22–88% of postal codes, depending on the measure. To expand upon this key finding, principal component analysis (PCA) was applied to further clarify relationships among these distinct measures. The final PCA identified three main sets of variables, with the first two components explaining 68% of the total variance. The first component was dominated by the percentages of public and private greenspace and bluespace and public greenspace within 250 meters, while the second component was driven by lack of access to bluespace within one kilometer. Taken together, these results indicate that incorporating additional measures of natural space exposure can result in more-robust exposure assessments and that the specific measures included in the NSI explain a significant amount of spatial variation across a metropolitan area. However, the exact components identified as a result of the PCA and their associated loadings are location-specific, applying only to the data entered into the model. All in all, the NSI described in Chapter 2 represents a novel approach at a regional scale with the potential to inform urban planning and policymaking by clarifying the public health impacts of measures already included in local sustainability plans, such as access to public greenspace within 400 meters (28) or the goal to plant 150,000 trees across the city of Vancouver between 2010 and 2020 (279).  5.1.2 Expanding epidemiological assessments to objective outcomes In Chapters 3 and 4, we applied the novel NSI approach to natural space exposure mapping and modeling to a range of objective mental health outcomes, representing a complementary approach to numerous population-level studies that have examined subjective outcomes, including health-related quality of life (22,65,139,249,280), perceived overall mental health (118,137,281), stress levels (275,282), or feelings of anxiety or depression (21).    121  In Chapter 3, this was achieved by using the PharmaNet administrative database, which records all prescriptions dispensed in outpatient and community pharmacies across the province of British Columbia (177). Even among the small set of studies that have examined psychotropic medications as an outcome, this represents the first attempt, to our knowledge, to integrate data from a population-based administrative prescription database into a non-ecological study design. This approach represents a marked advancement because ecological study designs are inherently subject to the ecological fallacy, in which relationships that are seen at the aggregate level of analysis do not necessarily represent those that exist at the individual level. In addition, exposure measurement within ecologic studies is generally less precise than in studies of individuals because these variables are calculated at a significantly larger scale. Of the five studies we could find examining psychotropic prescriptions as an outcome, three were ecological: one conducted at the borough level in London, an area that generally contains 250,000 residents across an area 50 square kilometers in size (165); one that examined exposures and effects across England within Lower Level Super Output Areas, with a mean population size of 1,600 individuals and an average size of 0.9 to 18 square kilometers (25); and one that focused on municipalities in the Netherlands, which ranged in population size from under 1,000 to over 800,000 residents and in areas from less than seven to more than 500 square kilometers (205). Among the two relevant studies based on individual exposures, both Trigeuro-Mas et al. (137) and Gascon et al. (23) relied on self-reported medication use, which is subject to both recall bias and social desirability bias. In addition, we were able to limit analyses to incident prescriptions by using a nested case-control design among a cohort of adults who had not received any antidepressant or anxiolytic prescription in the five years preceding exposure assignment. This design represents another methodological advancement, greatly reducing the potential for movers’ bias to play a role in our findings.     122  In Chapter 4, objective outcome data came from the 2012 Canadian Community Health Survey-Mental Health (CCHS-MH) (3). Drawing on a nationally representative sample of Canadians over the age of 14, the survey was specifically designed to provide reliable and robust mental health data. As a result, the definition of past-year major depressive disorder was based on the World Health Organization’s Composite International Diagnostic Interview (WHO-CIDI), a standardized questionnaire developed specifically to identify mental disorders in the context of epidemiologic surveys. Negative mental health was similarly derived from a reliable and valid instrument, the Mental Health Continuum Short Form (MHC-SF). Finally, psychological distress came from the Kessler-10 Psychological Distress Scale (K10), which is also intended for use in population health surveys, and which has demonstrated high concordance with clinical psychiatric assessments. The Canadian Community Health Surveys serve as a rich resource of data for health scientists (202,210,259), but it is uncommon for investigators to integrate external data with that from the survey. In fact, we required special approval to do so for the analyses reported here.  5.1.3 Elucidating pathways via mediation analyses The larger novelty of Chapter 4 was the inclusion of a mediation analysis based on modern statistical approaches. Examining the pathways linking natural space exposure to both physical and mental health has become more common in recent years, but some of this work has been carried out via systematic reviews (14,18), while other efforts have explored pathways without including a formal mediation analysis (22). The majority of studies that have integrated original data and systematically tested mediation have done so by following Baron and Kenny’s classical four-step approach (154,204,270,275), which is no longer considered ideal due to its lack of power to detect mediated effects and high type II error rates (267). As a result, statisticians are now recommending omitting Baron and Kenny’s requirement of a significant total effect before proceeding to a formal mediation test when significant associations are found    123  between an explanatory variable and mediator as well as between that mediator and an outcome (268).     A small study in Plovdiv, Bulgaria explored single and parallel mediation models as an alternative to the classical four-step approach, and reported indirect associations between objective greenspace exposures and scores on a psychiatric screening tool via social cohesion, even in the absence of a significant direct effect (271). Although we employed a slightly different methodology – Iacobucci’s zMediation approach (269) – we also identified an indirect effect of natural space on mental health via sense of community belonging. Our findings provide additional evidence for the statistical understanding that indirect associations may be missed by studies that strictly follow the four-step approach (267,268). Ideally, results such as ours should encourage more-widespread adoption of modern approaches to mediation within the field of natural space and health. These new approaches are particularly critical in this field of research because urban natural spaces are inherently embedded in complex environments that contain an array of spatially covarying exposures. These exposures may have both harmful and beneficial health impacts, increasing the possibility for unmeasured factors to result in a lack of direct effects even in the presence of true mediation.  5.2 Implications for urban design and policy development 5.2.1 Methodological limitations The NSI dataset contained more than 50 distinct measures across a range of buffers from 100 to 1,600 meters in size, all of which were informed by prior research and with the goal of directly informing municipal and regional policy and planning. However, some individual components may still be too coarse or not properly adapted to the Vancouver CMA context. Both the private and municipal databases used to map greenspace were highly disaggregated, including details on the specific type of greenspace, such as a school park, arboretum, or regional nature    124  preserve. These details allowed us to carry out an access-categorization procedure to identify sites with restrictions such as permitting requirements or entrance fees, and to subsequently exclude these locations from our measure of publicly accessible greenspace. For bluespace, however, no such information was readily available, leading us to assume all bluespace was fully accessible. However, there are many factors that may limit public use and enjoyment of bluespace, including concurrent industrial uses, topographical barriers such as cliffs or marshes, or inaccessibility due to private ownership of adjacent lands.   There were also constraints on individual items in terms of their geographic coverage, namely with respect to the POSDAT park quality appraisals and the street tree density measures. Although the POSDAT appraisals were done remotely to eliminate the time spent traveling between locations, they remained relatively time-intensive. As a result, only 200 of more than 2,205 eligible parcels were assessed. Previous work had indicated that 1,600 meters is the maximum distance individuals are willing to regularly travel to a park (144), and assigning exposure based on the nearest appraised park within this buffer allowed us to assign a quality score to 82% of postal codes versus the 23% covered by parks within 400 meters. However, in light of the fact that our greenspace accessibility metric was based on this 400-meter distance as a means of representing a five-minute walk, it would have been ideal to examine the effects of park quality within this buffer as well. In addition, even at this larger buffer we were unable to include park quality in the models in Chapter 4 that examined mediation via social interaction due to small cell sizes, leaving a critical gap in our analyses.  Another limitation with respect to exposure assignment arises from the fact that the epidemiological analyses reported in Chapters 3 and 4 relied on residential six-digit postal codes to assign exposure, omitting exposures to natural space that may occur at, or in transit to, workplaces, community centers, or houses of worship, all of which may be particularly important    125  sites when examining aspects of social interaction. A 2018 commentary by Helbich notes that the majority of quotidian experiences actually take place outside of the home and calls for expanded use of technologies such as GPS-enabled smartphones to more accurately capture the spaces and places people inhabit during their daily lives (283). Integrating such technologies into studies on nature and mental health could be particularly useful because it would also support the capture of both imagery and qualitative appraisals of natural space, which could then be directly compared with objective measures such as those included in the NSI. However, such studies would likely be on a significantly smaller scale than those included in this dissertation, and it would be challenging to correlate these exposure data with large-scale administrative and survey data of the type incorporated here.  The NSI, which provides strength to this set of interrelated studies, is also a source of potential limitation with respect to the identification of significant associations. Emerging evidence indicates that there may be a dose-response effect of exposure to natural space on mental health, with both a ceiling and a floor (153). With this in mind, it is possible that the ceiling was exceeded in many postal codes within the Vancouver CMA, so we were unable to identify actual differences that exist between individuals with sufficient versus insufficient exposure. Because this area of research is still understudied, it is challenging to integrate particular thresholds in variable definitions. However, non-linear approaches to modeling such as the Bayesian geoadditive quantile approach used in a recent examination of antidepressant prescription rates could offer a valid statistical alternative (205). Another indication that the relationship between natural space exposure and mental health may be non-linear is demonstrated by our results for  public park quality, in which moderate-quality parks were associated with a 2% decrease in the odds of any psychotropic medication or anxiolytic dispensation alone, while high-quality parks were associated with 3% increases for any psychotropic and antidepressants alone.     126  Finally, Vancouver is a highly green, very urbanized region that may differ from many other municipalities to which the findings of this dissertation might be applied. Recognizing this complexity, after developing an integrated measure by applying principal component analysis (PCA) in Chapter 2, we proceeded to examine multiple individual components of the NSI in all subsequent epidemiological studies. This decision makes our results as a whole more directly applicable to other urban locations, and also increases our capacity to inform ongoing policy and planning decisions. To take one example, we selected the presence of public greenspace within 400 meters as a specific exposure metric because providing access based on this criterion is currently included in Vancouver’s Greenest City 2020 Action Plan (28). Little progress has been made toward achieving this objective since baseline assessments were carried out in 2010 (128) but, in light of our findings, prioritizing other components of the plan to increase access to nature may be recommended, at least from a public health perspective. For example, the plan also includes a goal of planting 150,000 new trees in the decade between 2010 and 2020; by the end of 2017, just over 102,000 of these trees had been planted, with 55% located within parks and along streets, and the remaining 45% added to private land (284). The results of Chapter 3 provide strong support for this goal, and may also represent a novel argument that the associated costs of the tree-planting program could be offset by savings within the healthcare system via lower psychotropic prescription dispensations (among other possible benefits). Our study was not designed to answer this question, but future health-economics research could fill in this gap and allow municipalities to request funding from the provincial healthcare system to support efforts to expand their urban forests. Our findings also lend support for a shift in the distribution of new tree plantings. As of the end of 2017, 45% of the newly planted trees were placed on private property, while only 16% were placed as street trees (284). Our results across all three buffers demonstrate the potential mental health benefits of focusing the program’s resources on street trees. In addition, the reductions seen for street trees within a 100-meter buffer should be used to encourage individuals who acquire trees for    127  residential properties to place them in areas where they are most visible to neighbors and passing pedestrians and occupants of cars.     5.2.2 Recommendations for future research and collaboration In the future, we hope to provide the POSDAT quality assessment tool to local residents via a website and brief workshops offered to local neighborhood-based environmental advocacy organizations. The tool is relatively easy to implement even without extensive training, and it could provide a useful approach for individuals to identify parks in need of improvement and then advocate for additional infrastructure or maintenance. Developing this program would also support a citizen-science approach to collecting park quality appraisal data for the remaining 2,000 eligible parcels, via the creation of an online tool to capture scores for inclusion in the existing database. This database would then be used to create a regularly updated, interactive version of the NSI, allowing residents of the region to identify natural space resources that they might not be aware of and to see how their postal codes compare with others.    More broadly, there are a number of methodological issues that remain to be explored within the literature, as well as a few areas where convincing evidence exists, but has yet to be consistently applied. With respect to the latter, one of the easiest improvements to implement would be the widespread adoption of the Enhanced Vegetation Index in lieu of the Normalized Difference Vegetation Index in settings where it has been shown to perform markedly better, including those featuring high-density vegetation (114,115) and those where customary cloud cover interferes with satellite-based imaging (57). Both measures come from the same set of satellites, with identical temporal and spatial coverage, and both are made freely available by the US National Aeronautics and Space Administration (69), making such a transition relatively seamless.       128  In terms of areas that could benefit from additional methodological refinement, perhaps the most-pressing need is the development of detailed bluespace measures that integrate factors such as access and quality and properly contextualize the broader environment. A first step would be examining discrete forms of bluespace (such as rivers versus coasts) to identify their independent effects on mental health. A subsequent improvement would be integrating data from land-cover or cadastral datasets to identify surrounding or concomitant land-uses that may partially offset any positive impacts. Historically, many forms of bluespace have been used for high-volume transportation, as an outflow for effluent from water treatment plants, or as a source of hydroelectric power, severely degrading their ecological quality (285). In the past few decades, numerous efforts have been made to reclaim bluespaces as sites for physical activity, social interaction, and even entrepreneurial innovation (286) – London’s South Bank project on the Thames (287) and the Cheonggyecheon Stream linear park in Seoul serving as two prominent examples (288) – but industrial uses remain common. Finally, few efforts (including the current dissertation) have attempted to evaluate bluespace quality, in part because the majority of existing tools and frameworks are based on greenspace characteristics (180,289) or rely on Google Street View imagery that doesn’t readily extend to areas with water (290–292). The recent development of the Natural Environment Scoring Tool by Gidlow et al. represents one promising appraisal method with application to multiple forms of natural space, but it has yet to be properly validated or broadly applied, so the extent of its utility and feasibility remains unclear (112). Disentangling the various forms and uses of bluespace would also support efforts to define and apply exposure metrics based on hypothesized pathways rather than mere data availability. This dissertation takes some important steps along that path via the prespecified hypotheses regarding the relevance of both form and scale to the social interaction pathway explored in Chapter 4, but additional attention to this area is essential to move the field forward.      129  Finally, this dissertation benefitted immensely from the willingness of local urban planners and natural space resource managers to make information publicly accessible or to provide it upon request. However, this was a one-way data-retrieval process rather than a consistent collaboration, which highlights a larger failure for environmental epidemiologists to partner with the individuals and organizations responsible for shaping environments. There are indications that such collaborations are slowly developing, however. For example, the City of Vancouver’s original Greenest City 2020 Action Plan simply notes that “green spaces have been shown to benefit our physical and emotional health by reducing blood pressure, cholesterol, and stress” (28) and the updated version of the plan, issued in 2015, includes identical language (128). However, the city’s 2018 Urban Forest Strategy incorporates multiple health benefits, including increased cognitive functioning and attention improvements among children; greater odds of achieving recommended levels of physical activity; facilitated social ties within neighborhoods; and air purification and noise reduction, neatly summarizing all four of the pathways described in Chapter 1 of this thesis (284). These benefits are ascribed to the urban forest as a whole, however, lacking the additional insights into presence, form, accessibility, and quality offered by a multicomponent model such as the NSI.      In a rapidly urbanizing world, epidemiologists are uniquely positioned to illuminate the complex, multilayered relationships between the built, natural, and social environments of cities and the health of their residents. Sociologists are much better placed to elicit the specific needs and preferences of adolescents, women, older adults, and residents of low-income neighborhoods, all of whom are particularly vulnerable to mental health problems of different types. Demographers can offer critical insights into the extent to which these individual groups contribute to the overall make-up of the urban population; geographers can identify the neighborhoods in which they live and describe factors such as gentrification that may lead to their displacement and relocation. Together, members of all of these academic disciplines can,    130  and must, engage with the urban planners, municipal and provincial elected officials, and federal policymakers whose decisions can create cities that take full advantage of nature’s potential to foster social connections and improve mental health.     131  Bibliography 1.  Whiteford HA, Degenhardt L, Rehm J, Baxter AJ, Ferrari AJ, Erskine HE, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet. 2013 Nov 9;382(9904):1575–86.  2.  Pearson C, Janz T, Ali J. Mental and substance use disorders in Canada [Internet]. 2015 [cited 2016 Dec 28]. Available from: http://www.statcan.gc.ca/pub/82-624-x/2013001/article/11855-eng.htm#n3 3.  Statistics Canada. Canadian Community Health Survey – Mental Health (CCHS): Detailed Information for 2012 [Internet]. 2013 [cited 2017 Jun 23]. Available from: http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5015 4.  Mental Health Commission of Canada. Making the Case for Investing in Mental Health. Ottawa; 2016.  5.  Vos T, Abajobir AA, Abbafati C, Abbas KM, Abate KH, Abd-Allah F, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2017;390(10100):1211–59.  6.  Rizvi SJ, Cyriac A, Grima E, Tan M, Lin P, Gallaugher LA, et al. Depression and employment status in primary and tertiary care settings. Can J Psychiatry. 2015;60(1):14–22.  7.  Mental Health Commission of Canada. Changing Directions, Changing Lives: The Mental Health Strategy for Canada. Ottawa; 2012.  8.  Tam T. The Chief Public Health Officer’s Report on the State of Public Health in Canada: 2017. Ottawa; 2017.  9.  Tennant C. Life events, stress and depression: a review of recent findings. Aust N Z J Psychiatry. 2002 Apr;36(2):173–82.  10.  Baune B. Conceptual challenges of a tentative model of stress-induced depression. PLoS    132  One. 2009 Jan 30;4(1):e4266–e4266.  11.  Magalhaes AC, Holmes KD, Dale LB, Comps-Agrar L, Lee D, Yadav PN, et al. CRF receptor 1 regulates anxiety behavior via sensitization of 5-HT2 receptor signaling. Nat Neurosci. 2010 May;13(5):622-U140.  12.  Gascon M, Triguero-Mas M, Martínez D, Dadvand P, Forns J, Plasència A, et al. Mental health benefits of long-term exposure to residential green and blue spaces: a systematic review. Int J Environ Res Public Health. 2015;12(4):4354–79.  13.  Taylor L, Hochuli DF. Defining greenspace: Multiple uses across multiple disciplines. Landsc Urban Plan. 2017;158:25–38.  14.  James P, Banay RF, Hart JE, Laden F. A review of the health benefits of greenness. Curr Epidemiol Reports. 2015;2(2):131–42.  15.  Lee ACK, Maheswaran R. The health benefits of urban green spaces: a review of the evidence. J Public Health (Bangkok). 2011;33(2):212–22.  16.  Sugiyama T, Thompson CW. Outdoor environments, activity and the well-being of older people: conceptualising environmental support. Environ Plan A. 2007 Aug;39(8):1943–60.  17.  Fong KC, Hart JE, James P. A review of epidemiologic studies on greenness and health: Updated literature through 2017. Curr Environ Heal Reports. 2018;5(1):77–87.  18.  Markevych I, Schoierer J, Hartig T, Chudnovsky A, Hystad P, Dzhambov AM, et al. Exploring pathways linking greenspace to health: Theoretical and methodological guidance. Environ Res. 2017;158(Oct):301–17.  19.  Gascon M, Zijlema W, Vert C, White MP, Nieuwenhuijsen MJ. Outdoor blue spaces, human health and well-being: A systematic review of quantitative studies. Int J Hyg Environ Health. Elsevier; 2017;220(8):1207–21.  20.  White MP, Alcock I, Wheeler BW, Depledge MH. Would you be happier living in a greener urban area? A fixed-effects analysis of panel data. Psychol Sci. United States;    133  2013 Jun;24(6):920–8.  21.  Ellaway A, Morris G, Curtice J, Robertson C, Allardice G, Robertson R. Associations between health and different types of environmental incivility: a Scotland-wide study. Public Health. Elsevier Ltd; 2009 Nov;123(11):708–13.  22.  Sugiyama T, Leslie E, Giles-Corti B, Owen N. Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships? J Epidemiol Community Health. 2008;62(5):e9–e9.  23.  Gascon M, Sánchez-Benavides G, Dadvand P, Martínez D, Gramunt N, Gotsens X, et al. Long-term exposure to residential green and blue spaces and anxiety and depression in adults: A cross-sectional study. Environ Res. 2018;162:231–9.  24.  Houlden V, Weich S, Jarvis S. A cross-sectional analysis of green space prevalence and mental wellbeing in England. BMC Public Health. BMC Public Health; 2017;17:1–9.  25.  Gidlow CJ, Smith G, Wilson R, Trinder P, Martinez D, Grazuleviciene R, et al. Natural environments and prescribing in England. J Epidemiol Community Heal. 2016;151:103–8.  26.  Sugiyama, Villanueva K, Knuiman M, Francis J, Foster S, Wood L, et al. Can neighborhood greenspace mitigate health inequalities? A study of socio-economic status and mental health. Health Place. 2016;38:16–21.  27.  Generaal E, Timmermans EJ, Dekkers JEC, Smit JH, Penninx BWJH. Not urbanization level but socioeconomic, physical and social neighbourhood characteristics are associated with presence and severity of depressive and anxiety disorders. Psychol Med. 2018;March:1–13.  28.  City of Vancouver. Greenest City 2020 Action Plan [Internet]. Vancouver; 2012. Available from: https://vancouver.ca/green-vancouver/greenest-city-2020-action-plan.aspx 29.  City of Campbell River. Sustainable Official Community Plan [Internet]. 2012 [cited 2014 Mar 23]. Available from: http://www.campbellriver.ca/docs/default-source/Document-Library/bylaws/sustainable-official-community-plan-%28schedule-a-to-bylaw-3475-   134  2012%29.pdf?sfvrsn=4 30.  City of New York. One New York: The Plan for a Strong and Just City [Internet]. New York City; 2015. Available from: http://www.nyc.gov/html/onenyc/downloads/pdf/publications/OneNYC.pdf 31.  Caro JC, Armour T, Fernandez M. Madrid + Natural: Nature-based Climate Change Adaptation [Internet]. Madrid; 2016. Available from: http://publications.arup.com/publications/m/madrid_natural 32.  New York City Environmental Protection. NYC Green Infrastructure Plan [Internet]. New York City; 2013 [cited 2014 Oct 21]. Available from: http://www.nyc.gov/html/dep/html/stormwater/nyc_green_infrastructure_plan.shtml 33.  Pearce JR, Maddison R. Do enhancements to the urban built environment improve physical activity levels among socially disadvantaged populations? Int J Equity Health. 2011 Jul;10:28.  34.  Richardson E, Pearce J, Mitchell R, Day P, Kingham S. The association between green space and cause-specific mortality in urban New Zealand: an ecological analysis of green space utility. BMC Public Health. 2010;10:240.  35.  Masuda JR, Teelucksingh C, Zupancic T, Crabtree A, Haber R, Skinner E, et al. Out of our inner city backyards: Re-scaling urban environmental health inequity assessment. Soc Sci Med. 2012 Oct;75(7):1244–53.  36.  Bullard RD. Environmental justice: It’s more than waste facility siting. Soc Sci Q. 1996;77(3):493–9.  37.  Jennings V, Larson L, Yun J. Advancing sustainability through urban green space: Cultural ecosystem services, equity, and social determinants of health. Int J Environ Res Public Health. 2016;13(2).  38.  Wolch JR, Byrne J, Newell JP. Urban green space, public health, and environmental justice: The challenge of making cities “just green enough.” Landsc Urban Plan.    135  2014;125:234–44.  39.  Annerstedt M, Jönsson P, Wallergård M, Johansson G, Karlson B, Grahn P, et al. Inducing physiological stress recovery with sounds of nature in a virtual reality forest - results from a pilot study. Physiol Behav. 2013;118:240–50.  40.  Barton J, Griffin M, Pretty J. Exercise-, nature- and socially interactive-based initiatives improve mood and self-esteem in the clinical population. Perspect Public Health. 2012 Mar;132(2):89–96.  41.  Thompson Coon J, Boddy K, Stein K, Whear R, Barton J, Depledge MH. Does participating in physical activity in outdoor natural environments have a greater effect on physical and mental wellbeing than physical activity indoors? A systematic review. Environ Sci Technol. 2011;45(5):1761–72.  42.  Berger R, Tiry M. The enchanting forest and the healing sand–nature therapy with people coping with psychiatric difficulties. Arts Psychother. 2012;39(5):412–6.  43.  Berman MG, Kross E, Krpan KM, Askren MK, Burson A, Deldin PJ, et al. Interacting with nature improves cognition and affect for individuals with depression. J Affect Disord. 2012 Nov;140(3):300–5.  44.  Berman MG, Jonides J, Kaplan S. The cognitive benefits of interacting with nature. Psychol Sci. 2008 Dec;19(12):1207–12.  45.  O’Brien L, Burls A, Townsend M, Ebden M. Volunteering in nature as a way of enabling people to reintegrate into society. Perspect Public Health. 2011;131(2):71–81.  46.  Stark MA. Restoring attention in pregnancy: the natural environment. Clin Nurs Res. 2003;12(3):246–65.  47.  Valtchanov D, Barton KR, Ellard C. Restorative effects of virtual nature settings. Cyberpsychol Behav Soc Netw. 2010;13(5):503–12.  48.  Mackay GJ, Neill JT. The effect of “green exercise” on state anxiety and the role of exercise duration, intensity, and greenness: a quasi-experimental study. Psychol Sport    136  Exerc. 2010 May;11(3):238–45.  49.  Wells NM, Evans GW. Nearby nature – A buffer of life stress among rural children. Environ Behav. 2003;35(3):311–30.  50.  Faber Taylor A, Kuo FE. Children with attention deficits concentrate better after walk in the park. J Atten Disord. 2009;12(5):402–9.  51.  Francis J, Wood LJ, Knuiman M, Giles-Corti B. Quality or quantity? Exploring the relationship between Public Open Space attributes and mental health in Perth, Western Australia. Soc Sci Med. 2012 May;74(10):1570–7.  52.  Huynh Q, Craig W, Janssen I, Pickett W. Exposure to public natural space as a protective factor for emotional well-being among young people in Canada. BMC Public Health. 2013;13:407.  53.  Adevi AA, Mårtensson F. Stress rehabilitation through garden therapy: the garden as a place in the recovery from stress. Urban For Urban Green. 2013;12(2):230–7.  54.  McCaffrey R, Hanson C, McCaffrey W. Garden walking for depression: a research report. Holist Nurs Pract. 2010;24(5):252–9.  55.  Rappe E, Kivelä SL. Effects of garden visits on long-term care residents as related to depression. Horttechnology. 2005;15(2):298–303.  56.  Whitehouse S, Varni JW, Seid M, Cooper-Marcus C, Ensberg MJ, Jacobs JR, et al. Evaluating a children’s hospital garden environment: Utilization and consumer satisfaction. J Environ Psychol. 2001 Sep;21(3):301–14.  57.  Sung J, Woo JM, Kim W, Lim SK, Chung EJ. The effect of cognitive behavior therapy-based “forest therapy” program on blood pressure, salivary cortisol level, and quality of life in elderly hypertensive patients. Clin Exp Hypertens (New York). 2012;34(1):1–7.  58.  Nakau M, Imanishi JJ, Watanabe S, Imanishi A, Baba T, Hirai K, et al. Spiritual care of cancer patients by integrated medicine in urban green space: a pilot study. Explor J Sci Heal. 2013;9(2):87–90.     137  59.  Kim W, Lim SK, Chung EJ, Woo JM. The effect of cognitive behavior therapy-based psychotherapy applied in a forest environment on physiological changes and remission of major depressive disorder. Psychiatry Investig. 2009;6(4):245–54.  60.  Lee J, Park B-J, Tsunetsugu Y, Ohira T, Kagawa T, Miyazaki Y. Effect of forest bathing on physiological and psychological responses in young Japanese male subjects. Public Health. 2011;125(2):93–100.  61.  Mao GX, Lan XG, Cao YB, Chen ZM, He ZH, Lv YD, et al. Effects of short-term forest bathing on human health in a broad-leaved evergreen forest in Zhejiang Province, China. Biomed Environ Sci. 2012;25(3):317–24.  62.  Martens D, Gutscher H, Bauer N. Walking in “wild” and “tended” urban forests: the impact on psychological well-being. J Environ Psychol. 2011;31(1):36–44.  63.  Morita E, Fukuda S, Nagano J, Hamajima N, Yamamoto H, Iwai Y, et al. Psychological effects of forest environments on healthy adults: shinrin-yoku (forest-air bathing, walking) as a possible method of stress reduction. Public Heal. 2007;121(1):54–63.  64.  Shin WS, Shin CS, Yeoun PS, Sop W, Chang S, Shin S, et al. The influence of forest therapy camp on depression in alcoholics. Environ Heal Prev Med. 2012;17(1):73–6.  65.  van Dillen SM, de Vries S, Groenewegen PP, Spreeuwenberg P. Greenspace in urban neighbourhoods and residents’ health: adding quality to quantity. J Epidemiol Community Heal. 2012 Jun;66(6):e8.  66.  Rhew IC, Vander Stoep A, Kearney A, Smith NL, Dunbar MD. Validation of the Normalized Difference Vegetation Index as a measure of neighborhood greenness. Ann Epidemiol. 2011 Dec;21(12):946–52.  67.  Hystad P, Davies HW, Frank L, Van Loon J, Gehring U, Tamburic L, et al. Residential greenness and birth outcomes: evaluating the influence of spatially correlated built-environment factors. Environ Health Perspect. 2014;122(10):1095–102.  68.  Dadvand P, de Nazelle A, Triguero-Mas M, Schembari A, Cirach M, Amoly E, et al.    138  Surrounding greenness and exposure to air pollution during pregnancy: an analysis of personal monitoring data. Environ Health Perspect. 2012;120(9):1286–90.  69.  Weier J, Herring D. Measuring Vegetation (NDVI & EVI) [Internet]. 2000 [cited 2014 Nov 24]. Available from: http://earthobservatory.nasa.gov/Features/MeasuringVegetation/measuring_vegetation_1.php 70.  United States National Aeronautics and Space Administration. MODIS Vegetation Index Products (NDVI and EVI) [Internet]. 2015 [cited 2015 Jun 23]. Available from: http://modis.gsfc.nasa.gov/data/dataprod/mod13.php 71.  Chen P-Y, Fedosejevs G, Tiscareno-Lopez M, Arnold JG. Assessment of MODIS-EVI MODIS-NDVI and Vegetation-NDVI composite data using agricultural measurements: an example at corn fields in western Mexico. Environ Monit Assess. 2006;119:69–82.  72.  Leslie E, Sugiyama T, Ierodiaconou D, Kremer P. Perceived and objectively measured greenness of neighbourhoods: are they measuring the same thing? Landsc Urban Plan. 2010 Mar 30;95(1–2):28–33.  73.  Fan Y, Das K V, Chen Q. Neighborhood green, social support, physical activity, and stress: assessing the cumulative impact. Heal Place. 2011;17(6):1202–11.  74.  Rugel EJ. Green Space and Mental Health: Pathways, Impacts, and Gaps [Internet]. Vancouver; 2015. Available from: http://www.ncceh.ca/documents/evidence-review/green-space-and-mental-health-pathways-impacts-and-gaps 75.  Hartig T, Mitchell R, de Vries S, Frumkin H. Nature and health. Annu Rev Public Health. 2014;35:207–28.  76.  Tzoulas K, Korpela K, Venn S, Yli-Pelkonen V, Kaźmierczak A, Niemela J, et al. Promoting ecosystem and human health in urban areas using Green Infrastructure: A literature review. Landsc Urban Plan. 2007;81(3):167–78.  77.  Corburn J. City planning as preventive medicine. Prev Med (Baltim). 2015;77:48–51.     139  78.  Stigsdotter UK, Corazon SS, Sidenius U, Refshauge AD, Grahn P. Forest design for mental health promotion—Using perceived sensory dimensions to elicit restorative responses. Landsc Urban Plan. 2017;160:1–15.  79.  Ulrich RS. Natural versus urban scenes: Some psychophysiological effects. Environ Behav. 1981;13(5):523–56.  80.  Kaplan R. Some psychological benefits of gardening. Environ Behav. 1973;5(2):145–62.  81.  National Institute of Mental Health. Schizophrenia [Internet]. 2014 [cited 2014 Feb 2]. Available from: http://www.nimh.nih.gov/health/topics/schizophrenia/index.shtml 82.  National Institute of Mental Health. Attention Deficit Hyperactivity Disorder [Internet]. 2012 [cited 2014 Feb 2]. Available from: http://www.nimh.nih.gov/health/publications/attention-deficit-hyperactivity-disorder/index.shtml 83.  Ohly H, White MP, Wheeler BW, Bethel A, Ukoumunne OC, Nikolaou V, et al. Attention restoration theory: A systematic review of the attention restoration potential of exposure to natural environments. J Toxicol Environ Heal - Part B Crit Rev. 2016;19(7):305–43.  84.  Barton J, Hine R, Pretty J. The health benefits of walking in greenspaces of high natural and heritage value. J Integr Environ Sci. 2009;6(4):261–78.  85.  Roe J, Aspinall P. The restorative benefits of walking in urban and rural settings in adults with good and poor mental health. Health Place. 2011 Jan;17(1):103–13.  86.  Ward Thompson C, Roe J, Aspinall P, Mitchell R, Clow A, Miller D. More green space is linked to less stress in deprived communities: evidence from salivary cortisol patterns. Landsc Urban Plan. 2012;105(3):221–9.  87.  Tsunetsugu Y, Lee J, Park BJ, Tyrvainen L, Kagawa T, Miyazaki Y. Physiological and psychological effects of viewing urban forest landscapes assessed by multiple measurements. Landsc Urban Plan. 2013;113:90–3.  88.  Wells NM. At home with nature – effects of “greenness” on children’s cognitive functioning. Environ Behav. 2000 Nov;32(6):775–95.     140  89.  Kuo FE, Taylor AF. A potential natural treatment for attention-deficit/hyperactivity disorder: evidence from a national study. Am J Public Health. 2004 Sep;94(9):1580–6.  90.  Bratman GN, Hamilton JP, Hahn KS, Daily GC, Gross JJ. Nature experience reduces rumination and subgenual prefrontal cortex activation. Proc Natl Acad Sci U S A. 2015;112(28):8567–672.  91.  Dadvand P, Pujol J, Macià D, Martínez-vilavella G, Blanco-hinojo L, Mortamais M, et al. The association between lifelong greenspace exposure and 3-dimensional brain magnetic resonance imaging in Barcelona schoolchildren. Environ Health Perspect. 2018;126(2):027012-1-027012-8.  92.  Barton J, Pretty J. What is the best dose of nature and green exercise for improving mental health? A multi-study analysis. Environ Sci Technol. 2010 May 15;44(10):3947–55.  93.  Lachowycz K, Jones AP. Greenspace and obesity: a systematic review of the evidence. Obes Rev. 2011 May;12(501):e183–9.  94.  Giles-Corti B, Broomhall MH, Knuiman M, Collins C, Douglas K, Ng K, et al. Increasing walking – how important is distance to, attractiveness, and size of public open space? Am J Prev Med. 2005 Feb;28(2):169–76.  95.  Kreuzfeld S, Preuss M, Weippert M, Stoll R. Health effects and acceptance of a physical activity program for older long-term unemployed workers. Int Arch Occup Environ Health. 2013 Jan;86(1):99–105.  96.  Pedersen BK, Saltin B. Evidence for prescribing exercise as therapy in chronic disease. Scand J Med Sci Sport. 2006 Feb;16:3–63.  97.  Lawlor DA, Hopker SW. The effectiveness of exercise as an intervention in the management of depression: systematic review and meta-regression analysis of randomised controlled trials. Br Med J. 2001 Mar 31;322(7289):763–7.  98.  Hillsdon M, Panter J, Foster C, Jones A. The relationship between access and quality of    141  urban green space with population physical activity. Public Health. 2006 Dec;120(12):1127–32.  99.  Kaczynski AT, Potwarka LR, Saelens BE. Association of park size, distance, and features with physical activity in neighborhood parks. Am J Public Health. 2008 Aug;98(8):1451–6.  100.  Maas J, van Dillen SME, Verheij RA, Groenewegen PP. Social contacts as a possible mechanism behind the relation between green space and health. Health Place. 2009 Jun;15(2):586–95.  101.  Kim J, Kaplan R. Physical and psychological factors in sense of community: New Urbanist Kentlands and nearby orchard village. Environ Behav. 2004 May;36(3):313–40.  102.  Thoits PA. Mechanisms linking social ties and support to physical and mental health. J Health Soc Behav. 2011 Jun;52(2):145–61.  103.  Kawachi I, Berkman LF. Social ties and mental health. J Urban Heal New York Acad Med. 2001;78(3):458–67.  104.  Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983 Dec;24(4):385–96.  105.  O’Brien L, Burls A, Townsend M, Ebden M. Volunteering in nature as a way of enabling people to reintegrate into society. Perspect Public Health. 2011;131(2):71–81.  106.  Finlay J, Franke T, McKay H, Sims-Gould J. Therapeutic landscapes and wellbeing in later life: Impacts of blue and green spaces for older adults. Health Place. 2015;34:97–106.  107.  Hsieh M, Lee W. Effect of spatial elements in an open space on a subject’s psychological perception. J Asian Archit Build Eng. 2010;9(1):111–6.  108.  Gidlof-Gunnarsson A, Ohrstrom E. Noise and well-being in urban residential environments: The potential role of perceived availability to nearby green areas. Landsc Urban Plan. 2007 Nov 19;83(2–3):115–26.  109.  Alcock I, White M, Cherrie M, Wheeler B, Taylor J, McInnes R, et al. Land cover and air    142  pollution are associated with asthma hospitalisations: A cross-sectional study. Environ Int. 2017;109(March):29–41.  110.  Sbihi H, Tamburic L, Koehoorn M, Brauer M. Greenness and incident childhood asthma: a 10-year follow-up in a population-based birth cohort. Am J Respir Crit Care Med. 2015;192(9):1131–3.  111.  Bixby H, Hodgson S, Fortunato L, Hansell A, Fecht D. Associations between green space and health in English cities: an ecological, cross-sectional study. PLoS One. 2015;10(3):e0119495.  112.  Gidlow C, van Kempen E, Smith G, Triguero-Mas M, Kruize H, Gražulevičienė R, et al. Development of the natural environment scoring tool (NEST). Urban For Urban Green. 2018;29:322–33.  113.  Nieuwenhuijsen MJ, Khreis H, Triguero-Mas M, Gascon M, Dadvand P. Fifty shades of green. Epidemiology. 2017;28(1):63–71.  114.  de Vries S, Verheij RA, Groenewegen PP, Spreeuwenberg P. Natural environments - healthy environments? An exploratory analysis of the relationship between greenspace and health. Environ Plan A. 2003 Oct;35(10):1717–31.  115.  Maas J, Verheij RA, De Vries S, Spreeuwenberg P, Schellevis FG, Groenewegen PP. Morbidity is related to a green living environment. J Epidemiol Community Heal. 2009 Dec;63(12):967–73.  116.  Kohlleppel T, Bradley JC, Jacob S. A walk through the garden: can a visit to a botanic garden reduce stress? Horttechnology. 2002;12(3):489–92.  117.  Wilson E. Biophilia. Cambridge (MA): Harvard University Press; 1984.  118.  van den Berg AE, Maas J, Verheij RA, Groenewegen PP. Green space as a buffer between stressful life events and health. Soc Sci Med. 2010 Apr;70(8):1203–10.  119.  Statistics Canada. Focus on Geography Series, 2011 Census [Internet]. 2016 [cited 2016 Jun 11]. Available from: https://www150.statcan.gc.ca/n1/pub/92-151-g/92-151-   143  g2011001-eng.pdf 120.  Metro Vancouver. Member Municipalities [Internet]. 2016 [cited 2015 Oct 11]. Available from: http://www.metrovancouver.org/ 121.  Statistics Canada. Population and Dwelling Count Highlight Tables, 2016 Census [Internet]. 2017 [cited 2017 Mar 7]. Available from: https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/hlt-fst/pd-pl/Table.cfm?Lang=Eng&T=303&CMA=933&S=86&O=A&RPP=25 122.  BC Stats. Population Projections [Internet]. 2015 [cited 2015 Jun 11]. Available from: http://www.bcstats.gov.bc.ca/StatisticsBySubject/Demography/PopulationProjections.aspx 123.  Statistics Canada. Immigration and Ethnocultural Diversity: Key Results from the 2016 Census [Internet]. 2017 [cited 2018 Jun 23]. Available from: https://www150.statcan.gc.ca/n1/daily-quotidien/171025/dq171025b-eng.htm 124.  City of Vancouver. Our City: Geography [Internet]. 2018 [cited 2018 Jun 23]. Available from: http://vancouver.ca/news-calendar/geo.aspx 125.  Port of Vancouver. Port of Vancouver: About Us [Internet]. 2018 [cited 2018 Jun 23]. Available from: https://www.portvancouver.com/about-us/ 126.  City of Vancouver. City of Vancouver Urban Forest Strategy. Vancouver; 2014.  127.  Greater Vancouver Regional District Board. Metro Vancouver 2040: Shaping Our Future [Internet]. 2015. Available from: http://www.metrovancouver.org/services/regional-planning/PlanningPublications/RGSAdoptedbyGVRDBoard.pdf 128.  City of Vancouver. Greenest City 2020 Action Plan – Part Two: 2015-2020. Vancouver; 2015.  129.  Ulrich RS. Human responses to vegetation and landscapes. Landsc Urban Plan. 1986;13:29–44.  130.  Kaplan S. The restorative benefits of nature: toward an integrative framework. J Environ    144  Psychol. 1995;15(3):169–82.  131.  Ward Thompson C. Linking landscape and health: The recurring theme. Landsc Urban Plan. 2011;99(3–4):187–95.  132.  World Health Organization. Urban Population Growth [Internet]. 2016 [cited 2016 Jan 1]. Available from: http://www.who.int/gho/urban_health/situation_trends/urban_population_growth_text/en/ 133.  Takano T, Nakamura K, Watanabe M. Urban residential environments and senior citizens’ longevity in megacity areas: the importance of walkable green spaces. J Epidemiol Community Health. 2002 Dec;56(12):913–8.  134.  Dadvand P, de Nazelle A, Figueras F, Basagana X, Su J, Amoly E, et al. Green space, health inequality and pregnancy. Environ Int. 2012 Apr;40:110–5.  135.  Maas J, Verheij RA, Groenewegen PP, de Vries S, Spreeuwenberg P. Green space, urbanity, and health: how strong is the relation? J Epidemiol Community Heal. 2006 Jul;60(7):587–92.  136.  Alcock I, White MP, Wheeler BW, Fleming LE, Depledge MH. Longitudinal effects on mental health of moving to greener and less green urban areas. Environ Sci Technol. 2014;48(2):1247–55.  137.  Triguero-Mas M, Dadvand P, Cirach M, Martinez D, Medina A, Mompart A, et al. Natural outdoor environments and mental and physical health: relationships and mechanisms. Environ Int. 2015;77C:35–41.  138.  Kearns A, Whitley E, Bond L, Tannahill C. The residential psychosocial environment and mental wellbeing in deprived areas. Int J Hous Policy. 2012;12(4):413–38.  139.  van den Berg M, van Poppel M, van Kamp I, Andrusaityte S, Balseviciene B, Cirach M, et al. Visiting greenspace is associated with mental health and vitality: a cross-sectional study in four European cities. Health Place. 2016;38:8–15.  140.  Nutsford D, Pearson AL, Kingham S, Reitsma F. Residential exposure to visible blue    145  space (but not green space) associated with lower psychological distress in a capital city. Health Place. 2016;39:70–8.  141.  Kardan O, Gozdyra P, Misic B, Moola F, Palmer LJ, Paus T, et al. Neighborhood greenspace and health in a large urban center. Sci Rep. 2015;5:11610.  142.  Gascon M, Cirach M, Martínez D, Dadvand P, Valentín A, Plasència A, et al. Normalized Difference Vegetation Index (NDVI) as a marker of surrounding greenness in epidemiological studies: the case of Barcelona city. Urban For Urban Green. 2016;19:88–94.  143.  Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985;14(1):32–8.  144.  Kaczynski AT, Schipperijn J, Hipp JA, Besenyi GM, Wilhelm Stanis SA, Hughey SM, et al. ParkIndex: development of a standardized metric of park access for research and planning. Prev Med (Baltim). 2016;87:110–4.  145.  Annerstedt van den Bosch M, Mudu P, Uscila V, Barrdahl M, Kulinkina A, Staatsen B, et al. Development of an urban green space indicator and the public health rationale. Scand J Public Health. 2016;(44):159–67.  146.  Census of Canada. Postal Code Conversion File, June 2013 Postal Codes, 2011 [Internet]. 2013 [cited 2015 Oct 1]. Available from: http://hdl.handle.net/11272/10029 147.  Fuller D, Shareck M. Canada Post community mailboxes: implications for health research. Can J Public Heal. 2014;105(6):e453–5.  148.  Healy MA, Gilliland JA. Quantifying the magnitude of environmental exposure misclassification when using imprecise address proxies in public health research. Spat Spatiotemporal Epidemiol. 2012;3(1):55–67.  149.  United States National Aeronautics and Space Administration. Landsat Science - Landsat 8 [Internet]. 2016 [cited 2016 Jul 4]. Available from: http://landsat.gsfc.nasa.gov/?p=5081 150.  United States Geological Survey. EarthExplorer [Internet]. 2016 [cited 2015 Aug 13]. Available from: http://earthexplorer.usgs.gov    146  151.  US NASA Land Processes Distributed Active Archive Center. MODIS Reprojection Tool. 2014.  152.  Metro Vancouver. Lawn Sprinkling Regulations [Internet]. 2016 [cited 2016 Aug 13]. Available from: http://www.metrovancouver.org/services/water/conservation-reservoir-levels/lawn-sprinkling/pages/default.aspx 153.  Cox DTC, Shanahan DF, Hudson HL, Fuller RA, Anderson K, Hancock S, et al. Doses of nearby nature simultaneously associated with multiple health benefits. Int J Environ Res Public Health. 2017;14(2):172.  154.  Zijlema WL, Triguero-mas M, Smith G, Cirach M, Martinez D, Dadvand P, et al. The relationship between natural outdoor environments and cognitive functioning and its mediators. Environ Res. 2017;155(February):1–8.  155.  DMTI Spatial Inc. CanMap Parks, v2014.3 [Internet]. 2014 [cited 2015 Oct 21]. Available from: http://hdl.handle.net.ezproxy.library.ubc.ca/11272/10183 156.  ESRI. North America Parks [Internet]. 2014 [cited 2015 Oct 21]. Available from: http://www.arcgis.com/home/item.html?id=d618daa4a2ab4b13846da9837b28dc60 157.  City of White Rock. Parks Dataset [Internet]. 2014 [cited 2015 Oct 28]. Available from: http://data.whiterockcity.ca/dataset/parks 158.  District of North Vancouver. Open Data - Parks [Internet]. 2006 [cited 2015 Oct 28]. Available from: http://geoweb.dnv.org/data/metadata.php?dataset=PrkPark 159.  City of Surrey. Datasets - Parks [Internet]. 2014 [cited 2015 Oct 28]. Available from: http://data.surrey.ca/dataset/parks 160.  Township of Langley BC. Parks - CAD [Internet]. 2014 [cited 2015 Oct 28]. Available from: https://data.tol.ca/Recreation/Parks-CAD/dq3q-si3d 161.  City of Vancouver. Open Data Catalogue - Parks Data [Internet]. 2015 [cited 2015 Oct 28]. Available from: http://data.vancouver.ca/datacatalogue/parks.htm 162.  DMTI Spatial Inc. CanMap Water, v2014.3 [Internet]. 2014 [cited 2015 Jun 23]. Available    147  from: http://hdl.handle.net.ezproxy.library.ubc.ca/11272/10185 163.  Musqueam Indian Band. Musqueam [Internet]. 2018 [cited 2018 Oct 28]. Available from: https://www.musqueam.bc.ca/ 164.  City of Richmond. Finn Slough Heritage Area [Internet]. 2018 [cited 2018 Oct 28]. Available from: https://www.richmond.ca/plandev/planning2/heritage/HeritageInv/details.aspx?ID=167 165.  Taylor MS, Wheeler BW, White MP, Economu T, Osborne NJ. Urban street tree density and antidepressant prescription rates in London. Landsc Urban Plan. 2015;136:174–9.  166.  Ulmer JM, Wolf KL, Backman DR, Tretheway RL, Blain CJ, O’Neil-Dunne JP, et al. Multiple health benefits of urban tree canopy: the mounting evidence for a green prescription. Health Place. 2016;42:54–62.  167.  Holtan MT, Dieterlen SL, Sullivan WC. Social life under cover: tree canopy and social capital in Baltimore, Maryland. Environ Behav. 2015;47(5):502–25.  168.  Kaplan R. The nature of the view from home: psychological benefits. Environ Behav. 2001;33(4):507–42.  169.  Ulrich RS. View through a window may influence recovery from surgery. Science (80- ). 1984;224(4647):420–1.  170.  Pretty J. How nature contributes to mental and physical health. Spitituality Heal Int. 2004;5(April):68–78.  171.  City of White Rock. Trees Dataset [Internet]. 2014 [cited 2015 Oct 28]. Available from: http://data.whiterockcity.ca/dataset/tree 172.  District of North Vancouver. Open Data - Street Trees [Internet]. 2015 [cited 2015 Oct 28]. Available from: http://geoweb.dnv.org/data/metadata.php?dataset=EnvStreetTree 173.  City of Surrey. Datasets - Park Specimen Trees [Internet]. 2014 [cited 2015 Oct 28]. Available from: http://data.surrey.ca/dataset/park-specimen-trees 174.  City of New Westminster. Geocoded Tree Data. Personal Communication; 2014.     148  175.  City of Vancouver. Geocoded Tree Data. Personal Communication; 2017.  176.  Astell-Burt T, Mitchell R, Hartig T. The association between green space and mental health varies across the lifecourse. A longitudinal study. J Epidemiol Community Health. 2014;68(6):578–83.  177.  Reklaitiene R, Grazuleviciene R, Dedele A, Virviciute D, Vensloviene J, Tamosiunas A, et al. The relationship of green space, depressive symptoms, and perceived general health in urban population. Scand J Public Health. 2014;42:669–76.  178.  Kaplan S. Meditation, restoration, and the management of mental fatigue. Environ Behav. 2001 Jul;33(4):480–506.  179.  Lee ACK, Jordan HC, Horsley J. Value of urban green spaces in promoting healthy living and wellbeing: prospects for planning. Risk Manag Healthc Policy. 2015;8:131–7.  180.  Edwards N, Hooper P, Trapp GSA, Bull F, Boruff B, Giles-Corti B. Development of a Public Open Space Desktop Auditing Tool (POSDAT): a remote sensing approach. Appl Geogr. 2013;38(1):22–30.  181.  Sugiyama T, Gunn LD, Christian H, Francis J, Foster S, Hooper P, et al. Quality of public open spaces and recreational walking. Am J Public Health. 2015;105(12):2490–5.  182.  Smith W. The 2011 National Household Survey—the complete statistical story [Internet]. StatCan Blog. 2015 [cited 2016 Jan 23]. Available from: http://www.statcan.gc.ca/eng/blog-blogue/cs-sc/2011NHSstory 183.  Randomness and Integrity Services Ltd. Random Number Generator [Internet]. 2015 [cited 2015 Jun 23]. Available from: https://random.org 184.  Edwards N, Hooper P. POSDAT Training Manual [Internet]. Perth; 2011. Available from: http://www.see.uwa.edu.au/research/cbeh/projects/?a=2147374 185.  Mygind L, Bentsen P, Badland H, Edwards N, Hooper P, Villanueva K. Public Open Space Desktop Auditing Tool – establishing appropriateness for use in Australian regional and urban settings. Urban For Urban Green. 2016;20:65–70.     149  186.  Metro Vancouver. Municipal Boundaries [Internet]. 2015 [cited 2015 Jul 11]. Available from: http://www.metrovancouver.org/data/regional-planning/municipal boundaries 187.  Ekkel ED, de Vries S. Nearby green space and human health: Evaluating accessibility metrics. Landsc Urban Plan. 2017;157:214–20.  188.  Wheeler BW, White M, Stahl-Timmins W, Depledge MH. Does living by the coast improve health and wellbeing? Health Place. 2012;18(5):1198–201.  189.  R Core Development Team. R: a language and environment for statistical computing, 3.2.4 [Internet]. Vienna: R Foundation for Statistical Computing; 2015. Available from: https://www.r-project.org 190.  Villeneuve PJ, Jerrett M, Su JG, Burnett RT, Chen H, Wheeler AJ, et al. A cohort study relating urban green space with mortality in Ontario, Canada. Environ Res. 2012;115:51–8.  191.  United States Geological Survey. Using Landsat 7 Data [Internet]. 2013 [cited 2015 Aug 13]. Available from: http://landsat.usgs.gov/using_Landsat_7_data.php 192.  United States Geological Survey. Filling the Gaps to use in Scientific Analysis [Internet]. 2013 [cited 2015 Aug 13]. Available from: http://landsat.usgs.gov/sci_an.php 193.  Lovasi GS, Quinn JW, Neckerman KM, Perzanowski MS, Rundle A. Children living in areas with more street trees have lower prevalence of asthma. J Epidemiol Community Heal. 2008 Jul;62(7):647–9.  194.  Arku G, Yeboah IEA, Nyantakyi-Frimpong H. Public parks as an element of urban planning: a missing piece in Accra’s growth and development. Local Environ. 2016;9839(July):1–16.  195.  Lee KE, Williams KJH, Sargent LD, Williams NSG, Johnson KA. 40-second green roof views sustain attention: the role of micro-breaks in attention restoration. J Environ Psychol. 2015;42:182–9.  196.  Nielsen TS, Hansen KB. Do green areas affect health? Results from a Danish survey on    150  the use of green areas and health indicators. Health Place. 2007 Dec;13(4):839–50.  197.  Weden MM, Carpiano RM, Robert SA. Subjective and objective neighborhood characteristics and adult health. Soc Sci Med. 2008;66(6):1256–70.  198.  Raudenbush SW, Sampson RJ. Ecometrics: toward a science of assessing ecological settings, with application to the systematic social observation of neighborhoods. Sociol Methodol. 1999;29(1):1–41.  199.  Tooke TR, Coops NC, Goodwin NR, Voogt JA. Extracting urban vegetation characteristics using spectral mixture analysis and decision tree classifications. Remote Sens Environ. 2009;113(2):398–407.  200.  Yu S, Yu B, Song W, Wu B, Zhou J, Huang Y, et al. View-based greenery: A three-dimensional assessment of city buildings’ green visibility using Floor Green View Index. Landsc Urban Plan. 2016;152:13–26.  201.  Keniger LE, Gaston KJ, Irvine KN, Fuller RA. What are the benefits of interacting with nature? Int J Environ Res Public Health. 2013;10(3):913–35.  202.  Sunderland A, Findlay LC. Perceived need for mental health care in Canada: Results from the 2012 Canadian Community Health Survey-Mental Health. Heal Reports. 2013;24(9):3–9.  203.  Beyer KMM, Kaltenbach A, Szabo A, Bogar S, Javier Nieto F, Malecki KM. Exposure to neighborhood green space and mental health: Evidence from the survey of the health of Wisconsin. Int J Environ Res Public Health. 2014;11(3):3453–72.  204.  Gubbels JS, Kremers SPJ, Droomers M, Hoefnagels C, Stronks K, Hosman C, et al. The impact of greenery on physical activity and mental health of adolescent and adult residents of deprived neighborhoods: A longitudinal study. Heal Place. 2016;40:153–60.  205.  Helbich M, Klein N, Roberts H, Hagedoorn P, Groenewegen PP. More green space is related to less antidepressant prescription rates in the Netherlands: A Bayesian geoadditive quantile regression approach. Environ Res. 2018;166(May):290–7.     151  206.  Segal Z V, Kennedy SH, Cohen NL, Group CD. Clinical guidelines for the treatment of depressive disorders. V. Combining psychotherapy and pharmacotherapy. Can J Psychiatry-Revue Can Psychiatr. 2001 Jun;46(Suppl 1):59–62S.  207.  Spijker J, van Straten A, Bockting CLH, Meeuwissen JAC, van Balkom AJLM. Psychotherapy, antidepressants, and their combination for chronic major depressive disorder: a systematic review. Can J Psychiatry-Revue Can Psychiatr. 2013 Jul;58(7):386–92.  208.  Morgan S, Smolina K, Mooney D, Raymond C, Bowen ML, Gorczynski C, et al. The Canadian Rx Atlas. 3rd ed. The Centre for Health Services and Policy Research; 2013.  209.  Dadvand P, Tischer C, Estarlich M, Llop S, Dalmau-Bueno A, López-Vicente M, et al. Lifelong residential exposure to green space and attention: A population-based prospective study. Environ Health Perspect. 2017;125(9):1–8.  210.  Crouse DL, Pinault L, Balram A, Hystad P, Peters PA, Chen H, et al. Urban greenness and mortality in Canada’s largest cities: a national cohort study. Lancet Planet Heal. 2017;1(7):e289–97.  211.  Rugel EJ, Henderson SB, Carpiano RM, Brauer M. Beyond the Normalized Difference Vegetation Index (NDVI): Developing a Natural Space Index for population-level health research. Environ Res. 2017;159C:474–83.  212.  DMTI Spatial Inc. CanMap Water, v2010.3 [Internet]. [cited 2018 Jan 21]. Available from: http://hdl.handle.net/11272/HITUI 213.  Jiang B, Chang CY, Sullivan WC. A dose of nature: Tree cover, stress reduction, and gender differences. Landsc Urban Plan. 2014;132:26–36.  214.  BC Ministry of Health. PharmaNet. V2. Data Extract. Data Stewardship Committee [Internet]. BC Ministry of Health; 2016. Available from: http://www.popdata.bc.ca/data 215.  Population Data BC. PharmaNet [Internet]. 2018 [cited 2018 Oct 21]. Available from: https://www.popdata.bc.ca/data/health/PharmaNet    152  216.  WHO Collaborating Centre for Drug Methodology. Guidelines for ATC Cassification and DDD Assignment 2013. Vol. 16th ed. Oslo; 2012.  217.  Health Canada. Drug Product Database Online Query [Internet]. 2015 [cited 2015 Jan 21]. Available from: https://health-products.canada.ca/dpd-bdpp/index-eng.jsp 218.  BC PharmaNet. Product Identification Numbers [Internet]. 2018 [cited 2018 Nov 3]. Available from: https://www2.gov.bc.ca/gov/content/health/practitioner-professional-resources/pharmacare/pharmacies/product-identification-numbers 219.  British Columbia Ministry of Health. Consolidation File (MSP Registration & Premium Billing). V2. Data Extract. [Internet]. Population Data BC; 2016. Available from: http://www.popdata.bc.ca/data 220.  Statistics Canada. Statistics Canada Income Band Data. Catalogue Number: 13C0016. Data Extract. [Internet]. Population Data BC; 2016. Available from: http://www.popdata.bc.ca/data 221.  Statistics Canada. 2011 Census Dictionary - Dissemination Area (DA) Defintion [Internet]. 2015 [cited 2018 May 1]. Available from: http://www12.statcan.gc.ca/census-recensement/2011/ref/dict/geo021-eng.cfm 222.  Statistics Canada. Geographic Attribute File, Census Year 2011 [Internet]. 2012 [cited 2013 Mar 23]. Available from: http://www5.statcan.gc.ca/olc-cel/olc?ObjId=92-151-X2011001&ObjType=46&lang=en 223.  Walk Score Data Services. Walk Score® Data [Internet]. Redfin Real Estate; 2016 [cited 2016 Mar 25]. Available from: https://www.redfin.com 224.  Walk Score. Walk Score Methodology [Internet]. 2018 [cited 2018 Nov 3]. Available from: https://www.walkscore.com/methodology.shtml 225.  Duncan DT, Aldstadt J, Whalen J, Melly SJ, Gortmaker SL. Validation of Walk Score (R) for estimating neighborhood walkability: an analysis of four US metropolitan areas. Int J Environ Res Public Health. 2011 Nov;8(11):4160–79.     153  226.  Population Data BC. About PopData [Internet]. 2018 [cited 2018 Oct 21]. Available from: https://www.popdata.bc.ca/aboutus 227.  Census of Canada. Postal Code Conversion File, March 2008 Postal Codes, 2006 [Internet]. 2009 [cited 2018 Jan 21]. Available from: http://hdl.handle.net/11272/LNEEC 228.  Census of Canada. Postal Code Conversion File, December 2009 Postal Codes, 2006 [Internet]. 2010 [cited 2018 Jan 21]. Available from: http://hdl.handle.net/11272/EGNM7 229.  Census of Canada. Postal Code Conversion File, May 2011 Postal Codes, 2006 [Internet]. [cited 2018 Jan 21]. Available from: http://hdl.handle.net/11272/DCQV4 230.  Census of Canada. Postal Code Conversion File, October 2010 Postal Codes, 2006 [Internet]. 2011 [cited 2018 Jan 21]. Available from: http://hdl.handle.net/11272/KO7Q0 231.  Statistics Canada. Focus on Geography Series, 2011 Census - Census Subdivision of Vancouver, CY - British Columbia [Internet]. 2016 [cited 2018 Oct 28]. Available from: https://www12.statcan.gc.ca/census-recensement/2011/as-sa/fogs-spg/Facts-csd-eng.cfm?LANG=Eng&GK=CSD&GC=5915022 232.  White MP, Alcock I, Wheeler BW, Depledge MH. Coastal proximity, health and well-being: Results from a longitudinal panel survey. Health Place. 2013;23:97–103.  233.  Sharma DC. Ports in a storm. Environ Health Perspect. 2006;114(4):A222–5.  234.  Lefrancois C, Porte E, Needoba A, Cox T. Tools for stormwater management and urban forestry. Planning Institute of British Columbia Newsletter. 2016;12–4.  235.  United Nations Human Settlements Programme (UN-HABITAT). State of the World’s Cities 2012/2013: Prosperity of Cities. Nairobi; 2012.  236.  Boers S, Hagoort K, Scheepers F, Helbich M. Does residential green and blue space promote recovery in psychotic disorders? A cross-sectional study in the province of Utrecht, the Netherlands. Int J Environ Res Public Heal. 2018;15(10):2195–202.  237.  Beck CA, Williams JVA, Wang JL, Kassam A, El-Guebaly N, SR C, et al. Psychotropic medication use in Canada. Can J Psychiatry. 2005;50(10):605.     154  238.  Patten SB, Williams JV, Lavorato DH, Wang JL, Mcdonald K, Bulloch AG. Descriptive epidemiology of major depressive disorder in Canada in 2012. 2015;60(1):23–30.  239.  Groenewegen PP, Zock JP, Spreeuwenberg P, Helbich M, Hoek G, Ruijsbroek A, et al. Neighbourhood social and physical environment and general practitioner assessed morbidity. Heal Place. 2018;49:68–84.  240.  Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al. The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). Jama. 2003;289(23):3095–105.  241.  Blair A, Ross NA, Gariepy G, Schmitz N. How do neighborhoods affect depression outcomes? A realist review and a call for the examination of causal pathways. Soc Psychiatry Psychiatr Epidemiol. 2014;49(6):873–87.  242.  Gong Y, Palmer S, Gallacher J, Marsden T, Fone D. A systematic review of the relationship between objective measurements of the urban environment and psychological distress. Environ Int. 2016;96:48–57.  243.  British Columbia Ministry of Health. Fair PharmaCare Plan [Internet]. 2018 [cited 2018 Oct 28]. Available from: https://www2.gov.bc.ca/gov/content/health/health-drug-coverage/pharmacare-for-bc-residents/who-we-cover/fair-pharmacare-plan 244.  Gan WQ, Davies HW, Koehoorn M, Brauer M. Association of long-term exposure to community noise and traffic-related air pollution with coronary heart disease mortality. Am J Epidemiol. 2012;175(9):898–906.  245.  Canadian Urban Environmental Health Research Consortium (CANUE). Postal Code User Guide [Internet]. 2018. Available from: http://canue.ca/wp-content/uploads/2018/03/CANUE-Postal-Code-User-Guide.pdf 246.  Vlahov D, Galea S, Freudenberg N. The urban health “advantage.” J Urban Heal. 2005;82(1):1–4.  247.  Dodman D. Blaming cities for climate change? An analysis of urban greenhouse gas    155  emissions inventories. Environ Urban. 2009;21(1):185–201.  248.  Lambert KG, Nelson RJ, Jovanovic T, Cerdá M. Brains in the city: neurobiological effects of urbanization. Neurosci Biobehav Rev. 2015;58:107–22.  249.  Parra DC, Gomez LF, Sarmiento OL, Buchner D, Brownson R, Schimd T, et al. Perceived and objective neighborhood environment attributes and health related quality of life among the elderly in Bogota, Colombia. Soc Sci Med. 2010;70(7):1070–6.  250.  Evans GW. The built environment and mental health. J Urban Heal. 2003;80(4):536–55.  251.  Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspect Psychol Sci. 2015;10(2):227–37.  252.  Valtorta NK, Kanaan M, Gilbody S, Ronzi S, Hanratty B. Loneliness and social isolation as risk factors for coronary heart disease and stroke: Systematic review and meta-analysis of longitudinal observational studies. Heart. 2016;102(13):1009–16.  253.  Umberson D, Karas Montez J. Social relationships and health: A flashpoint for health policy. J Health Soc Behav. 2010;51(Suppl 1):S54–66.  254.  Pope D, Tisdall R, Middleton J, Verma A, van Ameijden E, Birt C, et al. Quality of and access to green space in relation to psychological distress. Eur J Public Health. 2015;1–4.  255.  Maas J, Spreeuwenberg P, Van Winsum-Westra M, Verheij RA, de Vries S, Groenewegen PP. Is green space in the living environment associated with people’s feelings of social safety? Environ Plan A. 2009 Jul;41(7):1763–77.  256.  Helbich M, Beurs D De, Kwan M, Connor RCO, Groenewegen PP. Natural environments and suicide mortality in the Netherlands: A cross-sectional, ecological study. Lancet Planet Heal. 2018;2(3):e134–9.  257.  de Vries S, van Dillen SME, Groenewegen PP, Spreeuwenberg P. Streetscape greenery and health: stress, social cohesion and physical activity as mediators. Soc Sci Med.    156  2013;94:26–33.  258.  Carpiano RM, Hystad PW. “Sense of community belonging” in health surveys: What social capital is it measuring? Heal Place. 2011;17(2):606–17.  259.  Schellenberg G, Lu C, Schimmele C, Hou F. The correlates of self-assessed community belonging in Canada: Social capital, neighbourhood characteristics, and rootedness. Soc Indic Res. Springer Netherlands; 2017;1–22.  260.  Statistics Canada. Infographic: Demographic Estimates, Census Metropolitan Areas - Canada, 2015 [Internet]. 2016 [cited 2016 Jun 11]. Available from: http://www.statcan.gc.ca/pub/11-627-m/11-627-m2016002-eng.htm 261.  Harvard Medical School Department of Health Care Policy. The World Health Organization World Mental Health Composite International Diagnostic Interview (WHO WMH-CIDI) [Internet]. 2018 [cited 2018 May 1]. Available from: https://www.hcp.med.harvard.edu/wmhcidi 262.  Kessler RC, Barker PR, Colpe LJ, Epstein JF, Gfroerer JC, Hiripi E, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003;60(2):184–9.  263.  Keyes CL. The mental health continuum: From languishing to flourishing in life. J Health Soc Behav. 2002;(Jun 1):207–22.  264.  Lamers SMA, Westerhof GJ, Bohlmeijer ET, Ten Klooster PM, Keyes CLM. Evaluating the psychometric properties of the Mental Health Continuum-Short Form (MHC-SF). J Clin Psychol. 2011;67(1):99–110.  265.  Orpana H, Vachon J, Dykxhoorn J, Jayaraman G. Measuring positive mental health in Canada: Construct validation of the Mental Health Continuum-Short Form. Heal Promot Chronic Dis Prev Canada Res Policy Pract. 2017;37(4):123–30.  266.  Kessler RC, Green JG, Gruber MJ, Sampson NA, Bromet E, Cultan M, et al. Screening for serious mental illness in the general population with the K6 screening scale: Results from the WHO World Mental Health (WMH) Survey Initiative. Int J Methods Psychiatr    157  Res. 2010;19(S1):4–22.  267.  MacKinnon D, Fairchild A, Fritz M. Mediation analysis. Annu Rev Psychol. 2007;58:593–602.  268.  Hayes AF. Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Commun Monogr. 2009;76(4):408–20.  269.  Iacobucci D. Mediation analysis and categorical variables: The final frontier. J Consum Psychol. 2012;22(4):582–94.  270.  Dadvand P, Bartoll X, Basagaña X, Dalmau-Bueno A, Martinez D, Ambros A, et al. Green spaces and general health: Roles of mental health status, social support, and physical activity. Environ Int. 2016;91:161–7.  271.  Dzhambov A, Hartig T, Markevych I, Tilov B, Dimitrova D. Urban residential greenspace and mental health in youth: Different approaches to testing multiple pathways yield different conclusions. Environ Res. 2018;160(June 2017):47–59.  272.  Leyden KM. Social capital and the built environment: The importance of walkable neighborhoods. Am J Public Health. 2003;93(9):1546–51.  273.  Wood L, Frank LD, Giles-Corti B. Sense of community and its relationship with walking and neighborhood design. Soc Sci Med. 2010;70(9):1381–90.  274.  United Kingdom Office for National Statistics. Census Geography [Internet]. 2018 [cited 2018 Mar 29]. Available from: https://www.ons.gov.uk/methodology/geography/ukgeographies/censusgeography 275.  Ward Thompson C, Aspinall P, Roe J, Robertson L, Miller D. Mitigating stress and supporting health in deprived urban communities: The importance of green space and the social environment. Int J Environ Res Public Health. 2016;13(4).  276.  Fuertes E, Markevych I, Bowatte G, Gruzieva O, Gehring U, Becker A, et al. Residential greenness is differentially associated with childhood allergic rhinitis and aeroallergen sensitization in seven birth cohorts. Allergy. 2016;71(10):1461–71.     158  277.  Vancouver Foundation. Connect and Engage: A Survey of Metro Vancouver, 2017 [Internet]. Vancouver; 2017. Available from: https://www.vancouverfoundation.ca/sites/default/files/publications/VF-Connect-Engage-report.pdf 278.  Rugel EJ, Carpiano RM, Henderson SB, Brauer M. Exposure to natural space, sense of community belonging, and adverse mental health outcomes across an urban region. Environ Res. 2019;171(April):365–77.  279.  City of Vancouver, Vancouver Park Board. Urban Forest Strategy: 2018 Update [Internet]. Vancouver; 2018. Available from: https://parkboardmeetings.vancouver.ca/2018/20180430/REPORT-UrbanForestStrategy2018Update-20180430.pdf 280.  Reid CE, Clougherty JE, Schmool JL, Kubzansky LD. Is all urban green space the same? A comparison of the health benefits of trees and grass in New York City. Int J Environ Res Public Heal. 2017;14(11):1411–28.  281.  Weimann H, Rylander L, Albin M, Skarback E, Grahn P, Östergren P-O, et al. Effects of changing exposure to neighbourhood greenness on general and mental health: a longitudinal study. Health Place. 2015;33C:48–56.  282.  Kim J, Kim H. Demographic and environmental factors associated with mental health: A cross-sectional study. Int J Environ Res Public Health. 2017;14(4).  283.  Helbich M. Toward dynamic urban environmental exposure assessments in mental health research. Environ Res. 2018;161(November 2017):129–35.  284.  City of Vancouver, Vancouver Park Board. Urban Forest Strategy: 2018 Update. Vancouver; 2018.  285.  Flörke M, Kynast E, Bärlund I, Eisner S, Wimmer F, Alcamo J. Domestic and industrial water uses of the past 60 years as a mirror of socio-economic development: A global simulation study. Glob Environ Chang. 2013;23(1):144–56.     159  286.  Jones AL. Regenerating urban waterfronts—creating better futures—from commercial and leisure market places to cultural quarters and innovation districts. Plan Pract Res. 2017;32(3):333–44.  287.  London Parks and Gardens Trust. South Bank Regeneration [Internet]. 2018 [cited 2018 Nov 6]. Available from: http://www.londongardenstrust.org/features/southbank.htm 288.  Lee JY, Anderson CD. The restored Cheonggyecheon and the quality of life in Seoul. J Urban Technol. 2013;20(4):3–22.  289.  Kaczynski AT, Stanis SAW, Besenyi GM, Wilhelm Stanis SA, Besenyi GM. Development and testing of a Community Stakeholder Park Audit Tool. Am J Prev Med. 2012;42(3).  290.  Clarke P, Ailshire J, Melendez R, Bader M, Morenoff J. Using Google Earth to conduct a neighborhood audit: reliability of a virtual audit instrument. Health Place. 2010 Nov;16(6):1224–9.  291.  Rundle AG, Bader MDM, Richards CA, Neckerman KM, Teitler JO. Using Google Street View to audit neighborhood environments. Am J Prev Med. 2011 Jan;40(1):94–100.  292.  Vanwolleghem G, Van Dyck D, Ducheyne F, De Bourdeaudhuij I, Cardon G. Assessing the environmental characteristics of cycling routes to school: a study on the reliability and validity of a Google Street View-based audit. Int J Health Geogr. 2014 Jun 10;13(1):19.               160  Appendices Appendix A  – Natural Space Index Variable Table Description Type Name Postal Code #s/%s Surrounding Greenness 2012 Wet Season – MODIS – NDVI (250 x 250) Continuous 12_W_M_N 60,242 (100%) 2012 Wet Season – MODIS – EVI (250 x 250) Continuous 12_W_M_E 60,242 (100%) 2012 Dry Season – MODIS – NDVI (250 x 250) Continuous 12_D_M_N 60,242 (100%) 2012 Dry Season – MODIS – EVI (250 x 250) Continuous 12_D_M_ E 60,242 (100%) 2015 Dry Season – MODIS – NDVI (500 x 500) Continuous 15_NMD_500 60,242 (100%) 2015 Dry Season – MODIS – EVI (500 x 500) Continuous 15_EMD_500 60,242 (100%) 2015 Wet Season – MODIS – NDVI (500 x 500) Continuous 15_NMW_500 60,242 (100%) 2015 Wet Season – MODIS – EVI (500 x 500) Continuous 15_EMW_500 60,242 (100%) 2015 Dry Season – Landsat – NDVI (30 x 30) Continuous 15_NLD_F_B 60,242 (100%) 2015 Dry Season – Landsat – NDVI (250 x 250) Continuous 15_NLDF_250 60,242 (100%) 2015 Dry Season – Landsat – NDVI (500 x 500) Continuous 15_NLDF_500 60,242 (100%) 2015 Dry Season – Landsat – NDVI (30 x 30) Continuous 15_N_L_W_F 60,242 (100%) 2015 Wet Season – Landsat – NDVI (250 x 250) Continuous 15_NLWF_250 60,242 (100%) 2015 Wet Season – Landsat – NDVI (500 x 500) Continuous 15_NLWF_500 60,242 (100%) 2015 Annual – MODIS – NDVI (250 x 250) Continuous 15NLAS250 60,242 (100%) 2015 Annual – MODIS – NDVI (250 x 250); bluespace-corrected Continuous 15NLAS250B 60,242 (100%) 2015 Annual – MODIS – NDVI (250 x 250) Ordinal 15NLAS250_CAT 60,242 (100%) 2015 Annual – MODIS – NDVI (250 x 250); bluespace-corrected Ordinal 15NLAS250B_CAT 60,242 (100%) Natural Space Presence & Accessibility Public & Private Greenspace (100m) Continuous AG_100_PER 16,070 (26.68%) Public & Private Greenspace (250m) Continuous AG_250_PER 38,776 (64.37%) Public & Private Greenspace (250m) Ordinal AG_250_PER_CAT 38,776 (64.37%) Public & Private Greenspace (500m) Continuous AG_500_PER  55,627 (92.34%) Public Greenspace (100m) Continuous PG_100_PER 15,466 (25.67%) Public Greenspace (250m) Continuous PG_250_PER 37,876 (62.87%)    161  Description Type Name Postal Code #s/%s Public Greenspace (500m) Continuous PG_500_PER 55,250 (91.71%) Public & Private Greenspace & Bluespace (100m) Continuous NS_100_PER 16,630 (27.61%) Public & Private Greenspace & Bluespace (250m) Continuous NS_250_PER 40,083 (66.54%) Public & Private Greenspace & Bluespace (250m) Ordinal NS_250_PER_CAT 40,083 (66.54%) Public & Private Greenspace & Bluespace (250m) Continuous NS_500_PER 56,789 (66.54%) Public Greenspace & Bluespace (100m) Continuous PNS_100_PR 16,037 (26.62%) Public Greenspace & Bluespace (250m) Continuous PNS_250_PR 39,347 (65.31%) Public Greenspace & Bluespace (500m) Continuous PNS_250_PR 56,513 (93.81%) Bluespace (100m) Continuous BS_100_PER 1,136 (1.89%) Bluespace (250m) Continuous BS_250_PER 5,156 (8.56%) Bluespace (500m) Continuous BS_500_PER  14,557 (24.16%) Binary Natural Space Bluespace (1,000m) [Yes (1) vs. No (0)]  Binary BS_1KM 34,258 (56.87%) Bluespace (1,600m/1 mi) [Yes (1) vs. No (0)]  Binary BS_1MI 49,696 (82.49%) Public Greenspace (400m) [Yes (1) vs. No (0)]  Binary PG_400 51,134 (84.88%) Street Tree Density Trees (100m) Continuous TREE_100 16,845 (27.96%) Trees (250m) Continuous TREE_250 17,081 (28.35%) Trees (250m) Ordinal TREE_250_CAT 17,081 (28.35%) Trees (500m) Continuous TREE_500 17,626 (29.26%) Park Quality (POSDAT) Park Name Nominal PARK_NAME 14,108 (23.42%)   Park Type Nominal FORM 14,108 (23.42%)   Activity Score (400m) Continuous ACTIVITY_400 14,108 (23.42%) Environmental Quality Score (400m) Continuous EQ_400 14,108 (23.42%) Dogs Off-Leash (400m) [Yes vs. No]  Continuous DOGS_400 14,108 (23.42%) Amenity Score (400m) Continuous AMEN_400 14,108 (23.42%) Safety Score (400m) Continuous SAFETY_400 14,108 (23.42%) Summary Score (400m) Continuous SUMMARY_400 14,108 (23.42%) Activity Score (1600m) Continuous ACTIVITY_1600 49,308 (81.85%) Environmental Quality Score (1600m) Continuous EQ_1600 49,308 (81.85%) Dogs Off-Leash (1600m)  [Yes (1) vs. No (0)]  Continuous DOGS_1600 49,308 (81.85%) Amenity Score (1600m) Continuous AMEN_1600 49,308 (81.85%) Safety Score (1600m) Continuous SAFETY_1600 49,308 (81.85%) Summary Score (1600m) Continuous SUMMARY_1600 49,308 (81.85%) Summary Score (1600m) Ordinal SUMMARY_1600_ CAT 49,308 (81.85%)    162  Appendix B  – World Health Organization Anatomical Therapeutic Chemical (ATC) Classification System Codes and Corresponding BC PharmaNet Drug Identification Number/Product Identification Number ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA CHLORDIAZEPOXIDE N05BA02 12629 LIBRIUM CAP 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 12637 LIBRIUM CAP 10MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 12645 LIBRIUM CAP 25MG N05B N05BA DIAZEPAM N05BA01 12874 VALIUM INJ ROCHE 5MG/ML N05B N05BA DIAZEPAM N05BA01 13110 VALIUM ORAL SUSPENSION 5MG/5ML N05B N05BA DIAZEPAM N05BA01 13277 VALIUM 2 TAB N05B N05BA DIAZEPAM N05BA01 13285 VALIUM 5 TAB N05B N05BA DIAZEPAM N05BA01 13293 VALIUM 10 TAB N05B N05BA CHLORDIAZEPOXIDE N05BA02 13463 SOLIUM 5 MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 13471 SOLIUM 10 MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 13498 SOLIUM 25 MG N05B N05BA DIAZEPAM N05BA01 13757 VIVOL 2MG N05B N05BA DIAZEPAM N05BA01 13765 VIVOL N05B N05BA DIAZEPAM N05BA01 13773 VIVOL N05B N05BC MEPROBAMATE N05BC01 13846 MILTOWN TABLETS 400MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 20915 NOVO-POXIDE 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 20923 NOVO-POXIDE 10MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 20931 NOVO-POXIDE 25MG N05B N05BC MEPROBAMATE N05BC01 21539 NOVO-MEPRO 200MG N05B N05BC MEPROBAMATE N05BC01 21547 NOVO-MEPRO 400MG N05B N05BB HYDROXYZINE N05BB01 24376 ATARAX CAP 10MG N05B N05BB HYDROXYZINE N05BB01 24384 ATARAX CAP 25MG N05B N05BB HYDROXYZINE N05BB01 24392 ATARAX CAP 50MG N05B N05BB HYDROXYZINE N05BB01 24589 ATARAX IM SOL 50MG/ML N05B N05BB HYDROXYZINE N05BB01 24694 ATARAX SYRUP 2MG/ML N05B N05BC MEPROBAMATE N05BC01 34142 EQUANIL TAB 400MG N05B N05BC MEPROBAMATE N05BC01 92711 MEPROBAMATE TAB 200MG N05B N05BC MEPROBAMATE N05BC01 92738 MEPROBAMATE TAB 400MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 134325 MEDILIUM CAP 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 134333 MEDILIUM CAP 10 10MG N05B N05BC MEPROBAMATE N05BC01 134368 MEDITRAN 400MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 156590 CHLORDIAZEPOXIDE CAP 10MG N05B N05BC MEPROBAMATE N05BC01 156620 MEPROBAMATE TAB 400MG N05B N05BC MEPROBAMATE N05BC01 210846 MEPROBAMATE TAB 400MG N05B N05BA OXAZEPAM N05BA04 231363 SERAX TAB 30MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 235873 CORAX CAP 10MG    163  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA CHLORDIAZEPOXIDE N05BA02 251259 CHLORDIAZEPOXIDE HCL CAP 25MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 251267 CHLORDIAZEPOXIDE HCL CAP 10MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 264911 TRANXENE CAP 15MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 264938 TRANXENE CAP 3.75MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 264946 TRANXENE CAP 7.5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 267090 CORAX CAP 25MG N05B N05BA DIAZEPAM N05BA01 272434 NOVO-DIPAM TAB 2MG N05B N05BA DIAZEPAM N05BA01 272442 NOVO-DIPAM TAB 5MG N05B N05BA DIAZEPAM N05BA01 272450 NOVO-DIPAM TAB 10MG N05B N05BA DIAZEPAM N05BA01 272639 E PAM TAB 10MG N05B N05BA DIAZEPAM N05BA01 272647 E PAM TAB 2MG N05B N05BA DIAZEPAM N05BA01 276642 MEVAL 5 N05B N05BA DIAZEPAM N05BA01 276650 MEVAL 2 N05B N05BA DIAZEPAM N05BA01 280429 E PAM TAB 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 295051 CORAX CAP 5MG N05B N05BA OXAZEPAM N05BA04 295698 SERAX TAB 15MG N05B N05BA OXAZEPAM N05BA04 295701 SERAX TAB 10MG N05B N05BA DIAZEPAM N05BA01 303461 DIAZEPAM TAB 5MG N05B N05BC MEPROBAMATE N05BC01 305294 MEPROBAMATE TAB 400MG N05B N05BA DIAZEPAM N05BA01 313580 DIAZEPAM 5 TAB N05B N05BC MEPROBAMATE N05BC01 337943 APO MEPROBAMATE TAB 400MG N05B N05BA LORAZEPAM N05BA06 348325 ATIVAN TAB 1MG N05B N05BA LORAZEPAM N05BA06 348333 ATIVAN TAB 2MG N05B N05BA DIAZEPAM N05BA01 362158 APO DIAZEPAM TAB 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 363596 CHLORDIAZEPOXIDE CAP 5MG N05B N05BA DIAZEPAM N05BA01 396230 DIAZEPAM TAB 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 398403 CHLORDIAZEPOXIDE CAP 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 398411 CHLORDIAZEPOXIDE CAP 10MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 398438 CHLORDIAZEPOXIDE CAP 25MG N05B N05BA LORAZEPAM N05BA06 399124 ATIVAN TAB 0.5MG N05B N05BA DIAZEPAM N05BA01 399728 DIAZEPAM INJECTION USP N05B N05BA OXAZEPAM N05BA04 402680 APO OXAZEPAM TAB 10MG N05B N05BA OXAZEPAM N05BA04 402737 APO OXAZEPAM TAB 30MG    164  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA OXAZEPAM N05BA04 402745 APO OXAZEPAM TAB 15MG N05B N05BA DIAZEPAM N05BA01 405329 APO DIAZEPAM TAB 2MG N05B N05BA DIAZEPAM N05BA01 405337 APO DIAZEPAM TAB 10MG N05B N05BA OXAZEPAM N05BA04 414247 OXPAM N05B N05BA OXAZEPAM N05BA04 414255 OXPAM N05B N05BA OXAZEPAM N05BA04 414263 OXPAM N05B N05BA CHLORDIAZEPOXIDE N05BA02 430927 CHLORDIAZEPOXYDE CAP 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 430935 CHLORDIAZEPOXYDE CAP 10MG N05B N05BC MEPROBAMATE N05BC01 431095 MEPROBAMATE TAB 400MG N05B N05BA DIAZEPAM N05BA01 434388 DIAZEPAM 10TAB 10MG N05B N05BA DIAZEPAM N05BA01 434396 DIAZEPAM 2TAB 2MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 434426 CHLORDIAZEPOXYDE HCL CAP 25MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 440183 CHLORDIAZEPOXIDE HCL CAP 5MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 448737 CHLORDIAZEPOXYDE 25 CAP N05B N05BA DIAZEPAM N05BA01 466891 DIAZEPAM TAB 10MG N05B N05BA DIAZEPAM N05BA01 466905 DIAZEPAM TAB 2MG N05B N05BA OXAZEPAM N05BA04 483893 OXAZEPAM TAB 10MG N05B N05BA OXAZEPAM N05BA04 483907 OXAZEPAM TABLETS 30MG N05B N05BA OXAZEPAM N05BA04 483915 OXAZEPAM TABLETS 15MG N05B N05BA OXAZEPAM N05BA04 496529 NOVOXAPAM TAB 15MG N05B N05BA OXAZEPAM N05BA04 496537 NOVOXAPAM TAB 30MG N05B N05BA OXAZEPAM N05BA04 497754 OXAZEPAM 10 TAB 10MG N05B N05BA OXAZEPAM N05BA04 497762 OXAZEPAM 15 TAB 15MG N05B N05BA OXAZEPAM N05BA04 497770 OXAZEPAM 30 TAB 30MG N05B N05BA OXAZEPAM N05BA04 500852 NOVOXAPAM TAB 10MG N05B N05BA KETAZOLAM N05BA10 514527 LOFTRAN CAP 30MG N05B N05BA BROMAZEPAM N05BA08 518123 LECTOPAM TAB 3MG N05B N05BA BROMAZEPAM N05BA08 518131 LECTOPAM TAB 6MG N05B N05BA CHLORDIAZEPOXIDE N05BA02 522724 CHLORDIAZEPOXIDE N05B N05BA CHLORDIAZEPOXIDE N05BA02 522988 CHLORDIAZEPOXIDE N05B N05BA CHLORDIAZEPOXIDE N05BA02 522996 CHLORDIAZEPOXIDE N05B N05BA ALPRAZOLAM N05BA12 548359 XANAX TAB 0.25MG N05B N05BA ALPRAZOLAM N05BA12 548367 XANAX TAB 0.5MG    165  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA LORAZEPAM N05BA06 557757 ATIVAN SUBL TAB 1MG N05B N05BA LORAZEPAM N05BA06 557765 ATIVAN TAB 2MG SUBL N05B N05BA LORAZEPAM N05BA06 557773 ATIVAN INJ 4MG/ML N05B N05BB HYDROXYZINE N05BB01 557900 MULTIPAX CAP 50MG N05B N05BB HYDROXYZINE N05BB01 557919 MULTIPAX CAP 25MG N05B N05BB HYDROXYZINE N05BB01 557927 MULTIPAX CAP 10MG N05B N05BA OXAZEPAM N05BA04 568392 RIVA OXAZEPAM N05B N05BA OXAZEPAM N05BA04 568406 RIVA OXAZEPAM N05B N05BA OXAZEPAM N05BA04 568414 RIVA OXAZEPAM N05B N05BA DIAZEPAM N05BA01 602825 DIAZEMULS EML 5MG/ML INJ N05B N05BE BUSPIRONE N05BE01 603813 BUSPAR 5MG TABLET N05B N05BE BUSPIRONE N05BE01 603821 BUSPAR TAB 10MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 628190 NOVO-CLOPATE CAP 3.75MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 628204 NOVO-CLOPATE CAP 7.5MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 628212 NOVO-CLOPATE CAP 15MG N05B N05BA LORAZEPAM N05BA06 637742 TEVA-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 637750 TEVA-LORAZEPAM N05B N05BB HYDROXYZINE N05BB01 646016 APO HYDROXYZINE CAP 50MG N05B N05BB HYDROXYZINE N05BB01 646024 APO HYDROXYZINE CAP 25MG N05B N05BB HYDROXYZINE N05BB01 646059 APO HYDROXYZINE CAP 10MG N05B N05BA LORAZEPAM N05BA06 655643 PRO-LORAZEPAM TAB 0.5MG N05B N05BA LORAZEPAM N05BA06 655651 PRO-LORAZEPAM TAB 1.0MG N05B N05BA LORAZEPAM N05BA06 655678 PRO-LORAZEPAM TAB 2MG N05B N05BA LORAZEPAM N05BA06 655740 APO-LORAZEPAM TAB 0.5MG N05B N05BA LORAZEPAM N05BA06 655759 APO-LORAZEPAM TAB 1MG N05B N05BA LORAZEPAM N05BA06 655767 APO-LORAZEPAM TAB 2MG N05B N05BA ALPRAZOLAM N05BA12 677477 RATIO-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 677485 RATIO-ALPRAZOLAM N05B N05BA BROMAZEPAM N05BA08 682314 LECTOPAM TAB 1.5MG N05B N05BA LORAZEPAM N05BA06 711101 TEVA-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 722138 ATIVAN SUBL TAB 0.5MG N05B N05BB HYDROXYZINE N05BB01 723479 MULTIPAX CAP 25MG N05B N05BB HYDROXYZINE N05BB01 723487 MULTIPAX CAP 10MG N05B N05BB HYDROXYZINE N05BB01 723592 MULTIPAX CAP 50MG N05B N05BA ALPRAZOLAM N05BA12 723770 XANAX TAB 1MG    166  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA OXAZEPAM N05BA04 726362 PMS-OXAZEPAM TAB 10MG N05B N05BA OXAZEPAM N05BA04 726370 PMS-OXAZEPAM TAB 15MG N05B N05BA OXAZEPAM N05BA04 726389 PMS-OXAZEPAM TAB 30MG N05B N05BA LORAZEPAM N05BA06 728187 PMS-LORAZEPAM TAB 0.5MG N05B N05BA LORAZEPAM N05BA06 728195 PMS-LORAZEPAM TAB 1MG N05B N05BA LORAZEPAM N05BA06 728209 PMS-LORAZEPAM TAB 2MG N05B N05BB HYDROXYZINE N05BB01 738824 NOVO-HYDROXYZIN CAP 10MG N05B N05BB HYDROXYZINE N05BB01 738832 NOVO-HYDROXYZIN CAP 25MG N05B N05BB HYDROXYZINE N05BB01 738840 NOVO-HYDROXYZIN CAP 50MG N05B N05BB HYDROXYZINE N05BB01 739618 HYDROXYZINE-10 N05B N05BB HYDROXYZINE N05BB01 739626 HYDROXYZINE-25 N05B N05BB HYDROXYZINE N05BB01 739634 HYDROXYZINE-50 N05B N05BB HYDROXYZINE N05BB01 741817 PMS HYDROXYZINE SYR 10MG/5ML N05B N05BB HYDROXYZINE N05BB01 741884 PMS HYDROXYZINE CAP 10MG N05B N05BB HYDROXYZINE N05BB01 741892 PMS HYDROXYZINE CAP 25MG N05B N05BB HYDROXYZINE N05BB01 741906 PMS HYDROXYZINE CAP 50MG N05B N05BB HYDROXYZINE N05BB01 742813 HYDROXYZINE HYDROCHLORIDE INJECTION USP N05B N05BA ALPRAZOLAM N05BA12 813958 XANAX TS TAB 2MG N05B N05BA CLOBAZAM N05BA09 846392 FRISIUM TAB 10MG N05B N05BA POTASSIUM CLORAZEPATE N05BA05 860689 CLORAZEPATE N05B N05BA POTASSIUM CLORAZEPATE N05BA05 860697 CLORAZEPATE N05B N05BA POTASSIUM CLORAZEPATE N05BA05 860700 CLORAZEPATE N05B N05BA ALPRAZOLAM N05BA12 865397 APO-ALPRAZ TAB 0.25MG N05B N05BA ALPRAZOLAM N05BA12 865400 APO-ALPRAZ TAB 0.5MG N05B N05BA LORAZEPAM N05BA06 865672 NU-LORAZ TAB 0.5MG N05B N05BA LORAZEPAM N05BA06 865680 NU-LORAZ TAB 1MG N05B N05BA LORAZEPAM N05BA06 865699 NU-LORAZ TAB 2MG N05B N05BA DIAZEPAM N05BA01 891797 PMS-DIAZEPAM SOLUTION 1MG/ML N05B N05BA ALPRAZOLAM N05BA12 1908170 ALPRAZOLAM-0.5 TAB 0.5MG    167  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA ALPRAZOLAM N05BA12 1908189 ALPRAZOLAM-0.25 TAB 0.25MG N05B N05BA ALPRAZOLAM N05BA12 1913239 NU-ALPRAZ TAB 0.25MG N05B N05BA ALPRAZOLAM N05BA12 1913247 NU-ALPRAZ TAB 0.5MG N05B N05BA ALPRAZOLAM N05BA12 1913484 TEVA-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 1913492 TEVA-ALPRAZOLAM N05B N05BA KETAZOLAM N05BA10 1919520 LOFTRAN CAP 30MG N05B N05BB HYDROXYZINE N05BB01 1927876 MULTIPAX CAP 10MG N05B N05BB HYDROXYZINE N05BB01 1927884 MULTIPAX CAP 50MG N05B N05BB HYDROXYZINE N05BB01 1938835 MULTIPAX CAP 25MG N05B N05BA CLOBAZAM N05BA09 1989634 FRISIUM TABLETS 10MG N05B N05BA DIAZEPAM N05BA01 2005492 DIAZEMULS 5MG/ML AMPOULE N05B N05BA ALPRAZOLAM N05BA12 2018179 RATIO-ALPRAZOLAM 1MG TAB N05B N05BA ALPRAZOLAM N05BA12 2018187 RATIO-ALPRAZOLAM 2MG TABLET N05B N05BA LORAZEPAM N05BA06 2041405 ATIVAN INJECTION LIQ 4MG/ML N05B N05BA LORAZEPAM N05BA06 2041413 ATIVAN N05B N05BA LORAZEPAM N05BA06 2041421 ATIVAN N05B N05BA LORAZEPAM N05BA06 2041448 ATIVAN N05B N05BA LORAZEPAM N05BA06 2041456 ATIVAN N05B N05BA LORAZEPAM N05BA06 2041464 ATIVAN N05B N05BA LORAZEPAM N05BA06 2041472 ATIVAN N05B N05BC MEPROBAMATE N05BC01 2041812 EQUANIL TABLETS 400MG N05B N05BA OXAZEPAM N05BA04 2043653 SERAX TABLETS 10MG N05B N05BA OXAZEPAM N05BA04 2043661 SERAX TABLETS 15MG N05B N05BA OXAZEPAM N05BA04 2043688 SERAX TABLETS 30MG N05B N05BA DIAZEPAM N05BA01 2065614 DIAZEMULS N05B N05BA ALPRAZOLAM N05BA12 2083418 TARO-ALPRAZOLAM 0.25MG TAB N05B N05BA ALPRAZOLAM N05BA12 2083426 TARO-ALPRAZOLAM 0.5MG TAB N05B N05BE BUSPIRONE N05BE01 2084201 SYN-BUSPIRONE 10MG TABLET N05B N05BA DIAZEPAM N05BA01 2137399 DOM-DIAZEPAM 1MG/ML ORAL LQ N05B N05BA ALPRAZOLAM N05BA12 2137534 MYLAN-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 2137542 MYLAN-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 2147572 ALPRAZOLAM 0.25MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2147580 ALPRAZOLAM 0.5MG TABLET N05B N05BB HYDROXYZINE N05BB01 2166313 NU-HYDROXYZINE    168  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BB HYDROXYZINE N05BB01 2166321 NU-HYDROXYZINE N05B N05BB HYDROXYZINE N05BB01 2166348 NU-HYDROXYZINE N05B N05BA BROMAZEPAM N05BA08 2167808 ALTI-BROMAZEPAM 1.5MG TABLETS N05B N05BA BROMAZEPAM N05BA08 2167816 ALTI-BROMAZEPAM 3MG TABLETS N05B N05BA BROMAZEPAM N05BA08 2167824 ALTI-BROMAZEPAM 6MG TABLETS N05B N05BA BROMAZEPAM N05BA08 2171856 NU-BROMAZEPAM - TAB 1.5MG N05B N05BA BROMAZEPAM N05BA08 2171864 NU-BROMAZEPAM - TAB 3MG N05B N05BA BROMAZEPAM N05BA08 2171872 NU-BROMAZEPAM - TAB 6MG N05B N05BE BUSPIRONE N05BE01 2176114 LINBUSPIRONE 5MG TABLET N05B N05BE BUSPIRONE N05BE01 2176122 LINBUSPIRONE N05B N05BA BROMAZEPAM N05BA08 2177153 APO-BROMAZEPAM - TAB 1.5MG N05B N05BA BROMAZEPAM N05BA08 2177161 APO-BROMAZEPAM - TAB 3MG N05B N05BA BROMAZEPAM N05BA08 2177188 APO-BROMAZEPAM - TAB 6MG N05B N05BA BROMAZEPAM N05BA08 2192705 MYLAN-BROMAZEPAM N05B N05BA BROMAZEPAM N05BA08 2192713 GEN-BROMAZEPAM - TAB 3MG N05B N05BA BROMAZEPAM N05BA08 2192721 GEN-BROMAZEPAM - TAB 6MG N05B N05BE BUSPIRONE N05BE01 2207664 NU-BUSPIRONE 5MG TABLET N05B N05BE BUSPIRONE N05BE01 2207672 NU-BUSPIRONE - TAB 10MG N05B N05BE BUSPIRONE N05BE01 2211068 APO-BUSPIRONE 5 MG TABLET N05B N05BE BUSPIRONE N05BE01 2211076 APO-BUSPIRONE - TAB 10MG N05B N05BA BROMAZEPAM N05BA08 2220512 BROMAZEPAM-1.5 - TAB 1.5MG N05B N05BA BROMAZEPAM N05BA08 2220520 BROMAZEPAM-3 - TAB 3MG N05B N05BA BROMAZEPAM N05BA08 2220539 BROMAZEPAM-6 - TAB 6MG N05B N05BA CLOBAZAM N05BA09 2221799 FRISIUM TABLET 10MG N05B N05BE BUSPIRONE N05BE01 2223155 BUSPIRONE 5MG TABLET N05B N05BE BUSPIRONE N05BE01 2223163 BUSPIRONE-10 - TAB 10MG N05B N05BA ALPRAZOLAM N05BA12 2228858 TRIA-ALPRAZOLAM 0.25MG TAB N05B N05BA ALPRAZOLAM N05BA12 2228866 TRIA-ALPRAZOLAM 0.5MG TAB    169  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA BROMAZEPAM N05BA08 2228874 TRIMAZEPAM 1.5MG TABLET N05B N05BA BROMAZEPAM N05BA08 2228882 TRIMAZEPAM 3MG TABLET N05B N05BA BROMAZEPAM N05BA08 2228890 TRIMAZEPAM 6MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2229813 MYLAN-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 2229814 MYLAN-ALPRAZOLAM N05B N05BA BROMAZEPAM N05BA08 2230039 BROMAZEPAM 1.5MG TABLET N05B N05BA BROMAZEPAM N05BA08 2230040 BROMAZEPAM 3MG TABLET N05B N05BA BROMAZEPAM N05BA08 2230041 BROMAZEPAM 6MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2230074 ALPRAZOLAM TABLETS 0.25 N05B N05BA ALPRAZOLAM N05BA12 2230075 ALPRAZOLAM TABLETS 0.5MG N05B N05BA BROMAZEPAM N05BA08 2230584 NOVO-BROMAZEPAM N05B N05BA BROMAZEPAM N05BA08 2230585 NOVO-BROMAZEPAM N05B N05BA BROMAZEPAM N05BA08 2230666 MED BROMAZEPAM 1.5MG TABLET N05B N05BA BROMAZEPAM N05BA08 2230667 MED BROMAZEPAM 3MG TABLET N05B N05BA BROMAZEPAM N05BA08 2230668 MED BROMAZEPAM 6MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2230744 ALPRAZOLAM 1MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2230745 ALPRAZOLAM 2MG TABLET N05B N05BE BUSPIRONE N05BE01 2230874 MYLAN-BUSPIRONE N05B N05BE BUSPIRONE N05BE01 2230941 PMS-BUSPIRONE N05B N05BE BUSPIRONE N05BE01 2230942 PMS-BUSPIRONE N05B N05BE BUSPIRONE N05BE01 2231000 BUSPIRONE 10MG TABLET N05B N05BE BUSPIRONE N05BE01 2231034 BUSTAB N05B N05BE BUSPIRONE N05BE01 2231035 BUSTAB N05B N05BB HYDROXYZINE N05BB01 2231209 HYDROXYZINE HYDROCHLORIDE INJECTION, USP N05B N05BE BUSPIRONE N05BE01 2231492 NOVO-BUSPIRONE N05B N05BA BROMAZEPAM N05BA08 2232556 PENTA-BROMAZEPAM TABLETS N05B N05BA BROMAZEPAM N05BA08 2232558 PENTA-BROMAZEPAM 6MG TABLET N05B N05BE BUSPIRONE N05BE01 2232564 DOM-BUSPIRONE N05B N05BE BUSPIRONE N05BE01 2237257 BUSPIRONE 10MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2237264 MED ALPRAZOLAM 0.25MG TAB    170  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA ALPRAZOLAM N05BA12 2237265 MED ALPRAZOLAM 0.5MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2237266 MED ALPRAZOLAM 1MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2237267 MED ALPRAZOLAM 2MG TABLET N05B N05BE BUSPIRONE N05BE01 2237268 MED BUSPIRONE 10MG TABLET N05B N05BE BUSPIRONE N05BE01 2237858 RATIO-BUSPIRONE N05B N05BA DIAZEPAM N05BA01 2238162 DIASTAT N05B N05BA CLOBAZAM N05BA09 2238334 NOVO-CLOBAZAM N05B N05BE BUSPIRONE N05BE01 2238447 FTP-BUSPIRONE N05B N05BE BUSPIRONE N05BE01 2238613 PENTA-BUSPIRONE 10MG TABLET N05B N05BA CLOBAZAM N05BA09 2238797 RATIO-CLOBAZAM N05B N05BA LORAZEPAM N05BA06 2240725 RIVA-LORAZEPAM 0.5 MG TABLETS N05B N05BA LORAZEPAM N05BA06 2240726 RIVA-LORAZEPAM 1 MG TABLETS N05B N05BA LORAZEPAM N05BA06 2240727 RIVA-LORAZEPAM 2MG TABLETS N05B N05BB HYDROXYZINE N05BB01 2241192 RIVA-HYDROXYZIN 10MG CAPSULES N05B N05BB HYDROXYZINE N05BB01 2241193 RIVA-HYDROXYZIN 25MG CAPSULES N05B N05BB HYDROXYZINE N05BB01 2241194 RIVA-HYDROXYZIN 50MG CAPSULES N05B N05BA ALPRAZOLAM N05BA12 2242107 RIVA-ALPRAZOLAM 0.25MG TAB N05B N05BA ALPRAZOLAM N05BA12 2242108 RIVA-ALPRAZOLAM 0.5MG TAB N05B N05BE BUSPIRONE N05BE01 2242149 RIVA-BUSPIRONE 10 MG N05B N05BA BROMAZEPAM N05BA08 2242152 RIVA-BROMAZEPAM 3MG TABLET N05B N05BA BROMAZEPAM N05BA08 2242153 RIVA-BROMAZEPAM 6MG TABLET N05B N05BB HYDROXYZINE N05BB01 2242864 HYDROXYZINE HYDROCHLORIDE INJECTION, USP N05B N05BA DIAZEPAM N05BA01 2243240 DIAZEPAM AUTOINJECTOR N05B N05BA LORAZEPAM N05BA06 2243278 LORAZEPAM INJECTION USP N05B N05BA ALPRAZOLAM N05BA12 2243611 APO-ALPRAZ N05B N05BA ALPRAZOLAM N05BA12 2243612 APO-ALPRAZ TS N05B N05BA CLOBAZAM N05BA09 2244474 PMS-CLOBAZAM N05B N05BA CLOBAZAM N05BA09 2244638 APO-CLOBAZAM TABLETS N05B N05BA LORAZEPAM N05BA06 2245784 DOM-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2245785 DOM-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2245786 DOM-LORAZEPAM    171  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA DIAZEPAM N05BA01 2247173 BIO-DIAZEPAM N05B N05BA DIAZEPAM N05BA01 2247174 BIO-DIAZEPAM N05B N05BA DIAZEPAM N05BA01 2247176 BIO-DIAZEPAM N05B N05BA OXAZEPAM N05BA04 2247177 BIO-OXAZEPAM N05B N05BA OXAZEPAM N05BA04 2247178 BIO-OXAZEPAM N05B N05BA OXAZEPAM N05BA04 2247179 BIO-OXAZEPAM N05B N05BA CLOBAZAM N05BA09 2247230 DOM-CLOBAZAM N05B N05BA DIAZEPAM N05BA01 2247490 PMS-DIAZEPAM N05B N05BA DIAZEPAM N05BA01 2247491 PMS-DIAZEPAM N05B N05BA DIAZEPAM N05BA01 2247492 PMS-DIAZEPAM N05B N05BA CLOBAZAM N05BA09 2248454 CLOBAZAM-10 N05B N05BA ALPRAZOLAM N05BA12 2248706 ALPRAZOLAM-1 N05B N05BE BUSPIRONE N05BE01 2252899 GMD-BUSPIRONE 10MG TABLET N05B N05BA OXAZEPAM N05BA04 2253968 DOM-OXAZEPAM 10MG TABLET N05B N05BA OXAZEPAM N05BA04 2253976 DOM-OXAZEPAM 15MG TABLET N05B N05BA OXAZEPAM N05BA04 2253984 DOM-OXAZEPAM 30MG TABLET N05B N05BE BUSPIRONE N05BE01 2262916 CO BUSPIRONE N05B N05BA LORAZEPAM N05BA06 2298201 PHL-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2298228 PHL-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2298236 PHL-LORAZEPAM N05B N05BA ALPRAZOLAM N05BA12 2346990 NTP-ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 2347008 NTP-ALPRAZOLAM N05B N05BA CLOBAZAM N05BA09 2347598 NTP-CLOBAZAM 10 MG TABLET N05B N05BA LORAZEPAM N05BA06 2347733 NTP-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2347741 NTP-LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2347768 NTP-LORAZEPAM N05B N05BA ALPRAZOLAM N05BA12 2349191 ALPRAZOLAM N05B N05BA ALPRAZOLAM N05BA12 2349205 ALPRAZOLAM N05B N05BA LORAZEPAM N05BA06 2351072 LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2351080 LORAZEPAM N05B N05BA LORAZEPAM N05BA06 2351099 LORAZEPAM N05B N05BA DIAZEPAM N05BA01 2386143 DIAZEPAM INJECTION SDZ N05B N05BA ALPRAZOLAM N05BA12 2397021 IPG-ALPRAZOLAM 0.25 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2397048 IPG-ALPRAZOLAM 0.5 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2397056 IPG-ALPRAZOLAM 1 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2397064 IPG-ALPRAZOLAM 2 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2400111 JAMP-ALPRAZOLAM 0.25 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2400138 JAMP-ALPRAZOLAM 0.5 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2400146 JAMP-ALPRAZOLAM 1 MG TABLET    172  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N05B N05BA ALPRAZOLAM N05BA12 2400154 JAMP-ALPRAZOLAM 2 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2404877 RIVA-ALPRAZOLAM 0.25 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2404885 RIVA-ALPRAZOLAM 0.5 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2404893 RIVA-ALPRAZOLAM 1 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2404907 RIVA-ALPRAZOLAM 2 MG TABLET N05B N05BA LORAZEPAM N05BA06 2410745 APO-LORAZEPAM SUBLINGUAL N05B N05BA LORAZEPAM N05BA06 2410753 APO-LORAZEPAM SUBLINGUAL N05B N05BA LORAZEPAM N05BA06 2410761 APO-LORAZEPAM SUBLINGUAL N05B N05BA ALPRAZOLAM N05BA12 2417634 NAT-ALPRAZOLAM 0.25 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2417642 NAT-ALPRAZOLAM 0.5 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2417650 NAT-ALPRAZOLAM 1 MG TABLET N05B N05BA ALPRAZOLAM N05BA12 2417669 NAT-ALPRAZOLAM 2 MG TABLET N06A N06AA DESIPRAMINE N06AA01 10448 PERTOFRANE 25MG N06A N06AA IMIPRAMINE N06AA02 10464 TOFRANIL 10MG N06A N06AA IMIPRAMINE N06AA02 10472 TOFRANIL 25MG N06A N06AA IMIPRAMINE N06AA02 10480 TOFRANIL 50MG N06A N06AA NORTRIPTYLINE N06AA10 15229 AVENTYL N06A N06AA NORTRIPTYLINE N06AA10 15237 AVENTYL N06A N06AA AMITRIPTYLINE N06AA09 16306 ELAVIL PAMOATE SYR 10MG/5ML N06A N06AA AMITRIPTYLINE N06AA09 16322 ELAVIL TAB 10MG N06A N06AA AMITRIPTYLINE N06AA09 16330 ELAVIL TAB 25MG N06A N06AA AMITRIPTYLINE N06AA09 16349 ELAVIL TABLET 50MG N06A N06AA IMIPRAMINE N06AA02 21504 NOVO-PRAMINE TAB 10MG N06A N06AA IMIPRAMINE N06AA02 21512 NOVO-PRAMINE TAB 25MG N06A N06AA IMIPRAMINE N06AA02 21520 NOVO-PRAMINE TAB 50MG N06A N06AA DOXEPIN N06AA12 24325 SINEQUAN CAP 10MG N06A N06AA DOXEPIN N06AA12 24333 SINEQUAN CAP 25MG N06A N06AA DOXEPIN N06AA12 24341 SINEQUAN CAP 50MG N06A N06AA TRIMIPRAMINE N06AA06 25828 SURMONTIL 17.43MG N06A N06AA TRIMIPRAMINE N06AA06 25836 SURMONTIL 34.86MG N06A N06AA TRIMIPRAMINE N06AA06 25844 SURMONTIL TAB 69.72MG N06A N06AA TRIMIPRAMINE N06AA06 25852 SURMONTIL 139.44MG N06A N06AF TRANYLCYPROMINE N06AF04 27111 PARNATE TAB 10MG    173  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA AMITRIPTYLINE N06AA09 37400 NOVO-TRIPTYN TAB 10MG N06A N06AA AMITRIPTYLINE N06AA09 37419 NOVO-TRIPTYN TAB 25MG N06A N06AA AMITRIPTYLINE N06AA09 37427 NOVO-TRIPTYN TAB 50MG N06A N06AA IMIPRAMINE N06AA02 209848 IMIPRAMINE TAB 50MG N06A N06AA IMIPRAMINE N06AA02 209856 IMIPRAMINE TAB 10MG N06A N06AA IMIPRAMINE N06AA02 209864 IMIPRAMINE TAB 25MG N06A N06AA IMIPRAMINE N06AA02 236721 IMPRIL N06A N06AA IMIPRAMINE N06AA02 236748 IMPRIL N06A N06AA IMIPRAMINE N06AA02 236756 IMPRIL N06A N06AA AMITRIPTYLINE N06AA09 251275 AMITRIPTYLINE HCL TAB 25MG N06A N06AA AMITRIPTYLINE N06AA09 251283 AMITRIPTYLINE HCL TAB 10MG N06A N06AA AMITRIPTYLINE N06AA09 271152 LEVATE N06A N06AA AMITRIPTYLINE N06AA09 293911 LEVATE N06A N06AA AMITRIPTYLINE N06AA09 306320 LEVATE N06A N06AA IMIPRAMINE N06AA02 306487 TOFRANIL 75MG N06A N06AA IMIPRAMINE N06AA02 312797 IMIPRAMINE N06A N06AA PROTRIPTYLINE N06AA11 322741 TRIPTIL TAB 10MG N06A N06AA CLOMIPRAMINE N06AA04 324019 ANAFRANIL N06A N06AA IMIPRAMINE N06AA02 326852 IMIPRAMINE N06A N06AA DOXEPIN N06AA12 326925 SINEQUAN CAP 100MG N06A N06AA CLOMIPRAMINE N06AA04 330566 ANAFRANIL N06A N06AA AMITRIPTYLINE N06AA09 335053 ELAVIL N06A N06AA AMITRIPTYLINE N06AA09 335061 ELAVIL N06A N06AA AMITRIPTYLINE N06AA09 335088 ELAVIL N06A N06AA DESIPRAMINE N06AA01 353868 NORPRAMIN TAB 25MG N06A N06AA DESIPRAMINE N06AA01 353876 NORPRAMIN TAB 50MG N06A N06AA AMITRIPTYLINE N06AA09 354295 ELAVIL TABLET 75MG N06A N06AA IMIPRAMINE N06AA02 360201 IMIPRAMINE N06A N06AA MAPROTILINE N06AA21 360481 LUDIOMIL TAB 25MG N06A N06AA MAPROTILINE N06AA21 360503 LUDIOMIL TAB 50MG N06A N06AA MAPROTILINE N06AA21 360511 LUDIOMIL TAB 75MG N06A N06AA AMITRIPTYLINE N06AA09 370991 AMITRIPTYLINE-10 N06A N06AA AMITRIPTYLINE N06AA09 371009 AMITRIPTYLINE-25 N06A N06AA IMIPRAMINE N06AA02 371017 IMIPRAMINE 10TAB N06A N06AA IMIPRAMINE N06AA02 371025 IMIPRAMINE 25TAB N06A N06AA AMITRIPTYLINE N06AA09 377872 AMITRIPTYLINE HCL TAB 10MG N06A N06AA AMITRIPTYLINE N06AA09 377880 AMITRIPTYLINE HCL TAB 25MG N06A N06AA AMITRIPTYLINE N06AA09 377899 AMITRIPTYLINE HCL TAB 50MG    174  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA IMIPRAMINE N06AA02 377902 IMIPRAMINE HCL TAB 10MG N06A N06AA IMIPRAMINE N06AA02 377910 IMIPRAMINE HYDROCHLORIDE TABLETS 25MG N06A N06AA IMIPRAMINE N06AA02 377929 IMIPRAMINE HYDROCHLORIDE TABLETS 50MG N06A N06AA AMITRIPTYLINE N06AA09 398462 AMITRIPTYLINE HCL TAB 50MG N06A N06AA DOXEPIN N06AA12 400750 SINEQUAN CAP 75MG N06A N06AA CLOMIPRAMINE N06AA04 402591 ANAFRANIL N06A N06AA IMIPRAMINE N06AA02 405604 IMPRIL N06A N06AA AMITRIPTYLINE N06AA09 405612 LEVATE N06A N06AA DESIPRAMINE N06AA01 425265 NORPRAMIN TAB 75MG N06A N06AA IMIPRAMINE N06AA02 431087 IMIPRAMINE TAB 25MG N06A N06AA TRIMIPRAMINE N06AA06 442437 SURMONTIL 75MG CAP N06A N06AA AMITRIPTYLINE N06AA09 448796 AMITRIPTYLINE 10 TAB N06A N06AA AMITRIPTYLINE N06AA09 448818 AMITRIPTYLINE 25 TAB N06A N06AA AMITRIPTYLINE N06AA09 456349 AMITRIPTYLINE-50 N06A N06AA IMIPRAMINE N06AA02 456357 IMIPRAMINE 50 TAB 50MG N06A N06AF PHENELZINE N06AF03 476552 NARDIL N06A N06AA AMOXAPINE N06AA17 527084 ASENDIN TAB 25MG N06A N06AA AMOXAPINE N06AA17 527092 ASENDIN TAB 50MG N06A N06AA AMOXAPINE N06AA17 527106 ASENDIN TAB 100MG N06A N06AX TRAZODONE N06AX05 579351 DESYREL TAB 50MG N06A N06AX TRAZODONE N06AX05 579378 DESYREL TAB 100MG N06A N06AA DOXEPIN N06AA12 584274 SINEQUAN CAP 150MG N06A N06AA DOXEPIN N06AA12 629251 TRIADAPIN CAP 10MG N06A N06AA DOXEPIN N06AA12 629278 TRIADAPIN CAP 25MG N06A N06AA DOXEPIN N06AA12 629286 TRIADAPIN CAP 50MG N06A N06AA DOXEPIN N06AA12 629294 TRIADAPIN CAP 75MG N06A N06AA DOXEPIN N06AA12 629308 TRIADAPIN CAP 100MG N06A N06AA DOXEPIN N06AA12 629316 TRIADAPIN CAP 150MG N06A N06AB FLUOXETINE N06AB03 636622 PROZAC CAPSULES 20MG N06A N06AA MAPROTILINE N06AA21 641855 LUDIOMIL TAB 10MG N06A N06AA IMIPRAMINE N06AA02 644579 IMIPRAMINE    175  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA AMITRIPTYLINE N06AA09 654507 PMS-AMITRIPTYLINE TABLETS 50MG N06A N06AA AMITRIPTYLINE N06AA09 654515 PMS-AMITRIPTYLINE TABLETS 25MG N06A N06AA AMITRIPTYLINE N06AA09 654523 PMS-AMITRIPTYLINE TABLETS 10MG N06A N06AX TRYPTOPHAN N06AX02 654531 TRYPTAN TAB 1000MG N06A N06AX TRAZODONE N06AX05 702277 DESYREL DIVIDOSE TABLET 150 MG N06A N06AX TRYPTOPHAN N06AX02 718149 TRYPTAN CAP 500MG N06A N06AA IMIPRAMINE N06AA02 726303 PMS IMIPRAMINE TAB 25MG N06A N06AA IMIPRAMINE N06AA02 726311 PMS IMIPRAMINE TAB 50MG N06A N06AA IMIPRAMINE N06AA02 726397 PMS IMIPRAMINE TAB 10MG N06A N06AA TRIMIPRAMINE N06AA06 740799 TRIMIPRAMINE N06A N06AA TRIMIPRAMINE N06AA06 740802 TRIMIPRAMINE N06A N06AA TRIMIPRAMINE N06AA06 740810 TRIMIPRAMINE N06A N06AA TRIMIPRAMINE N06AA06 740829 TRIMIPRAMINE N06A N06AA AMITRIPTYLINE N06AA09 754129 ELAVIL N06A N06AA TRIMIPRAMINE N06AA06 761605 RHOTRIMINE N06A N06AA TRIMIPRAMINE N06AA06 761613 RHOTRIMINE N06A N06AA TRIMIPRAMINE N06AA06 761621 RHOTRIMINE N06A N06AA TRIMIPRAMINE N06AA06 761648 RHOTRIMINE N06A N06AA TRIMIPRAMINE N06AA06 761656 RHOTRIMINE N06A N06AA TRIMIPRAMINE N06AA06 761702 TRIMIPRAMINE TAB 12.5MG N06A N06AA TRIMIPRAMINE N06AA06 761710 TRIMIPRAMINE TAB 25MG N06A N06AA TRIMIPRAMINE N06AA06 761729 TRIMIPRAMINE TAB 50MG N06A N06AA TRIMIPRAMINE N06AA06 761737 TRIMIPRAMINE TAB 100MG N06A N06AA DESIPRAMINE N06AA01 776157 NORPRAMIN TAB 10MG N06A N06AX TRAZODONE N06AX05 824135 DESYREL DIVIDOSE TAB 300MG N06A N06AA DOXEPIN N06AA12 842745 TRIADAPIN CAP 10MG N06A N06AA DOXEPIN N06AA12 842753 TRIADAPIN CAP 25MG N06A N06AA DOXEPIN N06AA12 842761 TRIADAPIN CAP 50MG N06A N06AA DOXEPIN N06AA12 842788 TRIADAPIN CAP 75MG N06A N06AA DOXEPIN N06AA12 842796 TRIADAPIN CAP 100MG N06A N06AA DOXEPIN N06AA12 842818 TRIADAPIN CAP 150MG    176  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA DESIPRAMINE N06AA01 878782 NORPRAMIN TAB 100MG N06A N06AA DESIPRAMINE N06AA01 893765 PERTOFRANE TAB 50MG N06A N06AG MOCLOBEMIDE N06AG02 899348 MANERIX TAB 100MG N06A N06AG MOCLOBEMIDE N06AG02 899356 MANERIX N06A N06AB FLUVOXAMINE N06AB08 1911856 LUVOX TAB 50MG N06A N06AB FLUVOXAMINE N06AB08 1911872 LUVOX TAB 100MG N06A N06AA DOXEPIN N06AA12 1913425 NOVO-DOXEPIN CAP 25MG USP N06A N06AA DOXEPIN N06AA12 1913433 NOVO-DOXEPIN CAP 50MG USP N06A N06AA DOXEPIN N06AA12 1913441 NOVO-DOXEPIN CAP 75MG USP N06A N06AA DOXEPIN N06AA12 1913468 NOVO-DOXEPIN CAP 100MG USP N06A N06AA DOXEPIN N06AA12 1913476 NOVO-DOXEPIN CAP 150MG USP N06A N06AB FLUOXETINE N06AB03 1917021 PROZAC LIQ 20MG/5ML N06A N06AB FLUVOXAMINE N06AB08 1919342 LUVOX N06A N06AB FLUVOXAMINE N06AB08 1919369 LUVOX N06A N06AF TRANYLCYPROMINE N06AF04 1919598 PARNATE N06A N06AA TRIMIPRAMINE N06AA06 1926284 SURMONTIL 100 N06A N06AA TRIMIPRAMINE N06AA06 1926322 SURMONTIL N06A N06AA TRIMIPRAMINE N06AA06 1926330 SURMONTIL N06A N06AA TRIMIPRAMINE N06AA06 1926349 SURMONTIL 75 N06A N06AA TRIMIPRAMINE N06AA06 1926357 SURMONTIL 12.5 N06A N06AX TRAZODONE N06AX05 1937227 PMS TRAZODONE HCL TAB 50MG N06A N06AX TRAZODONE N06AX05 1937235 PMS TRAZODONE HCL TAB 100MG N06A N06AA TRIMIPRAMINE N06AA06 1940430 NOVO-TRIPRAMINE TAB 25MG BP N06A N06AA TRIMIPRAMINE N06AA06 1940449 NOVO-TRIPRAMINE TAB 50MG BP N06A N06AA TRIMIPRAMINE N06AA06 1940457 NOVO-TRIPRAMINE TAB 100MG BP N06A N06AB PAROXETINE N06AB05 1940473 PAXIL TAB 30MG N06A N06AB PAROXETINE N06AB05 1940481 PAXIL TAB 20MG N06A N06AA DESIPRAMINE N06AA01 1946242 PMS DESIPRAMINE HYDRO TAB 75MG N06A N06AA DESIPRAMINE N06AA01 1946250 PMS DESIPRAMINE HYDRO TAB 10MG N06A N06AA DESIPRAMINE N06AA01 1946269 PMS DESIPRAMINE HYDRO TAB 25MG N06A N06AA DESIPRAMINE N06AA01 1946277 PMS DESIPRAMINE HYDRO TAB 50MG N06A N06AA DESIPRAMINE N06AA01 1948776 RATIO-DESIPRAMINE TAB 10MG N06A N06AA DESIPRAMINE N06AA01 1948784 RATIO-DESIPRAMINE TAB 25MG    177  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA DESIPRAMINE N06AA01 1948792 RATIO-DESIPRAMINE TAB 50MG N06A N06AA DESIPRAMINE N06AA01 1948806 RATIO-DESIPRAMINE TAB 75MG N06A N06AA DESIPRAMINE N06AA01 1948814 RATIO-DESIPRAMINE TAB 100MG N06A N06AB SERTRALINE N06AB06 1962779 ZOLOFT CAP 100MG N06A N06AB SERTRALINE N06AB06 1962817 ZOLOFT CAP 50MG N06A N06AA AMITRIPTYLINE N06AA09 1985426 AMITRIPTYLINE TABLETS B.P 25MG N06A N06AB FLUOXETINE N06AB03 2018985 PROZAC CAPSULES 10MG N06A N06AA TRIMIPRAMINE N06AA06 2020599 NU-TRIMIPRAMINE TAB 12.5MG N06A N06AA TRIMIPRAMINE N06AA06 2020602 NU-TRIMIPRAMINE TAB 25MG N06A N06AA TRIMIPRAMINE N06AA06 2020610 NU-TRIMIPRAMINE TAB 50MG N06A N06AA TRIMIPRAMINE N06AA06 2020629 NU-TRIMIPRAMINE TAB 100MG N06A N06AA DESIPRAMINE N06AA01 2024888 NORPRAMIN TAB 10MG N06A N06AA DESIPRAMINE N06AA01 2024896 NORPRAMIN TAB 25MG N06A N06AA DESIPRAMINE N06AA01 2024918 NORPRAMIN TAB 50MG N06A N06AA DESIPRAMINE N06AA01 2024926 NORPRAMIN TAB 75MG N06A N06AA DESIPRAMINE N06AA01 2024934 NORPRAMIN TAB 100MG N06A N06AB PAROXETINE N06AB05 2027887 PAXIL TAB 10MG N06A N06AX TRYPTOPHAN N06AX02 2029456 TRYPTAN TAB 500MG N06A N06AA CLOMIPRAMINE N06AA04 2040751 APO-CLOMIPRAMINE TABLETS 50MG N06A N06AA CLOMIPRAMINE N06AA04 2040778 APO-CLOMIPRAMINE TABLETS 25MG N06A N06AA CLOMIPRAMINE N06AA04 2040786 APO-CLOMIPRAMINE TABLETS 10MG N06A N06AA DOXEPIN N06AA12 2049996 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050005 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050013 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050021 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050048 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050056 APO-DOXEPIN N06A N06AA DOXEPIN N06AA12 2050778 DOXEPINE 100MG CAPSULE N06A N06AX TRAZODONE N06AX05 2053187 RATIO-TRAZODONE TABLETS 50MG N06A N06AX TRAZODONE N06AX05 2053195 RATIO-TRAZODONE TABLETS 100MG N06A N06AX TRAZODONE N06AX05 2053209 RATIO-TRAZODONE TAB 150MG    178  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA TRIMIPRAMINE N06AA06 2070987 TRIMIPRAMINE N06A N06AA DOXEPIN N06AA12 2072734 NOVO-DOXEPIN 10MG CAPSULE N06A N06AA DOXEPIN N06AA12 2072742 NOVO-DOXEPIN 25MG CAPSULE N06A N06AA DOXEPIN N06AA12 2072750 NOVO-DOXEPIN 50MG CAPSULE N06A N06AA DOXEPIN N06AA12 2072769 NOVO-DOXEPIN 75MG CAPSULE N06A N06AA DOXEPIN N06AA12 2072777 NOVO-DOXEPIN 100MG CAPSULE N06A N06AX NEFAZODONE N06AX06 2087294 SERZONE-5HT2 N06A N06AX NEFAZODONE N06AX06 2087375 SERZONE-5HT2 N06A N06AX NEFAZODONE N06AX06 2087383 SERZONE-5HT2 N06A N06AX NEFAZODONE N06AX06 2087391 SERZONE-5HT2 N06A N06AA DESIPRAMINE N06AA01 2099128 NORPRAMIN 25MG TAB N06A N06AA DESIPRAMINE N06AA01 2099136 NORPRAMIN 50MG TAB N06A N06AA DESIPRAMINE N06AA01 2099144 NORPRAMIN 75MG TAB N06A N06AA DESIPRAMINE N06AA01 2103583 NORPRAMIN 10MG TAB N06A N06AA DESIPRAMINE N06AA01 2103591 NORPRAMIN 100MG TAB N06A N06AX VENLAFAXINE N06AX16 2103680 EFFEXOR TABLETS, 37.5MG N06A N06AX VENLAFAXINE N06AX16 2103699 EFFEXOR TABLETS, 50MG N06A N06AX VENLAFAXINE N06AX16 2103702 EFFEXOR TABLETS, 75MG N06A N06AA NORTRIPTYLINE N06AA10 2128829 KENRAL-NORTRIPTYLINE 10MG C N06A N06AA NORTRIPTYLINE N06AA10 2128837 KENRAL-NORTRIPTYLINE 25MG C N06A N06AX TRAZODONE N06AX05 2128950 DOM-TRAZODONE TABLETS - 50MG N06A N06AX TRAZODONE N06AX05 2128969 DOM-TRAZODONE TABLETS-100MG N06A N06AA DESIPRAMINE N06AA01 2130084 DOM-DESIPRAMINE 10MG TABLET N06A N06AA DESIPRAMINE N06AA01 2130092 DOM-DESIPRAMINE TABLETS - 25MG N06A N06AA DESIPRAMINE N06AA01 2130106 DOM-DESIPRAMINE TABLETS - 50MG N06A N06AA DESIPRAMINE N06AA01 2130114 DOM-DESIPRAMINE 75MG TABLET N06A N06AA CLOMIPRAMINE N06AA04 2130122 CLOMIPRAMINE-10 - TAB 10MG N06A N06AA CLOMIPRAMINE N06AA04 2130130 CLOMIPRAMINE-25 - TAB 25MG    179  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA CLOMIPRAMINE N06AA04 2130149 CLOMIPRAMINE-50 - TAB 50MG N06A N06AA CLOMIPRAMINE N06AA04 2130165 NOVO-CLOPAMINE TABLETS - 25MG N06A N06AA CLOMIPRAMINE N06AA04 2130173 NOVO-CLOPAMINE TABLETS- 50MG N06A N06AB SERTRALINE N06AB06 2132702 ZOLOFT N06A N06AA CLOMIPRAMINE N06AA04 2139340 GEN-CLOMIPRAMINE - TAB 10MG N06A N06AA CLOMIPRAMINE N06AA04 2139359 GEN-CLOMIPRAMINE - TAB 25MG N06A N06AA CLOMIPRAMINE N06AA04 2139367 GEN-CLOMIPRAMINE - TAB 50MG N06A N06AA DOXEPIN N06AA12 2140071 ALTI-DOXEPIN - CAP 10MG N06A N06AA DOXEPIN N06AA12 2140098 ALTI-DOXEPIN-CAP 25MG N06A N06AA DOXEPIN N06AA12 2140101 ALTI-DOXEPIN-CAP 50MG N06A N06AA DOXEPIN N06AA12 2140128 ALTI-DOXEPIN-CAP 75MG N06A N06AA DOXEPIN N06AA12 2144123 RHO-DOXEPIN-10 MG CAP N06A N06AA DOXEPIN N06AA12 2144131 RHO-DOXEPIN  - CAP 25MG N06A N06AA DOXEPIN N06AA12 2144158 RHO-DOXEPIN - CAP 50MG N06A N06AX TRAZODONE N06AX05 2144263 TEVA-TRAZODONE N06A N06AX TRAZODONE N06AX05 2144271 TEVA-TRAZODONE N06A N06AX TRAZODONE N06AX05 2144298 TEVA-TRAZODONE N06A N06AA CLOMIPRAMINE N06AA04 2147491 CLOMIPRAMINE HCL 10MG TAB N06A N06AA CLOMIPRAMINE N06AA04 2147505 CLOMIPRAMINE HCL 25MG TAB N06A N06AA CLOMIPRAMINE N06AA04 2147513 CLOMIPRAMINE HCL 50MG TAB N06A N06AA TRIMIPRAMINE N06AA06 2147599 TRIMIPRAMINE-75 - CAP 75MG N06A N06AX TRAZODONE N06AX05 2147637 APO-TRAZODONE TABLETS 50MG N06A N06AX TRAZODONE N06AX05 2147645 APO-TRAZODONE TABLETS 100MG N06A N06AX TRAZODONE N06AX05 2147653 APO-TRAZODONE D TABLETS 150MG N06A N06AA DOXEPIN N06AA12 2150727 DOXEPINE-10 - CAP N06A N06AA DOXEPIN N06AA12 2150735 DOXEPINE-25 -CAP N06A N06AA DOXEPIN N06AA12 2150743 DOXEPINE-50 - CAP N06A N06AA DOXEPIN N06AA12 2150751 DOXEPINE-75 - CAP N06A N06AA DOXEPIN N06AA12 2150778 DOXEPINE-100 - CAP N06A N06AA DOXEPIN N06AA12 2150786 DOXEPINE-150 - CAP    180  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA NORTRIPTYLINE N06AA10 2155710 STCC-NORTRIPTYLINE 10MG CAP N06A N06AA NORTRIPTYLINE N06AA10 2155729 STCC-NORTRIPTYLINE 25MG CAP N06A N06AB FLUOXETINE N06AB03 2155826 STCC-FLUOXETINE - CAP 10MG N06A N06AB FLUOXETINE N06AB03 2155834 STCC-FLUOXETINE - CAP 20MG N06A N06AB FLUOXETINE N06AB03 2155842 STCC-FLUOXETINE 20MG/5ML N06A N06AA MAPROTILINE N06AA21 2158604 NOVO-MAPROTILINE  - TAB 10MG N06A N06AA MAPROTILINE N06AA21 2158612 TEVA-MAPROTILINE N06A N06AA MAPROTILINE N06AA21 2158620 TEVA-MAPROTILINE N06A N06AA MAPROTILINE N06AA21 2158639 TEVA-MAPROTILINE N06A N06AX TRAZODONE N06AX05 2164353 TRAZODONE-50 - TAB 50MG N06A N06AX TRAZODONE N06AX05 2164361 TRAZODONE - 100 - TAB 100MG N06A N06AX TRAZODONE N06AX05 2164388 TRAZODONE-150 D - TAB 150MG N06A N06AX TRAZODONE N06AX05 2165384 NU-TRAZODONE - TAB 50MG N06A N06AX TRAZODONE N06AX05 2165392 NU-TRAZODONE - TAB 100MG N06A N06AX TRAZODONE N06AX05 2165406 NU-TRAZODONE-D - TAB 150MG N06A N06AG MOCLOBEMIDE N06AG02 2166747 MANERIX N06A N06AA DESIPRAMINE N06AA01 2168952 PMS-DESIPRAMINE - TAB 100MG N06A N06AA AMOXAPINE N06AA17 2169886 ASENDIN - TAB 25MG N06A N06AA AMOXAPINE N06AA17 2169894 ASENDIN - TAB 50MG N06A N06AA AMOXAPINE N06AA17 2169908 ASENDIN - TAB 100MG N06A N06AB FLUOXETINE N06AB03 2177579 PMS-FLUOXETINE - CAP 10MG N06A N06AB FLUOXETINE N06AB03 2177587 PMS-FLUOXETINE - CAP 20MG N06A N06AB FLUOXETINE N06AB03 2177595 PMS-FLUOXETINE LIQUID - 20MG/5ML N06A N06AB FLUOXETINE N06AB03 2177609 DOM-FLUOXETINE 20MG/5ML SOL N06A N06AB FLUOXETINE N06AB03 2177617 DOM-FLUOXETINE - CAP 10MG N06A N06AB FLUOXETINE N06AB03 2177625 DOM-FLUOXETINE N06A N06AA NORTRIPTYLINE N06AA10 2177692 PMS-NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2177706 PMS-NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2178729 DOM-NORTRIPTYLINE 10MG CAPSULES    181  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA NORTRIPTYLINE N06AA10 2178737 DOM-NORTRIPTYLINE 25MG CAPSULES N06A N06AA CLOMIPRAMINE N06AA04 2189003 MED CLOMIPRAMINE TABLETS - 25MG N06A N06AA CLOMIPRAMINE N06AA04 2189011 MED CLOMIPRAMINE TABLETS - 50MG N06A N06AB FLUOXETINE N06AB03 2192756 NU-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2192764 NU-FLUOXETINE N06A N06AB FLUVOXAMINE N06AB08 2204460 SYN-FLUVOXAMINE 50MG TAB N06A N06AB FLUVOXAMINE N06AB08 2204479 SYN-FLUVOXAMINE 100MG TAB N06A N06AA DESIPRAMINE N06AA01 2211939 NU-DESIPRAMINE - TAB 10MG N06A N06AA DESIPRAMINE N06AA01 2211947 NU-DESIPRAMINE - TAB 25MG N06A N06AA DESIPRAMINE N06AA01 2211955 NU-DESIPRAMINE - TAB 50MG N06A N06AA DESIPRAMINE N06AA01 2211963 NU-DESIPRAMINE - TAB 75MG N06A N06AA DESIPRAMINE N06AA01 2211971 NU-DESIPRAMINE - TAB  100MG N06A N06AB SERTRALINE N06AB06 2212787 RATIO-SERTRALINE 25MG CAP N06A N06AA DESIPRAMINE N06AA01 2216248 DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2216256 DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2216264 DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2216272 DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2216280 DESIPRAMINE N06A N06AB FLUOXETINE N06AB03 2216353 APO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2216361 APO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2216582 TEVA-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2216590 TEVA-FLUOXETINE N06A N06AG MOCLOBEMIDE N06AG02 2218402 ALTI-MOCLOBEMIDE 100MG TAB N06A N06AG MOCLOBEMIDE N06AG02 2218410 RATIO-MOCLOBEMIDE - TAB 150MG N06A N06AG MOCLOBEMIDE N06AG02 2218429 RATIO-MOCLOBEMIDE - TAB 300MG N06A N06AB FLUVOXAMINE N06AB08 2218453 RATIO-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2218461 RATIO-FLUVOXAMINE N06A N06AB FLUOXETINE N06AB03 2220121 FLUOXETINE-10 - CAP N06A N06AB FLUOXETINE N06AB03 2220148 FLUOXETINE-20 - CAP N06A N06AA DESIPRAMINE N06AA01 2222981 DESIPRAMINE-10 - TAB 10MG N06A N06AA DESIPRAMINE N06AA01 2223007 DESIPRAMINE-25 - TAB 25MG    182  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA DESIPRAMINE N06AA01 2223015 DESIPRAMINE-50 - TAB 50MG N06A N06AA DESIPRAMINE N06AA01 2223023 DESIPRAMINE-75 - TAB 75MG N06A N06AA DESIPRAMINE N06AA01 2223031 DESIPRAMINE 100MG TABLET N06A N06AA NORTRIPTYLINE N06AA10 2223139 NU-NORTRIPTYLINE - CAP 10MG N06A N06AA NORTRIPTYLINE N06AA10 2223147 NU-NORTRIPTYLINE - CAP 25MG N06A N06AB FLUOXETINE N06AB03 2223181 FLUOXETINE 10MG CAPSULE N06A N06AA DESIPRAMINE N06AA01 2223325 NOVO-DESIPRAMINE SC - TAB 25MG N06A N06AA DESIPRAMINE N06AA01 2223333 NOVO-DESIPRAMINE FC - 50MG TAB N06A N06AA DESIPRAMINE N06AA01 2223341 NOVO-DESIPRAMINE SC - TAB 10MG N06A N06AA DESIPRAMINE N06AA01 2223368 NOVO-DESIPRAMINE SC -TAB 75MG N06A N06AB FLUOXETINE N06AB03 2223481 PHL-FLUOXETINE - CAP 10MG N06A N06AB FLUOXETINE N06AB03 2223503 PHL-FLUOXETINE - CAP 20MG N06A N06AA NORTRIPTYLINE N06AA10 2223511 APO-NORTRIPTYLINE - CAP 10MG N06A N06AA NORTRIPTYLINE N06AA10 2223538 APO-NORTRIPTYLINE - CAP 25MG N06A N06AB FLUOXETINE N06AB03 2225174 PHL-FLUOXETINE - SOLUTION 20MG/5ML N06A N06AA CLOMIPRAMINE N06AA04 2229589 PENTA-CLOMIPRAMINE - 25MG N06A N06AA CLOMIPRAMINE N06AA04 2229590 PENTA-CLOMIPRAMINE - 50MG N06A N06AA NORTRIPTYLINE N06AA10 2229763 NORTRIPTYLINE-10 N06A N06AA NORTRIPTYLINE N06AA10 2229764 NORTRIPTYLINE-25 N06A N06AB FLUOXETINE N06AB03 2229819 PENTA-FLUOXETINE 10MG CAP N06A N06AB FLUOXETINE N06AB03 2229820 PENTA-FLUOXETINE 20MG CAP N06A N06AA CLOMIPRAMINE N06AA04 2230063 CLOMIPRAMINE TABLETS 10MG N06A N06AA CLOMIPRAMINE N06AA04 2230064 CLOMIPRAMINE TABLETS 25MG N06A N06AA CLOMIPRAMINE N06AA04 2230065 CLOMIPRAMINE TABLETS 50MG N06A N06AX TRYPTOPHAN N06AX02 2230202 PMS-TRYPTOPHAN N06A N06AA CLOMIPRAMINE N06AA04 2230256 NOVO-CLOPAMINE TAB 10MG N06A N06AX TRAZODONE N06AX05 2230284 TRAZOREL N06A N06AX TRAZODONE N06AX05 2230285 TRAZOREL    183  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX TRAZODONE N06AX05 2230286 TRAZOREL N06A N06AA NORTRIPTYLINE N06AA10 2230361 NORVENTYL N06A N06AA NORTRIPTYLINE N06AA10 2230362 NORVENTYL N06A N06AB FLUOXETINE N06AB03 2231174 NU-FLUOXETINE 20MG/5ML SOLN N06A N06AB FLUVOXAMINE N06AB08 2231192 NU-FLUVOXAMINE 50 MG TAB N06A N06AB FLUVOXAMINE N06AB08 2231193 NU-FLUVOXAMINE 100 MG TAB N06A N06AX TRYPTOPHAN N06AX02 2231255 DOM-TRYPTOPHAN 1GM TABLET N06A N06AB FLUOXETINE N06AB03 2231328 APO-FLUOXETINE ORAL SOLUTION N06A N06AB FLUVOXAMINE N06AB08 2231329 APO-FLUVOXAMINE TABLETS N06A N06AB FLUVOXAMINE N06AB08 2231330 APO-FLUVOXAMINE TABLETS N06A N06AA MAPROTILINE N06AA21 2231559 PMS-MAPROTILINE 10MG TABLET N06A N06AA MAPROTILINE N06AA21 2231560 PMS-MAPROTILINE 25MG TABLET N06A N06AA MAPROTILINE N06AA21 2231561 PMS-MAPROTILINE 50MG TABLET N06A N06AA MAPROTILINE N06AA21 2231562 PMS-MAPROTILINE 75MG TABLET N06A N06AA CLOMIPRAMINE N06AA04 2231608 PMS-CLOMIPRAMINE 10 MG TAB N06A N06AA CLOMIPRAMINE N06AA04 2231610 PMS-CLOMIPRAMINE 25MG TAB N06A N06AA CLOMIPRAMINE N06AA04 2231611 PMS-CLOMIPRAMINE 50MG TAB N06A N06AA CLOMIPRAMINE N06AA04 2231667 DOM-CLOMIPRAMINE N06A N06AA CLOMIPRAMINE N06AA04 2231669 DOM-CLOMIPRAMINE N06A N06AX TRAZODONE N06AX05 2231683 MYLAN-TRAZODONE N06A N06AX TRAZODONE N06AX05 2231684 MYLAN-TRAZODONE N06A N06AX TRAZODONE N06AX05 2231685 MYLAN-TRAZODONE N06A N06AA NORTRIPTYLINE N06AA10 2231686 GEN-NORTRIPTYLINE CAPSULES 10MG N06A N06AA NORTRIPTYLINE N06AA10 2231687 GEN-NORTRIPTYLINE CAPSULES 25MG N06A N06AA NORTRIPTYLINE N06AA10 2231781 TEVA-NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2231782 TEVA-NORTRIPTYLINE N06A N06AG MOCLOBEMIDE N06AG02 2232148 APO-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2232150 APO-MOCLOBEMIDE N06A N06AX TRAZODONE N06AX05 2232543 PENTA-TRAZODONE 50MG TABLET N06A N06AX TRAZODONE N06AX05 2232544 PENTA-TRAZODONE 100MG TAB N06A N06AX TRAZODONE N06AX05 2232545 PENTA-TRAZODONE 150MG TAB    184  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA DESIPRAMINE N06AA01 2232559 PENTA-DESIPRAMINE 10MG TAB N06A N06AA DESIPRAMINE N06AA01 2232561 PENTA-DESIPRAMINE 25MG TAB N06A N06AA DESIPRAMINE N06AA01 2232562 PENTA-DESIPRAMINE 50MG TAB N06A N06AA DESIPRAMINE N06AA01 2232563 PENTA-DESIPRAMINE 75MG TAB N06A N06AB FLUVOXAMINE N06AB08 2236753 FLUVOXAMINE-50 N06A N06AB FLUVOXAMINE N06AB08 2236754 FLUVOXAMINE-100 N06A N06AG MOCLOBEMIDE N06AG02 2236928 MOCLOBEMIDE-100 N06A N06AG MOCLOBEMIDE N06AG02 2236929 MOCLOBEMIDE-150 N06A N06AA DESIPRAMINE N06AA01 2236936 PHL-DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2236937 PHL-DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2236938 PHL-DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2236939 PHL-DESIPRAMINE N06A N06AA DESIPRAMINE N06AA01 2236940 DESIPRAMINE 100MG TABLET N06A N06AX TRAZODONE N06AX05 2236941 PHL-TRAZODONE N06A N06AX TRAZODONE N06AX05 2236942 PHL-TRAZODONE N06A N06AX TRYPTOPHAN N06AX02 2236957 PHL-TRYPTOPHAN N06A N06AB FLUOXETINE N06AB03 2237005 FLUOXETINE 20MG/5ML SYRUP N06A N06AG MOCLOBEMIDE N06AG02 2237111 NU-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2237112 NU-MOCLOBEMIDE N06A N06AX TRYPTOPHAN N06AX02 2237250 TEVA-TRYPTOPHAN N06A N06AX VENLAFAXINE N06AX16 2237279 EFFEXOR XR N06A N06AX VENLAFAXINE N06AX16 2237280 EFFEXOR XR N06A N06AX VENLAFAXINE N06AX16 2237282 EFFEXOR XR N06A N06AX TRAZODONE N06AX05 2237339 PMS-TRAZODONE N06A N06AA NORTRIPTYLINE N06AA10 2237376 NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2237377 NORTRIPTYLINE N06A N06AX NEFAZODONE N06AX06 2237397 LIN-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2237398 LIN-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2237399 LIN-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2237400 LIN-NEFAZODONE N06A N06AB FLUOXETINE N06AB03 2237813 MYLAN-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2237814 MYLAN-FLUOXETINE N06A N06AB SERTRALINE N06AB06 2238280 APO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2238281 APO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2238282 APO-SERTRALINE N06A N06AX TRYPTOPHAN N06AX02 2239326 TRYPTAN TAB 250MG N06A N06AX TRYPTOPHAN N06AX02 2239327 TRYPTAN TAB 750MG N06A N06AB CITALOPRAM N06AB04 2239606 CELEXA 10MG TABLET N06A N06AB CITALOPRAM N06AB04 2239607 CELEXA 20 MG N06A N06AB CITALOPRAM N06AB04 2239608 CELEXA 40 MG N06A N06AG MOCLOBEMIDE N06AG02 2239746 TEVA-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2239747 TEVA-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2239748 TEVA-MOCLOBEMIDE N06A N06AB FLUOXETINE N06AB03 2239751 MED FLUOXETINE 10MG CAPSULE    185  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB FLUOXETINE N06AB03 2239752 MED FLUOXETINE 20MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2239817 FLUOXETINE CAPSULES N06A N06AB FLUOXETINE N06AB03 2239818 FLUOXETINE CAPSULES N06A N06AB FLUVOXAMINE N06AB08 2239953 NOVO-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2239954 NOVO-FLUVOXAMINE N06A N06AX TRYPTOPHAN N06AX02 2240333 TEVA-TRYPTOPHAN N06A N06AX TRYPTOPHAN N06AX02 2240334 TEVA-TRYPTOPHAN N06A N06AX TRYPTOPHAN N06AX02 2240445 PMS-TRYPTOPHAN N06A N06AG MOCLOBEMIDE N06AG02 2240456 APO-MOCLOBEMIDE N06A N06AB SERTRALINE N06AB06 2240481 TEVA-SERTRALINE N06A N06AB SERTRALINE N06AB06 2240484 TEVA-SERTRALINE N06A N06AB SERTRALINE N06AB06 2240485 TEVA-SERTRALINE N06A N06AB FLUVOXAMINE N06AB08 2240682 PMS-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2240683 PMS-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2240723 RIVA-FLUVOX 50MG TABLETS N06A N06AB FLUVOXAMINE N06AB08 2240724 RIVA-FLUVOX 100MG TABLETS N06A N06AG MOCLOBEMIDE N06AG02 2240736 RIVA-MOCLOBEMIDE 150MG TABLETS N06A N06AG MOCLOBEMIDE N06AG02 2240737 RIVA-MOCLOBEMIDE 300MG TABLETS N06A N06AA NORTRIPTYLINE N06AA10 2240789 RATIO-NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2240790 RATIO-NORTRIPTYLINE N06A N06AB FLUVOXAMINE N06AB08 2240849 GEN-FLUVOXAMINE 50MG N06A N06AB FLUVOXAMINE N06AB08 2240850 GEN-FLUVOXAMINE 100MG N06A N06AB PAROXETINE N06AB05 2240907 APO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2240908 APO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2240909 APO-PAROXETINE N06A N06AG MOCLOBEMIDE N06AG02 2240970 MOCLOBEMIDE 300MG TABLET N06A N06AX TRYPTOPHAN N06AX02 2241023 PMS-TRYPTOPHAN N06A N06AB SERTRALINE N06AB06 2241302 SERTRALINE-25 N06A N06AB SERTRALINE N06AB06 2241303 SERTRALINE-50 N06A N06AB SERTRALINE N06AB06 2241304 SERTRALINE-100 N06A N06AB FLUVOXAMINE N06AB08 2241347 DOM-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2241348 DOM-FLUVOXAMINE N06A N06AB FLUOXETINE N06AB03 2241371 RATIO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2241374 RATIO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2242123 RIVA-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2242124 RIVA-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2242177 CO FLUOXETINE 10MG N06A N06AB FLUOXETINE N06AB03 2242178 CO FLUOXETINE 20MG    186  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX TRAZODONE N06AX05 2242391 SCHEINPHARM TRAZODONE N06A N06AX TRAZODONE N06AX05 2242392 SCHEINPHARM TRAZODONE N06A N06AB SERTRALINE N06AB06 2242519 MYLAN-SERTRALINE N06A N06AB SERTRALINE N06AB06 2242520 MYLAN-SERTRALINE N06A N06AB SERTRALINE N06AB06 2242521 MYLAN-SERTRALINE N06A N06AB FLUOXETINE N06AB03 2242820 ALTI-FLUOXETINE 20MG/5ML N06A N06AX NEFAZODONE N06AX06 2242822 APO-NEFAZODONE TABLETS N06A N06AX NEFAZODONE N06AX06 2242823 APO-NEFAZODONE TABLETS N06A N06AX NEFAZODONE N06AX06 2242824 APO-NEFAZODONE TABLETS N06A N06AX NEFAZODONE N06AX06 2242825 APO-NEFAZODONE TABLETS N06A N06AG MOCLOBEMIDE N06AG02 2243218 PMS-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2243219 PMS-MOCLOBEMIDE N06A N06AG MOCLOBEMIDE N06AG02 2243348 DOM-MOCLOBEMIDE TABLETS 150MG N06A N06AG MOCLOBEMIDE N06AG02 2243349 DOM-MOCLOBEMIDE TABLETS 300MG N06A N06AB FLUOXETINE N06AB03 2243486 SANDOZ FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2243487 SANDOZ FLUOXETINE N06A N06AX NEFAZODONE N06AX06 2243791 BIONEFAZODONE 50MG TABLET N06A N06AX NEFAZODONE N06AX06 2243792 BIONEFAZODONE 100MG TABLET N06A N06AX NEFAZODONE N06AX06 2243793 BIONEFAZODONE 150MG TABLET N06A N06AX NEFAZODONE N06AX06 2243794 BIONEFAZODONE 200MG TABLET N06A N06AB FLUOXETINE N06AB03 2243833 PMS-FLUOX 20MG CAPSULE N06A N06AX MIRTAZAPINE N06AX11 2243909 REMERON 15 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2243910 REMERON N06A N06AX MIRTAZAPINE N06AX11 2243911 REMERON 45 MG TABLET N06A N06AA CLOMIPRAMINE N06AA04 2244816 CO CLOMIPRAMINE TABLETS 10MG N06A N06AA CLOMIPRAMINE N06AA04 2244817 CO CLOMIPRAMINE TABLETS 25MG N06A N06AA CLOMIPRAMINE N06AA04 2244818 CO CLOMIPRAMINE TABLETS 50MG N06A N06AB SERTRALINE N06AB06 2244838 PMS-SERTRALINE N06A N06AB SERTRALINE N06AB06 2244839 PMS-SERTRALINE N06A N06AB SERTRALINE N06AB06 2244840 PMS-SERTRALINE N06A N06AX NEFAZODONE N06AX06 2245101 PMS-NEFAZODONE    187  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX NEFAZODONE N06AX06 2245102 PMS-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2245103 PMS-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2245111 PMS-NEFAZODONE N06A N06AB SERTRALINE N06AB06 2245159 SANDOZ SERTRALINE N06A N06AB SERTRALINE N06AB06 2245160 SANDOZ SERTRALINE N06A N06AB SERTRALINE N06AB06 2245161 SANDOZ SERTRALINE N06A N06AX NEFAZODONE N06AX06 2245202 GEN-NEFAZODONE 50 MG TABLETS N06A N06AX NEFAZODONE N06AX06 2245203 GEN-NEFAZODONE 100 MG TABLETS N06A N06AX NEFAZODONE N06AX06 2245204 GEN-NEFAZODONE 150 MG TABLETS N06A N06AX NEFAZODONE N06AX06 2245205 GEN-NEFAZODONE 200 MG TABLETS N06A N06AB FLUOXETINE N06AB03 2245281 FXT 10 N06A N06AB FLUOXETINE N06AB03 2245282 FXT 20 N06A N06AB FLUOXETINE N06AB03 2245283 FXT 40 N06A N06AX NEFAZODONE N06AX06 2245434 NOVO-NEFAZODONE-5HT2 TABLETS N06A N06AX NEFAZODONE N06AX06 2245435 NOVO-NEFAZODONE-5HT2 TABLETS N06A N06AX NEFAZODONE N06AX06 2245436 NOVO-NEFAZODONE-5HT2 TABLETS N06A N06AX NEFAZODONE N06AX06 2245437 NOVO-NEFAZODONE-5HT2 TABLETS N06A N06AB SERTRALINE N06AB06 2245748 DOM-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245749 DOM-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245750 DOM-SERTRALINE N06A N06AX NEFAZODONE N06AX06 2245754 DOM-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2245755 DOM-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2245756 DOM-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2245757 DOM-NEFAZODONE N06A N06AB SERTRALINE N06AB06 2245787 RATIO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245788 RATIO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245789 RATIO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245824 PHL-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245825 PHL-SERTRALINE N06A N06AB SERTRALINE N06AB06 2245826 PHL-SERTRALINE N06A N06AB CITALOPRAM N06AB04 2246056 APO-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2246057 APO-CITALOPRAM N06A N06AX NEFAZODONE N06AX06 2246548 PHL-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2246549 PHL-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2246550 PHL-NEFAZODONE N06A N06AX NEFAZODONE N06AX06 2246551 PHL-NEFAZODONE N06A N06AB CITALOPRAM N06AB04 2246593 MYLAN-CITALOPRAM 10MG TABLET N06A N06AB CITALOPRAM N06AB04 2246594 MYLAN-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2246595 MYLAN-CITALOPRAM    188  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB FLUVOXAMINE N06AB08 2246822 FLUVOXAMINE 50MG TABLET N06A N06AB FLUVOXAMINE N06AB08 2246823 FLUVOXAMINE 100MG TABLET N06A N06AX NEFAZODONE N06AX06 2246946 RATIO-NEFAZODONE 50MG TAB N06A N06AX NEFAZODONE N06AX06 2246947 RATIO-NEFAZODONE 100MG TAB N06A N06AX NEFAZODONE N06AX06 2246948 RATIO-NEFAZODONE 150MG TAB N06A N06AX NEFAZODONE N06AX06 2246949 RATIO-NEFAZODONE 200MG TAB N06A N06AB SERTRALINE N06AB06 2247047 NU-SERTRALINE CAPSULES N06A N06AB SERTRALINE N06AB06 2247048 NU-SERTRALINE CAPSULES N06A N06AB SERTRALINE N06AB06 2247050 NU-SERTRALINE CAPSULES N06A N06AB FLUVOXAMINE N06AB08 2247054 SANDOZ FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2247055 SANDOZ FLUVOXAMINE N06A N06AA AMITRIPTYLINE N06AA09 2247166 BIO-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2247167 BIO-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2247168 BIO-AMITRIPTYLINE N06A N06AX NEFAZODONE N06AX06 2247246 NU-NEFAZODONE 50MG TABLET N06A N06AX NEFAZODONE N06AX06 2247247 NU-NEFAZODONE 100MG TABLET N06A N06AX NEFAZODONE N06AX06 2247248 NU-NEFAZODONE 150MG TABLET N06A N06AX NEFAZODONE N06AX06 2247249 NU-NEFAZODONE 200MG TABLET N06A N06AA AMITRIPTYLINE N06AA09 2247302 PMS-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2247303 PMS-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2247304 PMS-AMITRIPTYLINE N06A N06AB FLUOXETINE N06AB03 2247528 FLUOXETINE 10MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2247529 FLUOXETINE 20MG CAPSULE N06A N06AB PAROXETINE N06AB05 2247750 PMS-PAROXETINE N06A N06AB PAROXETINE N06AB05 2247751 PMS-PAROXETINE N06A N06AB PAROXETINE N06AB05 2247752 PMS-PAROXETINE N06A N06AB PAROXETINE N06AB05 2247810 RATIO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2247811 RATIO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2247812 RATIO-PAROXETINE N06A N06AB CITALOPRAM N06AB04 2248010 PMS-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248011 PMS-CITALOPRAM N06A N06AB PAROXETINE N06AB05 2248012 MYLAN-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248013 MYLAN-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248014 MYLAN-PAROXETINE N06A N06AB CITALOPRAM N06AB04 2248050 CO CITALOPRAM    189  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB CITALOPRAM N06AB04 2248051 CO CITALOPRAM N06A N06AA AMITRIPTYLINE N06AA09 2248131 DOM-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2248132 DOM-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2248133 DOM-AMITRIPTYLINE N06A N06AB CITALOPRAM N06AB04 2248170 SANDOZ CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248171 SANDOZ CITALOPRAM N06A N06AB PAROXETINE N06AB05 2248447 DOM-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248448 DOM-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248449 DOM-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248450 PAROXETINE N06A N06AB PAROXETINE N06AB05 2248451 PAROXETINE N06A N06AB PAROXETINE N06AB05 2248452 PAROXETINE N06A N06AB SERTRALINE N06AB06 2248496 RIVA-SERTRALINE 25 N06A N06AB SERTRALINE N06AB06 2248497 RIVA-SERTRALINE 50 N06A N06AB SERTRALINE N06AB06 2248498 RIVA-SERTRALINE 100 N06A N06AB PAROXETINE N06AB05 2248503 PAXIL CR N06A N06AB PAROXETINE N06AB05 2248504 PAXIL CR N06A N06AX TRYPTOPHAN N06AX02 2248537 APO-TRYPTOPHAN 250MG TAB N06A N06AX TRYPTOPHAN N06AX02 2248538 APO-TRYPTOPHAN N06A N06AX TRYPTOPHAN N06AX02 2248539 APO-TRYPTOPHAN N06A N06AX TRYPTOPHAN N06AX02 2248540 APO-TRYPTOPHAN N06A N06AX MIRTAZAPINE N06AX11 2248542 REMERON RD N06A N06AX MIRTAZAPINE N06AX11 2248543 REMERON RD N06A N06AX MIRTAZAPINE N06AX11 2248544 REMERON RD N06A N06AB PAROXETINE N06AB05 2248556 TEVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248557 TEVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248558 TEVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248559 RIVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248560 RIVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248561 RIVA-PAROXETINE N06A N06AB PAROXETINE N06AB05 2248719 NU-PAROXETINE 10MG TABLET N06A N06AB PAROXETINE N06AB05 2248720 NU-PAROXETINE 20MG TABLET N06A N06AB PAROXETINE N06AB05 2248721 NU-PAROXETINE 30MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2248762 PMS-MIRTAZAPINE N06A N06AB PAROXETINE N06AB05 2248913 PAROXETINE-10 N06A N06AB PAROXETINE N06AB05 2248914 PAROXETINE-20 N06A N06AB PAROXETINE N06AB05 2248915 PAROXETINE-30 N06A N06AB CITALOPRAM N06AB04 2248942 DOM-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248943 DOM-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248944 PHL-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248945 PHL-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2248996 NU-CITALOPRAM 20MG TABLET N06A N06AB CITALOPRAM N06AB04 2248997 NU-CITALOPRAM 40MG TABLET    190  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB CITALOPRAM N06AB04 2249219 PREM-CITALOPRAM 10MG TABLET N06A N06AB CITALOPRAM N06AB04 2249227 PREM-CITALOPRAM 20MG TABLET N06A N06AB CITALOPRAM N06AB04 2249235 PREM-CITALOPRAM 40MG TABLET N06A N06AB CITALOPRAM N06AB04 2249251 RIVA-CITALOPRAM 10MG TABLET N06A N06AB CITALOPRAM N06AB04 2249278 RIVA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2249286 RIVA-CITALOPRAM N06A N06AX TRAZODONE N06AX05 2249804 ZYM-TRAZODONE N06A N06AX MIRTAZAPINE N06AX11 2250594 SANDOZ MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2250608 SANDOZ MIRTAZAPINE N06A N06AB PAROXETINE N06AB05 2251345 PREM-PAROXETINE 10MG TABLET N06A N06AB PAROXETINE N06AB05 2251353 PREM-PAROXETINE 20MG TABLET N06A N06AB PAROXETINE N06AB05 2251361 PREM-PAROXETINE 30MG TABLET N06A N06AB CITALOPRAM N06AB04 2251558 NOVO-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2251566 NOVO-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2252112 RATIO-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2252120 RATIO-CITALOPRAM N06A N06AX MIRTAZAPINE N06AX11 2252279 MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2252287 DOM-MIRTAZAPINE N06A N06AB PAROXETINE N06AB05 2254743 SANDOZ PAROXETINE 10MG TAB N06A N06AB PAROXETINE N06AB05 2254751 SANDOZ PAROXETINE N06A N06AB PAROXETINE N06AB05 2254778 SANDOZ PAROXETINE N06A N06AB FLUVOXAMINE N06AB08 2255529 CO FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2255537 CO FLUVOXAMINE N06A N06AX MIRTAZAPINE N06AX11 2256096 MYLAN-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2256118 MYLAN-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2256126 MYLAN-MIRTAZAPINE N06A N06AB CITALOPRAM N06AB04 2257513 CITALOPRAM-20 N06A N06AB CITALOPRAM N06AB04 2257521 CITALOPRAM-40 N06A N06AB FLUVOXAMINE N06AB08 2257661 BCI FLUVOXAMINE TABLETS N06A N06AB FLUVOXAMINE N06AB08 2257688 BCI FLUVOXAMINE TABLETS N06A N06AX TRYPTOPHAN N06AX02 2259176 RATIO-TRYPTOPHAN 750MG TAB N06A N06AX MIRTAZAPINE N06AX11 2259354 NOVO-MIRTAZAPINE TABLETS N06A N06AX TRYPTOPHAN N06AX02 2262436 PHL-TRYPTOPHAN N06A N06AX TRYPTOPHAN N06AX02 2262444 PHL-TRYPTOPHAN N06A N06AB FLUVOXAMINE N06AB08 2262622 PHL-FLUVOXAMINE    191  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB FLUVOXAMINE N06AB08 2262630 PHL-FLUVOXAMINE N06A N06AB PAROXETINE N06AB05 2262746 CO PAROXETINE N06A N06AB PAROXETINE N06AB05 2262754 CO PAROXETINE N06A N06AB PAROXETINE N06AB05 2262762 CO PAROXETINE N06A N06AB ESCITALOPRAM N06AB10 2263238 CIPRALEX   -10MG N06A N06AB ESCITALOPRAM N06AB10 2263254 CIPRALEX   -20MG N06A N06AX MIRTAZAPINE N06AX11 2265265 RIVA-MIRTAZAPINE N06A N06AB FLUOXETINE N06AB03 2266776 BCI FLUOXETINE 20MG CAPSULE N06A N06AX MIRTAZAPINE N06AX11 2267284 SANDOZ MIRTAZAPINE FC 15MG N06A N06AX MIRTAZAPINE N06AX11 2267292 SANDOZ MIRTAZAPINE FC N06A N06AB CITALOPRAM N06AB04 2268000 RAN-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2268019 RAN-CITALOPRAM N06A N06AB PAROXETINE N06AB05 2269422 SANDOZ PAROXETINE N06A N06AB PAROXETINE N06AB05 2269430 SANDOZ PAROXETINE N06A N06AB PAROXETINE N06AB05 2269449 SANDOZ PAROXETINE N06A N06AB CITALOPRAM N06AB04 2270609 PMS-CITALOPRAM N06A N06AX MIRTAZAPINE N06AX11 2270927 RATIO-MIRTAZAPINE N06A N06AB CITALOPRAM N06AB04 2273055 DOM-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2273543 PHL-CITALOPRAM N06A N06AB SERTRALINE N06AB06 2273683 GD-SERTRALINE N06A N06AB SERTRALINE N06AB06 2273691 GD-SERTRALINE N06A N06AB SERTRALINE N06AB06 2273705 GD-SERTRALINE N06A N06AX MIRTAZAPINE N06AX11 2273942 PMS-MIRTAZAPINE N06A N06AX VENLAFAXINE N06AX16 2273969 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2273977 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2273985 VENLAFAXINE XR N06A N06AX MIRTAZAPINE N06AX11 2274361 CO MIRTAZAPINE N06A N06AX VENLAFAXINE N06AX16 2275023 TEVA-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2275031 TEVA-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2275058 TEVA-VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2275562 AURO-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2275570 AURO-CITALOPRAM N06A N06AX TRAZODONE N06AX05 2277344 RATIO-TRAZODONE N06A N06AX TRAZODONE N06AX05 2277352 RATIO-TRAZODONE N06A N06AX TRAZODONE N06AX05 2277360 RATIO-TRAZODONE N06A N06AX VENLAFAXINE N06AX16 2278545 PMS-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2278553 PMS-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2278561 PMS-VENLAFAXINE XR    192  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX MIRTAZAPINE N06AX11 2279894 NOVO-MIRTAZAPINE OD TABLETS N06A N06AX MIRTAZAPINE N06AX11 2279908 NOVO-MIRTAZAPINE OD TABLETS N06A N06AX MIRTAZAPINE N06AX11 2279916 NOVO-MIRTAZAPINE OD TABLETS N06A N06AX MIRTAZAPINE N06AX11 2281716 DOM-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2281732 MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2282356 MIRTAZAPINE 15 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2282364 MIRTAZAPINE 30 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2282372 MIRTAZAPINE 45 MG TABLET N06A N06AB PAROXETINE N06AB05 2282844 PAROXETINE N06A N06AB PAROXETINE N06AB05 2282852 PAROXETINE N06A N06AB PAROXETINE N06AB05 2282860 PAROXETINE N06A N06AB SERTRALINE N06AB06 2284804 SERTRALINE N06A N06AB SERTRALINE N06AB06 2284812 SERTRALINE N06A N06AB SERTRALINE N06AB06 2284820 SERTRALINE N06A N06AB CITALOPRAM N06AB04 2284839 CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2284847 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2284855 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2285622 RAN-CITALO N06A N06AB CITALOPRAM N06AB04 2285630 RAN-CITALO N06A N06AB FLUOXETINE N06AB03 2286068 FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2286076 FLUOXETINE N06A N06AX MIRTAZAPINE N06AX11 2286610 APO-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2286629 APO-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2286637 APO-MIRTAZAPINE N06A N06AB SERTRALINE N06AB06 2287390 CO SERTRALINE N06A N06AB SERTRALINE N06AB06 2287404 CO SERTRALINE N06A N06AB SERTRALINE N06AB06 2287412 CO SERTRALINE N06A N06AX VENLAFAXINE N06AX16 2289059 VENLAFAXINE XR 37.5 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2289067 VENLAFAXINE XR 75 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2289075 VENLAFAXINE XR 150 MG CAPSULE N06A N06AB CITALOPRAM N06AB04 2293218 TEVA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2293226 TEVA-CITALOPRAM N06A N06AB PAROXETINE N06AB05 2293749 PMS-PAROXETINE N06A N06AB CITALOPRAM N06AB04 2295431 CO CITALOPRAM 30 MG TABLET N06A N06AB CITALOPRAM N06AB04 2296152 CTP 30 N06A N06AX MIRTAZAPINE N06AX11 2296675 MIRTAZAPINE - 15 N06A N06AX MIRTAZAPINE N06AX11 2296683 MIRTAZAPINE - 30 N06A N06AX MIRTAZAPINE N06AX11 2296691 MIRTAZAPINE - 45 N06A N06AB PAROXETINE N06AB05 2298961 DOM-PAROXETINE 40 MG TABLET    193  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB PAROXETINE N06AB05 2298988 PAROXETINE 40 MG TABLET N06A N06AX VENLAFAXINE N06AX16 2299291 DOM-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2299305 DOM-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2299313 DOM-VENLAFAXINE XR N06A N06AX MIRTAZAPINE N06AX11 2299801 AURO-MIRTAZAPINE OD N06A N06AX MIRTAZAPINE N06AX11 2299828 AURO-MIRTAZAPINE OD N06A N06AX MIRTAZAPINE N06AX11 2299836 AURO-MIRTAZAPINE OD N06A N06AX DULOXETINE N06AX21 2301482 CYMBALTA N06A N06AX DULOXETINE N06AX21 2301490 CYMBALTA N06A N06AB CITALOPRAM N06AB04 2301822 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2301830 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2301849 CITALOPRAM N06A N06AB PAROXETINE N06AB05 2302012 PAROXETINE N06A N06AB PAROXETINE N06AB05 2302020 PAROXETINE N06A N06AB PAROXETINE N06AB05 2302039 PAROXETINE N06A N06AB PAROXETINE N06AB05 2302047 ZYM-PAROXETINE 40 MG TABLET N06A N06AB CITALOPRAM N06AB04 2302535 IPG-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2302543 IPG-CITALOPRAM N06A N06AB FLUOXETINE N06AB03 2302659 ZYM-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2302667 ZYM-FLUOXETINE N06A N06AB CITALOPRAM N06AB04 2303256 RIVA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2303264 RIVA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2303272 RIVA-CITALOPRAM N06A N06AB FLUVOXAMINE N06AB08 2303345 RIVA-FLUVOX N06A N06AB FLUVOXAMINE N06AB08 2303361 RIVA-FLUVOX N06A N06AB SERTRALINE N06AB06 2303779 SERTRALINE N06A N06AB SERTRALINE N06AB06 2303809 SERTRALINE N06A N06AB SERTRALINE N06AB06 2303817 SERTRALINE N06A N06AX VENLAFAXINE N06AX16 2304317 CO VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2304325 CO VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2304333 CO VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2304686 MINT-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2304694 MINT-CITALOPRAM N06A N06AB FLUOXETINE N06AB03 2305461 RIVA-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2305488 RIVA-FLUOXETINE N06A N06AB CITALOPRAM N06AB04 2306239 CITALOPRAM-ODAN N06A N06AB CITALOPRAM N06AB04 2306247 CITALOPRAM-ODAN N06A N06AX VENLAFAXINE N06AX16 2307774 RIVA-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2307782 RIVA-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2307790 RIVA-VENLAFAXINE XR    194  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX VENLAFAXINE N06AX16 2310279 MYLAN-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2310287 MYLAN-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2310295 MYLAN-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2310317 SANDOZ VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2310325 SANDOZ VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2310333 SANDOZ VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2312336 TEVA-CITALOPRAM N06A N06AX MIRTAZAPINE N06AX11 2312778 PRO-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2312786 PRO-MIRTAZAPINE N06A N06AB CITALOPRAM N06AB04 2313405 JAMP-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2313413 JAMP-CITALOPRAM N06A N06AB FLUOXETINE N06AB03 2314991 PRO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2315009 PRO-FLUOXETINE N06A N06AX DESVENLAFAXINE N06AX23 2321092 PRISTIQ N06A N06AX DESVENLAFAXINE N06AX23 2321106 PRISTIQ N06A N06AB CITALOPRAM N06AB04 2322781 NG CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2322803 NG CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2325047 CITALOPRAM-10 N06A N06AX TRAZODONE N06AX05 2325101 ZYM-TRAZODONE N06A N06AX TRAZODONE N06AX05 2325128 ZYM-TRAZODONE N06A N06AX MIRTAZAPINE N06AX11 2325179 ZYM-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2325187 ZYM-MIRTAZAPINE N06A N06AX BUPROPION N06AX12 2325357 BUPROPION SR N06A N06AX BUPROPION N06AX12 2325373 PMS-BUPROPION SR N06A N06AA AMITRIPTYLINE N06AA09 2326051 TEVA-AMITRIPTYLINE 25 MG TAB N06A N06AA AMITRIPTYLINE N06AA09 2326078 TEVA-AMITRIPTYLINE 50 MG TAB N06A N06AB SERTRALINE N06AB06 2331535 SIG-SERTRALINE 25 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2331543 SIG-SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2331578 SIG-SERTRALINE 100 MG CAPSULE N06A N06AX BUPROPION N06AX12 2331616 BUPROPION SR N06A N06AX VENLAFAXINE N06AX16 2331683 APO-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2331691 APO-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2331705 APO-VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2331950 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2331977 CITALOPRAM 40 MG TABLET    195  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB CITALOPRAM N06AB04 2333961 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2333988 CITALOPRAM 40 MG TABLET N06A N06AB FLUOXETINE N06AB03 2335131 FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2335156 FLUOXETINE 20 MG CAPSULE N06A N06AX MIRTAZAPINE N06AX11 2335913 MIRTAZAPINE 15 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2335921 MIRTAZAPINE 30 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2336189 NU-MIRTAZAPINE 15 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2336197 NU-MIRTAZAPINE 30 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2336200 NU-MIRTAZAPINE 45 MG TABLET N06A N06AB CITALOPRAM N06AB04 2338769 Q-CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2338777 Q-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2338785 Q-CITALOPRAM 40 MG TABLET N06A N06AB PAROXETINE N06AB05 2338793 Q-PAROXETINE N06A N06AB PAROXETINE N06AB05 2338807 Q-PAROXETINE N06A N06AB PAROXETINE N06AB05 2338815 Q-PAROXETINE N06A N06AX VENLAFAXINE N06AX16 2338823 Q-VENLAFAXINE XR 37.5 MG CAP N06A N06AX VENLAFAXINE N06AX16 2338831 Q-VENLAFAXINE XR 75 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2338858 Q-VENLAFAXINE XR 150 MG CAP N06A N06AX VENLAFAXINE N06AX16 2339242 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2339250 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2339269 VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2340712 APX-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2340720 APX-CITALOPRAM 40 MG TABLET N06A N06AB FLUOXETINE N06AB03 2341743 APX-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2341751 APX-FLUOXETINE 20 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2341778 APX-FLUOXETINE 20 MG/5 ML SOLN N06A N06AB FLUVOXAMINE N06AB08 2341824 APX-FLUVOXAMINE 50 MG TABLET N06A N06AB FLUVOXAMINE N06AB08 2341832 APX-FLUVOXAMINE 100 MG TABLET N06A N06AB SERTRALINE N06AB06 2341840 APX-SERTRALINE 25 MG CAPSULE    196  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB SERTRALINE N06AB06 2341859 APX-SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2341867 APX-SERTRALINE 100 MG CAPSULE N06A N06AB PAROXETINE N06AB05 2342251 APX-PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2342278 APX-PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2342286 APX-PAROXETINE 30 MG TABLET N06A N06AB SERTRALINE N06AB06 2342359 SERTRALINE 25 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2342367 SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2342375 SERTRALINE 100 MG CAPSULE N06A N06AB PAROXETINE N06AB05 2342383 PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2342391 PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2342405 PAROXETINE 30 MG TABLET N06A N06AB CITALOPRAM N06AB04 2342766 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2342774 CITALOPRAM 40 MG TABLET N06A N06AB FLUOXETINE N06AB03 2344149 FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2344157 FLUOXETINE N06A N06AA AMITRIPTYLINE N06AA09 2344203 APX-AMITRIPTYLINE 75 MG TABLET N06A N06AX VENLAFAXINE N06AX16 2344742 VENLAFAXINE XR 37.5 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2344750 VENLAFAXINE XR 75 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2344769 VENLAFAXINE XR 150 MG CAPSULE N06A N06AX TRAZODONE N06AX05 2345943 NTP-TRAZODONE N06A N06AX TRAZODONE N06AX05 2345951 NTP-TRAZODONE N06A N06AX TRAZODONE N06AX05 2345978 NTP-TRAZODONE N06A N06AX VENLAFAXINE N06AX16 2346303 SIG-VENLAFAXINE XR 37.5 MG CAP N06A N06AX VENLAFAXINE N06AX16 2346311 SIG-VENLAFAXINE XR 75 MG CAP N06A N06AX VENLAFAXINE N06AX16 2346338 SIG-VENLAFAXINE XR 150 MG CAP N06A N06AA DOXEPIN N06AA12 2348349 NTP-DOXEPIN 25 MG CAPSULE N06A N06AA DOXEPIN N06AA12 2348357 NTP-DOXEPIN 50 MG CAPSULE N06A N06AA DOXEPIN N06AA12 2348365 NTP-DOXEPIN 75 MG CAPSULE N06A N06AA DOXEPIN N06AA12 2348373 NTP-DOXEPIN 100 MG CAPSULE    197  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX TRAZODONE N06AX05 2348772 TRAZODONE N06A N06AX TRAZODONE N06AX05 2348780 TRAZODONE N06A N06AX TRAZODONE N06AX05 2348799 TRAZODONE N06A N06AA TRIMIPRAMINE N06AA06 2349973 AA-TRIMIP 12.5 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2349981 AA-TRIMIP 25 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2350025 AA-TRIMIP 50 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2350033 AA-TRIMIP 100 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2350041 AA-TRIMIP 75 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2351838 NTP-VENLAFAXINE XR 37.5 MG CAP N06A N06AX VENLAFAXINE N06AX16 2351846 NTP-VENLAFAXINE XR 75 MG CAP N06A N06AX VENLAFAXINE N06AX16 2351854 NTP-VENLAFAXINE XR 150 MG CAP N06A N06AB CITALOPRAM N06AB04 2351900 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2351919 CITALOPRAM 40 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2352826 GD-MIRTAZAPINE OD N06A N06AX MIRTAZAPINE N06AX11 2352834 GD-MIRTAZAPINE OD N06A N06AX MIRTAZAPINE N06AX11 2352842 GD-MIRTAZAPINE OD N06A N06AB SERTRALINE N06AB06 2353520 SERTRALINE N06A N06AB SERTRALINE N06AB06 2353539 SERTRALINE N06A N06AB SERTRALINE N06AB06 2353547 SERTRALINE N06A N06AB CITALOPRAM N06AB04 2353636 CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2353644 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2353652 CITALOPRAM 40 MG TABLET N06A N06AB CITALOPRAM N06AB04 2353660 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2353679 CITALOPRAM N06A N06AX VENLAFAXINE N06AX16 2354713 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2354721 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2354748 VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2355248 ACCEL-CITALOPRAM TABLETS N06A N06AB CITALOPRAM N06AB04 2355256 ACCEL-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2355264 ACCEL-CITALOPRAM 40 MG TABLET N06A N06AB CITALOPRAM N06AB04 2355272 SEPTA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2355280 SEPTA-CITALOPRAM N06A N06AB SERTRALINE N06AB06 2357143 JAMP-SERTRALINE N06A N06AB SERTRALINE N06AB06 2357151 JAMP-SERTRALINE N06A N06AB SERTRALINE N06AB06 2357178 JAMP-SERTRALINE    198  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX MIRTAZAPINE N06AX11 2358832 NTP-MIRTAZAPINE 30 MG TABLET N06A N06AX VENLAFAXINE N06AX16 2360020 GD-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2360039 GD-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2360047 GD-VENLAFAXINE XR N06A N06AB CITALOPRAM N06AB04 2360292 NTP-CITALOPRAM 10MG TABLET N06A N06AB CITALOPRAM N06AB04 2360306 NTP-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2360314 NTP-CITALOPRAM 40 MG TABLET N06A N06AB CITALOPRAM N06AB04 2360683 GD-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2360691 GD-CITALOPRAM 40 MG TABLET N06A N06AX TRAZODONE N06AX05 2361868 OLEPTRO N06A N06AX TRAZODONE N06AX05 2361876 OLEPTRO N06A N06AX MIRTAZAPINE N06AX11 2362929 AVA-MIRTAZAPINE N06A N06AX MIRTAZAPINE N06AX11 2362937 AVA-MIRTAZAPINE N06A N06AX BUPROPION N06AX12 2363399 AVA-BUPROPION SR N06A N06AX BUPROPION N06AX12 2363402 AVA-BUPROPION SR N06A N06AB FLUVOXAMINE N06AB08 2363763 AVA-FLUVOXAMINE N06A N06AB FLUVOXAMINE N06AB08 2363771 AVA-FLUVOXAMINE N06A N06AB CITALOPRAM N06AB04 2364255 AVA-CITALOPRAM N06A N06AA NORTRIPTYLINE N06AA10 2364301 AVA-NORTRIPTYLINE N06A N06AA NORTRIPTYLINE N06AA10 2364328 AVA-NORTRIPTYLINE N06A N06AB SERTRALINE N06AB06 2364360 AVA-SERTRALINE 25 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2364379 AVA-SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2364387 AVA-SERTRALINE 100 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2364522 AVA-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2364530 AVA-FLUOXETINE N06A N06AB PAROXETINE N06AB05 2365103 AVA-PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2365111 AVA-PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2365138 AVA-PAROXETINE 30 MG TABLET N06A N06AX TRAZODONE N06AX05 2365626 AVA-TRAZODONE 50 MG TABLET N06A N06AX TRAZODONE N06AX05 2365634 AVA-TRAZODONE 100 MG TABLET N06A N06AX TRAZODONE N06AX05 2365642 AVA-TRAZODONE 150 MG TABLET N06A N06AB CITALOPRAM N06AB04 2367793 AVA-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2367807 AVA-CITALOPRAM 40 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2368560 JAMP-MIRTAZAPINE    199  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX MIRTAZAPINE N06AX11 2368579 JAMP-MIRTAZAPINE N06A N06AB PAROXETINE N06AB05 2368862 JAMP-PAROXETINE N06A N06AB PAROXETINE N06AB05 2368870 JAMP-PAROXETINE N06A N06AB PAROXETINE N06AB05 2368889 JAMP-PAROXETINE N06A N06AB CITALOPRAM N06AB04 2369664 IPG-CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2370077 MINT-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2370085 JAMP-CITALOPRAM N06A N06AX MIRTAZAPINE N06AX11 2370093 NTP-MIRTAZAPINE OD 15 MG TAB N06A N06AX MIRTAZAPINE N06AX11 2370107 NTP-MIRTAZAPINE OD 30 MG TAB N06A N06AX MIRTAZAPINE N06AX11 2370115 NTP-MIRTAZAPINE OD 45 MG TAB N06A N06AB FLUOXETINE N06AB03 2370573 Q-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2370581 Q-FLUOXETINE N06A N06AX MIRTAZAPINE N06AX11 2370689 MIRTAZAPINE N06A N06AB SERTRALINE N06AB06 2370948 Q-SERTRALINE N06A N06AB SERTRALINE N06AB06 2370956 Q-SERTRALINE N06A N06AB SERTRALINE N06AB06 2370964 Q-SERTRALINE N06A N06AB CITALOPRAM N06AB04 2371871 MAR-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2371898 MAR-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2371901 MAR-CITALOPRAM N06A N06AX VENLAFAXINE N06AX16 2373858 VENLAFAXINE HCL XR 37.5 MG CAP N06A N06AX VENLAFAXINE N06AX16 2373866 VENLAFAXINE HCL XR 75 MG CAP N06A N06AX VENLAFAXINE N06AX16 2373874 VENLAFAXINE HCL XR 150 MG CAP N06A N06AB CITALOPRAM N06AB04 2373971 MANDA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2373998 MANDA-CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2374005 MANDA-CITALOPRAM N06A N06AB FLUOXETINE N06AB03 2374447 FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2374455 FLUOXETINE N06A N06AB SERTRALINE N06AB06 2374552 RAN-SERTRALINE N06A N06AB SERTRALINE N06AB06 2374560 RAN-SERTRALINE N06A N06AB SERTRALINE N06AB06 2374579 RAN-SERTRALINE N06A N06AB CITALOPRAM N06AB04 2374617 AG-CITALOPRAM 10 MG TABLET N06A N06AX VENLAFAXINE N06AX16 2380072 RAN-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2380080 RAN-VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2380099 RAN-VENLAFAXINE XR N06A N06AB FLUOXETINE N06AB03 2380560 MINT-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2380579 MINT-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2382393 FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2382407 FLUOXETINE 20 MG CAPSULE    200  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AB PAROXETINE N06AB05 2382504 NTP-PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2382512 NTP-PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2382520 NTP-PAROXETINE 30 MG TABLET N06A N06AB FLUOXETINE N06AB03 2383241 FLUOXETINE CAPSULES BP N06A N06AB PAROXETINE N06AB05 2383276 AURO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2383284 AURO-PAROXETINE N06A N06AB PAROXETINE N06AB05 2383292 AURO-PAROXETINE N06A N06AB FLUOXETINE N06AB03 2385627 AURO-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2385635 AURO-FLUOXETINE N06A N06AX VENLAFAXINE N06AX16 2385929 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2385937 VENLAFAXINE XR N06A N06AX VENLAFAXINE N06AX16 2385945 VENLAFAXINE XR N06A N06AB SERTRALINE N06AB06 2386070 SERTRALINE N06A N06AB SERTRALINE N06AB06 2386089 SERTRALINE N06A N06AB SERTRALINE N06AB06 2386097 SERTRALINE N06A N06AB FLUOXETINE N06AB03 2386402 JAMP-FLUOXETINE N06A N06AB CITALOPRAM N06AB04 2387948 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2387956 CITALOPRAM N06A N06AB CITALOPRAM N06AB04 2387964 CITALOPRAM N06A N06AB PAROXETINE N06AB05 2388227 PAROXETINE N06A N06AB PAROXETINE N06AB05 2388235 PAROXETINE N06A N06AB PAROXETINE N06AB05 2388243 PAROXETINE N06A N06AB SERTRALINE N06AB06 2390906 AURO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2390914 AURO-SERTRALINE N06A N06AB SERTRALINE N06AB06 2390922 AURO-SERTRALINE N06A N06AX VENLAFAXINE N06AX16 2391392 VENLAFAXINE XR 37.5 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2391406 VENLAFAXINE XR 75 MG CAPSULE N06A N06AX VENLAFAXINE N06AX16 2391414 VENLAFAXINE XR 150 MG CAPSULE N06A N06AB ESCITALOPRAM N06AB10 2391449 CIPRALEX MELTZ N06A N06AB ESCITALOPRAM N06AB10 2391457 CIPRALEX MELTZ N06A N06AX BUPROPION N06AX12 2391562 BUPROPION SR N06A N06AX BUPROPION N06AX12 2391570 BUPROPION SR N06A N06AB FLUOXETINE N06AB03 2391627 IPG-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2391635 IPG-FLUOXETINE 20 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2392658 GD-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2392666 GD-FLUOXETINE 20 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2392909 MAR-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2392917 MAR-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2393441 FLUOXETINE CAPSULES BP N06A N06AA DOXEPIN N06AA12 2398257 SILENOR    201  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AA DOXEPIN N06AA12 2398265 SILENOR N06A N06AB SERTRALINE N06AB06 2398303 IPG-SERTRALINE 25 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2398311 IPG-SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2398338 IPG-SERTRALINE 100 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2399415 MAR-SERTRALINE 25 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2399423 MAR-SERTRALINE 50 MG CAPSULE N06A N06AB SERTRALINE N06AB06 2399431 MAR-SERTRALINE 100 MG CAPSULE N06A N06AB CITALOPRAM N06AB04 2400316 CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2400324 CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2400332 CITALOPRAM 40 MG TABLET N06A N06AB FLUOXETINE N06AB03 2400391 ACCEL-FLUOXETINE 10 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2400405 ACCEL-FLUOXETINE 20 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2401894 JAMP-FLUOXETINE N06A N06AB SERTRALINE N06AB06 2402378 MINT-SERTRALINE N06A N06AB SERTRALINE N06AB06 2402394 MINT-SERTRALINE N06A N06AB SERTRALINE N06AB06 2402408 MINT-SERTRALINE N06A N06AA AMITRIPTYLINE N06AA09 2403137 APO-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2403145 APO-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2403153 APO-AMITRIPTYLINE N06A N06AA AMITRIPTYLINE N06AA09 2403161 APO-AMITRIPTYLINE N06A N06AA TRIMIPRAMINE N06AA06 2405369 APO-TRIMIPRAMINE 12.5 MG TAB N06A N06AA TRIMIPRAMINE N06AA06 2405377 APO-TRIMIPRAMINE 25 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2405385 APO-TRIMIPRAMINE 50 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2405393 APO-TRIMIPRAMINE 100 MG TABLET N06A N06AA TRIMIPRAMINE N06AA06 2405407 APO-TRIMIPRAMINE 75 MG CAPSULE N06A N06AB FLUOXETINE N06AB03 2405695 RAN-FLUOXETINE N06A N06AB FLUOXETINE N06AB03 2405709 RAN-FLUOXETINE N06A N06AB PAROXETINE N06AB05 2410834 IPG-PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2410842 IPG-PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2410850 IPG-PAROXETINE 30 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2411695 AURO-MIRTAZAPINE 15 MG TABLET    202  ATC 3 CD ATC 4 CD ATC 5 ATC 5 CD DIN/ PIN Brand Name N06A N06AX MIRTAZAPINE N06AX11 2411709 AURO-MIRTAZAPINE 30 MG TABLET N06A N06AX MIRTAZAPINE N06AX11 2411717 AURO-MIRTAZAPINE 45 MG TABLET N06A N06AB PAROXETINE N06AB05 2411946 MAR-PAROXETINE 10 MG TABLET N06A N06AB PAROXETINE N06AB05 2411954 MAR-PAROXETINE 20 MG TABLET N06A N06AB PAROXETINE N06AB05 2411962 MAR-PAROXETINE 30 MG TABLET N06A N06AB CITALOPRAM N06AB04 2414570 ABBOTT-CITALOPRAM 10 MG TABLET N06A N06AB CITALOPRAM N06AB04 2414589 ABBOTT-CITALOPRAM 20 MG TABLET N06A N06AB CITALOPRAM N06AB04 2414597 ABBOTT-CITALOPRAM 40 MG TABLET       203  Appendix C  – 2012 Canadian Community Health Survey-Mental Health Questionnaire        204  Sex  SEX_Q01 INTERVIEWER: Enter [respondent name]’s sex. If necessary, ask: (Is [respondent name] male or female?) 1 Male 2 Female (DK, RF are not allowed)  SEX_END   Marital status  MSNC_Q01 What is [respondent name]’s marital status? Is [he/she]: INTERVIEWER: Read categories to respondent. 1 … married? 2 … living common-law? 3 … widowed? 4 … separated? 5 … divorced? 6 … single, never married? MSNC_END   Relationship without confirmation  RNC_Q1 What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] 01 Husband/Wife  02 Common-law partner  03 Father/Mother (Go to RNC_Q2A) 04 Son/Daughter (Go to RNC_Q2B) 05 Brother/Sister (Go to RNC_Q2C) 06 Foster father/mother  07 Foster son/daughter  08 Grandfather/mother  09 Grandson/daughter  10 In-law (Go to RNC_Q2D) 11 Other related (Go to RNC_Q2E) 12 Unrelated (Go to RNC_Q2F) RNC_Q2A What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... birth father/mother? 2 ... step father/mother? 3 ... adoptive father/mother? RNC_Q2B What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... birth son/daughter? 2 ... step son/daughter? 3 ... adopted son/daughter?    205  RNC_Q2C What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... full brother/sister? 2 ... half brother/sister? 3 ... step brother/sister? 4 ... adopted brother/sister? 5 ... foster brother/sister? RNC_Q2D What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... father/mother-in-law? 2 ... son/daughter-in-law? 3 ... brother/sister-in-law? 4 ... other in-law? RNC_Q2E What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... uncle/aunt? 2 ... cousin? 3 ... nephew/niece? 4 ... other relative? RNC_Q2F What is the relationship of: to: [respondent name] [(Text sex, age)] [respondent name]? [(Text sex, age)] Is that a(n): 1 ... boyfriend/girlfriend? 2 ... room-mate? 3 ... other?  RNC_END   Age of respondent (AN3) 1 - Copied Version - 1001  AN3_BEG Content block External variables required: CURRDATE: Current date from operating system DOANC: Do block flag, from the sample file. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space AN3_C01A If DOANC = 1, go to AN3_R01. Otherwise, go to AN3_END.    206  AN3_R01 For some of the questions I’ll be asking, I need to know your exact date of birth. INTERVIEWER: Press <1> to continue. AN3_B01 Call sub block "Date" (DATE). AN3_C02 If AN3_B01.Year = DK, RF, go to AN3_Q03. Otherwise, go to AN3_D02. AN3_D02  Programmer: Calculate DV_AGE based on the entered date of birth and CURRDATE. AN3_E02A An impossible day/month/year combination has been entered. Please return and correct Rule : Trigger hard edit if DV_AGE > CURRDATE. AN3_Q02 So your age is ^DV_AGE. Is that correct?  1 Yes (Go to AN3_C03) 2 No, return and correct date of birth 3 No, collect age (Go to AN3_Q03) (DK, RF not allowed)  AN3_E02B Return to AN3_B01 and correct the date of birth. Rule : Trigger hard edit if AN3_Q02 = 2. AN3_C03 If DV_AGE < 15 years, go to AN3_R04. Otherwise, go to AN3_END. AN3_Q03 What is your age?  (MIN: 0) (MAX: 130)  (DK, RF not allowed) AN3_D03  Programmer: If AN3_Q02 = 3, DV_AGE = AN3_Q03. AN3_C04 If DV_AGE < 15 years, go to AN3_R04. Otherwise, go to AN3_END. AN3_R04 Because you are less than 15 years old, you are not eligible to participate in the Canadian Community Health Survey on Mental Health. INTERVIEWER: Press <1> to continue. Programmer: Auto code as 90 Unusual/Special circumstances and call the exit block. AN3_END    GEN_Q10 How would you describe your sense of belonging to your local community? Would you say it is…? INTERVIEWER: Read categories to respondent. 1 Very strong 2 Somewhat strong 3 Somewhat weak 4 Very weak DK, RF GEN_END   Depression (DEP) 1 - Copied Version - 1001  DEP_BEG Content block External variables required: DODEP: do block flag, from the sample file. SCR_Q21, SCR_Q22, SCR_Q23: from the SCR block    207  SUI_Q01, SUI_Q02, SUI_Q04A, SUI_Q06A, SUI_Q13, SUI_Q16, SUI_Q19 : from the Suicide sub-block SUI DV_AGE: age of selected respondent from AN3 block. SEX_Q01: sex of specific respondent (1 = male, 2 = female) from Sex block. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space Programmer: For the KEY PHRASES that are associated with DEP_Q24A to DEP_Q26BB as well as SUI_Q01, SUI_Q04A and SUI_Q06A, please create a parallel block to display the list of KEY PHRASES that can be triggered at any point during the module by pressing <CTRL D>. Only the KEY PHRASES for answers that the respondent reported will be displayed. Content type: NOTE TO DATA USERS: There are 4 different ways to enter the depression module, as described in the conditions below. The flow is based on responses to three screener questions: SCR_Q21, SCR_Q22 and SCR_Q23. Those respondents who said “yes” to the first screener question (SCR_Q21) are not asked the second and third screener questions, so they will go to DEP_Q01A. Respondents who say “yes” to the second screener question (SCR_Q22) are not asked the third screener question, so they will go to DEP_Q02. Those who say “yes” to the third screener question (SCR_Q23) will go to DEP_Q09. Finally, those who say no to all three screener questions will go to the start of the suicide sub-block (DEP_B27). DEP_C01A If DODEP = 1, go to DEP_C01B. Otherwise, go to DEP_END. DEP_C01B If SCR_Q21 = 1 (Yes), go to DEP_Q01A. Otherwise, go to DEP_C01C. DEP_C01C If SCR_Q22 = 1 (Yes), go to DEP_Q02. Otherwise, go to DEP_C01D. DEP_C01D If SCR_Q23 = 1 (Yes), go to DEP_Q09. Otherwise, go to DEP_B27. DEP_Q01A Earlier, you mentioned having periods that lasted several days or longer when you felt sad, empty or depressed most of the day. During such episodes, did you ever feel discouraged about how things were going in your life? 1 Yes  2 No (Go to DEP_Q01B) DK (Go to DEP_Q01B) RF (Go to DEP_B27) DEP_Q01A_1 During the episodes of being sad, empty or depressed, did you ever lose interest in most things like work, hobbies or other things you usually enjoyed? 1 Yes 2 No DK, RF Go to DEP_D12 DEP_Q01B During the episodes of being sad, empty or depressed, did you ever lose interest in most things like work, hobbies or other things you usually enjoyed? 1 Yes 2 No DK, RF Go to DEP_D12 DEP_Q02 Earlier, you mentioned having periods that lasted several days or longer    208  when you felt discouraged about how things were going in your life. During such episodes, did you ever lose interest in most things like work, hobbies or other things you usually enjoy? 1 Yes  2 No  DK  RF (Go to DEP_B27) Go to DEP_D12 DEP_Q09 Earlier, you mentioned having periods that lasted several days or longer when you lost interest in most things like work, hobbies or other things you usually enjoy. Did you ever have such a period that lasted for most of the day, nearly every day, for 2 weeks or longer? 1 Yes  2 No (Go to DEP_B27) DK, RF (Go to DEP_B27) DEP_D12 If DEP_Q01A_1 = 1, DT_KEYPHRASE1 = "sad, discouraged or uninterested in things". If DEP_Q01A_1 = 2, DK, RF, DT_KEYPHRASE1 = "sad or discouraged". If DEP_Q01B = 1, DT_KEYPHRASE1 = "sad or uninterested in things". If DEP_Q01B = 2, DK, RF, DT_KEYPHRASE1 = "sad". If DEP_Q02 = 1, DT_KEYPHRASE1 = "discouraged or uninterested in things". If DEP_Q02 = 2, DK, DT_KEYPHRASE1 = "discouraged". If DEP_Q09 = 1, DT_KEYPHRASE1 = "uninterested in things". If DEP_Q01A_1 = 1, DT_KEYPHRASE3 = "sad, discouraged or uninterested". If DEP_Q01A_1 = 2, DK, RF, DT_KEYPHRASE3 = "sad or discouraged". If DEP_Q01B = 1, DT_KEYPHRASE3 = "sad or uninterested". If DEP_Q01B = 2, DK, RF, DT_KEYPHRASE3 = "sad". If DEP_Q02 = 1, DT_KEYPHRASE3 = "discouraged or uninterested". If DEP_Q02 = 2, DK, DT_KEYPHRASE3 = "discouraged". If DEP_Q09 = 1, DT_KEYPHRASE3 = "uninterested". If DEP_Q01A_1 = 1, DT_PROBLEMS = "these problems". If DEP_Q01A_1 = 2, DK, RF, DT_PROBLEMS = "these problems". If DEP_Q01B = 1, DT_PROBLEMS = "these problems". If DEP_Q01B = 2, DK, RF, DT_PROBLEMS = "this problem". If DEP_Q02 = 1, DT_PROBLEMS = "these problems". If DEP_Q02 = 2, DK, DT_PROBLEMS = "this problem". If DEP_Q09 = 1, DT_PROBLEMS = "this problem". If DEP_Q01A_1 = 1, DT_WERE = "were". If DEP_Q01A_1 = 2, DK, RF, DT_WERE = "were". If DEP_Q01B = 1, DT_WERE = "were". If DEP_Q01B = 2, DK, RF, DT_WERE = "was". If DEP_Q02 = 1, DT_WERE = "were". If DEP_Q02 = 2, DK, DT_WERE = "was". If DEP_Q09 = 1, DT_WERE = "was". Processing: NOTE TO DATA USERS: The dynamic text for KEYPHRASE1 and KEYPHRASE3 above is set based on the flow from the screener to the first few depression questions. Each respondent is asked 3 questions up until this point (including the screener), and these three questions will identify the feelings the respondent has. The key words (sad, discouraged or uninterested) encompass every question the respondent said “yes” to up until this point, starting with the first depression screener question. They will be used to describe their    209  episode(s) of depression and will be referenced in many questions after this point. DEP_C12 If DEP_Q09 = 1,  go to DEP_Q16. Otherwise, go to DEP_Q12. DEP_Q12 Did you ever have a period of being ^DT_KEYPHRASE1 that lasted for most of the day, nearly every day, for 2 weeks or longer? 1 Yes  2 No (Go to DEP_B27) DK, RF (Go to DEP_B27) DEP_Q16 Think of periods lasting 2 weeks or longer when ^DT_PROBLEMS with your mood ^DT_WERE most severe and frequent. During those periods, did your feelings of being ^DT_KEYPHRASE3 usually last …? INTERVIEWER: Read categories to respondent. 1 Less than one hour 2 1 hour to less than 3 hours 3 3 hours to less than 5 hours 4 5 hours or more DK, RF DEP_Q17 During those periods, how severe was your emotional distress? INTERVIEWER: Read categories to respondent. 1 Mild 2 Moderate 3 Severe 4 Very severe DK, RF DEP_Q18 During those periods, how often was your emotional distress so severe that nothing could cheer you up? INTERVIEWER: Read categories to respondent. 1 Often 2 Sometimes 3 Rarely 4 Never DK, RF DEP_Q19 During those periods, how often was your emotional distress so severe that you could not carry out your daily activities? INTERVIEWER: Read categories to respondent. 1 Often 2 Sometimes 3 Rarely 4 Never DK, RF DEP_C20 If (DEP_Q17 = 1 (mild) or RF) and (DEP_Q18 = 4 (never) or RF) and (DEP_Q19 = 4 (never) or RF), go to DEP_B27. Otherwise, go to DEP_R21. DEP_R21 People with episodes of being ^DT_KEYPHRASE3 often have other problems at the same time. These include things like feelings of low self-worth and changes in sleep, appetite, energy and ability to concentrate and remember. INTERVIEWER: Press <1> to continue. DEP_Q21 Did you ever have problems like this during one of your episodes of being ^DT_KEYPHRASE3? 1 Yes  2 No (Go to DEP_B27) DK, RF (Go to DEP_B27)    210  DEP_Q22A Please think of an episode of being ^DT_KEYPHRASE3 that lasted 2 weeks or longer when, at the same time, you also had the largest number of these other problems. Is there one particular episode that stands out as the worst one you ever had? 1 Yes  2 No (Go to DEP_Q23A) DK, RF (Go to DEP_Q23A) DEP_Q22A_1 How old were you when that worst episode started? INTERVIEWER: Minimum is 0; maximum is ^DV_AGE. (MIN: 0) (MAX: 130) DK, RF DEP_E22A_1 The reported age is invalid, please return and correct. Rule : Trigger hard edit if DEP_Q22A_1 > DV_AGE. DEP_Q22B How long did it last (in terms of days, weeks, months or years)? INTERVIEWER: If the episode is ongoing, enter how long it has lasted to date. (MIN: 1) (MAX: 900)  DK, RF (Go to DEP_R24) DEP_N22C INTERVIEWER: Was that in days, weeks, months or years? 1 Days 2 Weeks 3 Months 4 Years (DK, RF not allowed) Go to DEP_R24 DEP_E22C An unusual value has been entered. Please confirm or return and change the reporting unit. Rule : Trigger soft edit if (DEP_Q22B > 365 and DEP_N22C = 1) or (DEP_Q22B > 52 and DEP_N22C = 2) or (DEP_Q22B > 24 and DEP_N22C = 3). DEP_E22D The number of years is invalid, please return and correct. Rule : Trigger hard edit if DEP_N22C = 4 and ((DEP_Q22B > (DV_AGE - DEP_Q22A_1)) and (DEP_Q22A_1 <> DK, RF) or (DEP_Q22A_1 = DK,RF and DEP_Q22B > DV_AGE)). DEP_E22E Respondent previously reported the episode lasted at least 2 weeks. The minimum length of time is 2 weeks or 14 days, please return and correct. Rule : Trigger hard edit if (DEP_Q22B = 1 and DEP_N22C = 2) or (DEP_Q22B < 14 and DEP_N22C = 1) DEP_Q23A Think of the last time you had a bad episode of being ^DT_KEYPHRASE3 like this. How old were you when that last episode occurred? INTERVIEWER: Minimum is 0; Maximum is ^DV_AGE. (MIN: 0) (MAX: 130) DK, RF DEP_E23A The reported age is invalid. Please return and correct. Rule : Trigger hard edit if DEP_Q23A > DV_AGE. DEP_C23A If DEP_Q23A = RF, go to DEP_R24. Otherwise, go to DEP_Q23B. DEP_Q23B How long did that episode last? INTERVIEWER: If the episode is ongoing, enter how long it has lasted to date. (MIN: 1) (MAX: 900)  DK, RF (Go to DEP_R24) DEP_N23C INTERVIEWER: Was that in days, weeks, months or years? 1 Days 2 Weeks 3 Months    211  4 Years (DK, RF not allowed) Go to DEP_R24 DEP_E23C An unusual value has been entered. Please confirm or return and change the reporting unit. Rule : Trigger soft edit if (DEP_Q23B > 365 and DEP_N23C = 1) or (DEP_Q23B > 52 and DEP_N23C = 2) or (DEP_Q23B > 24 and DEP_N23C = 3). DEP_E23D The number of years is invalid, please return and correct. Rule : Trigger hard edit if DEP_N23C = 4 and ((DEP_Q23B > (DV_AGE - DEP_Q23A)) and (DEP_Q23A <> DK, RF) or (DEP_Q23A = DK,RF and DEP_Q23B > DV_AGE)). DEP_E23E Respondent previously reported the episode lasted at least 2 weeks. The minimum length of time is 2 weeks or 14 days, please return and correct. Rule : Trigger hard edit if (DEP_Q23B = 1 and DEP_N23C = 2) or (DEP_Q23B < 14 and DEP_N23C = 1) DEP_R24 In answering the next questions, think about the period of 2 weeks or longer when your feelings of being ^DT_KEYPHRASE3 and other problems were most severe and frequent. During that period, tell me which of the following problems you had for most of the day, nearly every day. INTERVIEWER: Press <1> to continue. DEP_Q24A Did you feel sad, empty or depressed most of the day, nearly every day, during that period of 2 weeks or longer? 1 Yes (KEY_PHRASE = feeling sad, empty or depressed) 2 No (Go to DEP_Q24C) DK, RF (Go to DEP_Q24C) DEP_Q24B Nearly every day, did you feel so sad that nothing could cheer you up? 1 Yes (KEY_PHRASE = feeling that nothing could cheer you up) 2 No DK, RF DEP_Q24C During that period of 2 weeks or longer, did you feel discouraged most of the day, nearly every day, about how things were going in your life? 1 Yes (KEY_PHRASE = feeling discouraged about things in your life) 2 No (Go to DEP_Q24E) DK, RF (Go to DEP_Q24E) DEP_Q24D Did you feel hopeless about the future nearly every day? 1 Yes (KEY_PHRASE = feeling hopeless about the future) 2 No DK, RF DEP_Q24E During that period of 2 weeks or longer, did you lose interest in almost all things like work, hobbies and things you like to do for fun? 1 Yes (KEY_PHRASE = losing interest in almost all things) 2 No DK, RF DEP_Q24F Did you feel like nothing was fun even when good things were happening? 1 Yes (KEY_PHRASE nothing was fun) = feeling that 2 No DK, RF Content type: NOTE TO DATA USERS: The condition DEP_C25 is a checkpoint to see if the respondent has said yes to at least one of the Q24 series of questions. If they did not, then they do not meet the criteria for the major depressive episode derived variable and are skipped to suicide sub block (DEP_B27).    212  DEP_C25 If any one of DEP_Q24A, DEP_Q24B, DEP_Q24C, DEP_Q24D, DEP_Q24E or DEP_Q24F = 1 (Yes), go to DEP_Q26A. Otherwise, go to DEP_B27. DEP_Q26A During that period of 2 weeks or longer, did you, nearly every day, have a much smaller appetite than usual? 1 Yes (KEY_PHRASE = having a much smaller appetite)  (Go to DEP_Q26E) 2 No  DK, RF  DEP_Q26B Did you have a much larger appetite than usual nearly every day? 1 Yes much (KEY_PHRASE = having larger appetite) a 2 No DK, RF DEP_Q26C During that period of 2 weeks or longer, did you gain weight without trying to? 1 Yes  2 No (Go to DEP_Q26E) DK, RF (Go to DEP_Q26E) DEP_D26C_1 If SEX_Q01 = 1, DT_PREGNANCY = "empty". If SEX_Q01 = 2, DT_PREGNANCY = " or pregnancy". DEP_Q26C_1 Was this weight gain due to a physical growth^DT_PREGNANCY? 1 Yes (Go to DEP_Q26G) 2 No (KEY_PHRASE = gaining weight without trying to) DK, RF  DEP_Q26D How much did you gain? INTERVIEWER: Enter amount only. (MIN: 1) (MAX: 300)  DK, RF (Go to DEP_Q26G) DEP_N26D INTERVIEWER: Was that in pounds or kilograms? 1 Pounds 2 Kilograms (DK, RF not allowed) Go to DEP_Q26G DEP_E26D An unusual value has been entered. Please confirm. Rule : Trigger soft edit if (DEP_Q26D > 100 and DEP_N26D = 1) or (DEP_Q26D > 50 and DEP_N26D = 2). DEP_Q26E Did you lose weight without trying to? INTERVIEWER: If respondent reports being on a diet or physically ill, select “No”. 1 Yes  2 No (Go to DEP_Q26G) DK, RF (Go to DEP_Q26G) DEP_Q26E_1 Was this weight loss a result of a diet or a physical illness? 1 Yes (Go to DEP_Q26G) 2 No (KEY_PHRASE = losing weight without trying to) DK, RF  DEP_Q26F How much did you lose?  INTERVIEWER: Enter amount only. (MIN: 1) (MAX: 300)  DK, RF (Go to DEP_Q26G) DEP_N26F INTERVIEWER: Was that in pounds or kilograms? 1 Pounds 2 Kilograms (DK, RF not allowed) DEP_E26F An unusual value has been entered. Please confirm. Rule : Trigger soft edit if (DEP_Q26F > 100 and DEP_N26F = 1) or (DEP_Q26F >    213  50 and DEP_N26F = 2). DEP_Q26G During that period of 2 weeks or longer, did you have a lot more trouble than usual either falling asleep, staying asleep or waking up too early nearly every night? 1 Yes (KEY_PHRASE = having trouble falling or staying asleep or waking up too early)                           (Go to DEP_Q26I) 2 No  DK, RF  DEP_Q26H During that period of 2 weeks or longer, did you sleep a lot more than usual nearly every night? 1 Yes (KEY_PHRASE = sleeping a lot more than usual)  (Go to DEP_Q26J) 2 No  DK, RF  DEP_Q26I Did you sleep much less than usual and still not feel tired or sleepy? 1 Yes much (KEY_PHRASE = less than usual)  sleeping 2 No DK, RF DEP_Q26J During that period of 2 weeks or longer, did you feel tired or low in energy nearly every day, even when you had not been working very hard? 1 Yes (KEY_PHRASE = feeling tired or low in energy)  (Go to DEP_Q26L) 2 No  DK, RF  DEP_Q26K During that period of 2 weeks or longer, did you have a lot more energy than usual nearly every day? Yes more (KEY_PHRASE = having energy than usual) a lot No DK, RF DEP_Q26L Did you talk or move more slowly than is normal for you nearly every day? 1 Yes (KEY_PHRASE = talking or moving more slowly than normal) 2 No (Go to DEP_Q26N) DK, RF (Go to DEP_Q26N) DEP_Q26M Did anyone else notice that you were talking or moving slowly? 1 Yes 2 No DK, RF Go to DEP_Q26P DEP_Q26N Were you so restless or jittery nearly every day that you paced up and down or couldn't sit still? 1 Yes (KEY_PHRASE = feeling restless or jittery, or couldn’t sit still) 2 No (Go to DEP_Q26P) DK, RF (Go to DEP_Q26P) DEP_Q26O Did anyone else notice that you were restless? 1 Yes 2 No DK, RF DEP_Q26P During that period of 2 weeks or longer, did your thoughts come much more slowly than usual or seem mixed up nearly every day? 1 Yes (KEY_PHRASE = thinking much more slowly than usual)     214  (Go to DEP_Q26R) 2 No  DK, RF  DEP_Q26Q Did your thoughts seem to jump from one thing to another or race through your head so fast you couldn't keep track of them? 1 Yes (KEY_PHRASE = having thoughts race through your head) 2 No DK, RF DEP_Q26R Nearly every day, did you have a lot more trouble concentrating than is normal for you? 1 Yes (KEY_PHRASE = having more trouble concentrating) 2 No DK, RF DEP_Q26S Were you unable to make up your mind about things you ordinarily have no trouble deciding about? 1 Yes (KEY_PHRASE = being unable to make your mind about things) 2 No DK, RF DEP_Q26T Did you lose your self-confidence? 1 Yes (KEY_PHRASE = losing your self-confidence) 2 No DK, RF DEP_Q26U Nearly every day, did you feel that you were not as good as other people? 1 Yes (KEY_PHRASE = feeling not as good as other people) 2 No (Go to DEP_Q26W) DK, RF (Go to DEP_Q26W) DEP_Q26V Did you feel totally worthless nearly every day? 1 Yes (KEY_PHRASE = feeling worthless) 2 No DK, RF DEP_Q26W Did you feel guilty nearly every day? 1 Yes guilty (KEY_PHRASE every day) = feeling 2 No DK, RF DEP_Q26X Did you feel irritable, grouchy or in a bad mood nearly every day? 1 Yes (KEY_PHRASE = feeling grouchy) 2 No DK, RF DEP_Q26Y Did you feel nervous or anxious most days? 1 Yes (KEY_PHRASE nervous or anxious) = feeling 2 No DK, RF DEP_Q26Z During that period of 2 weeks or longer, did you have any sudden attacks of intense fear or panic? 1 Yes (KEY_PHRASE = having attacks of fear or panic) 2 No DK, RF DEP_Q26ZFF Did you feel that you could not cope with your everyday responsibilities? 1 Yes (KEY_PHRASE couldn’t cope with responsibilities) = feeling your you 2 No DK, RF    215  DEP_Q26ZGG Did you feel like you wanted to be alone rather than spend time with friends or relatives? 1 Yes (KEY_PHRASE = wanting to be alone) 2 No DK, RF DEP_Q26ZHH Did you feel less talkative than usual? 1 Yes (KEY_PHRASE = feeling less talkative) 2 No DK, RF DEP_Q26ZII Were you often in tears? 1 Yes (KEY_PHRASE = being often in tears) 2 No DK, RF DEP_Q26AA Did you often think a lot about death, either your own, someone else’s or death in general? 1 Yes (KEY_PHRASE = thinking about death) 2 No DK, RF DEP_Q26BB During that period, did you ever think that it would be better if you were dead? 1 Yes were (KEY_PHRASE better dead) = thinking you 2 No DK, RF DEP_B27 Call Suicide block (SUI). Content type: NOTE TO DATA USERS: The Suicide (SUI) sub-block is triggered here. All respondents in the survey will be sent here, as all will receive the suicide questions. Some respondents will enter the SUI module after they have gone through the previous depression questions, while respondents who do not meet the depression criteria will directly go to the suicide questions (note the flows in the previous questions that send respondents to DEP_B27). DEP_D27A  Programmer: Set count of DEP_D27A = 0. If any of DEP_Q24A through DEP_Q24F = 1 (Yes), DEP_D27A = DEP_D27A + 1. For each 1 (Yes) in DEP_Q26A, DEP_Q26B, DEP_Q26G, DEP_Q26H, DEP_Q26I, DEP_Q26J, DEP_Q26K, DEP_Q26L, DEP_Q26M, DEP_Q26N, DEP_Q26O, DEP_Q26P, DEP_Q26Q, DEP_Q26R, DEP_Q26S, DEP_Q26T, DEP_Q26U, DEP_Q26V, DEP_Q26W, DEP_Q26X, DEP_Q26Y, DEP_Q26Z, DEP_Q26ZFF, DEP_Q26ZGG, DEP_Q26ZHH, DEP_Q26ZII, DEP_Q26AA, DEP_Q26BB, SUI_Q01, SUI_Q04A, SUI_Q06A, DEP_D27A = DEP_D27A + 1. For each 2 (No) in DEP_Q26C_1 and DEP_Q26E_1, DEP_D27A = DEP_D27A + 1. (min 0; max 34) DEP_D27B  Programmer: If SUI_Q01 = 1, KEY_PHRASE = having EXPERIENCE A  If SUI_Q04A = 1, KEY_PHRASE = having EXPERIENCE B If SUI_Q06A = 1, KEY_PHRASE = having EXPERIENCE C DEP_C27 If DEP_D27A >= 5, go to DEP_C28. Otherwise, go to DEP_END. Content type: NOTE TO DATA USERS: The condition DEP_C27 is a checkpoint to see if the respondent has said yes to at least 5 of the key questions above. If they did not, then they do not meet    216  the criteria for the major depressive episode derived variable and are sent to the end of the module. DEP_C28 If SUI_Q01 = 1 (Yes) or SUI_Q02 = 1 (Yes), go to DEP_Q28B. Otherwise, go to DEP_Q28A. DEP_Q28A You mentioned having a number of the problems that I just asked you about. During that episode, how much did your feelings of being ^DT_KEYPHRASE3 and having these other problems interfere with either your work, your social life or your personal relationships? INTERVIEWER: Read categories to respondent. If respondent does not remember the problems, press <Ctrl D> to show the list of situations. 1 Not at all (Go to DEP_Q29A) 2 A little (Go to DEP_Q28C) 3 Some (Go to DEP_Q28C) 4 A lot (Go to DEP_Q28C) 5 Extremely (Go to DEP_Q28C) DK, RF (Go to DEP_Q28C) DEP_Q28B Earlier, you mentioned having a number of problems during the period of 2 weeks or longer when your feelings of being ^DT_KEYPHRASE3 were most frequent and severe. During that episode, how much did your feelings of being ^DT_KEYPHRASE3 and having these other problems interfere with either your work, your social life or your personal relationships? INTERVIEWER: Read categories to respondent. If respondent does not remember the problems, press <Ctrl D> to show the list of situations. 1 Not at all (Go to DEP_Q29A) 2 A little  3 Some  4 A lot  5 Extremely  DK, RF  DEP_Q28C During that episode, how often were you unable to carry out your daily activities because of your feelings of being ^DT_KEYPHRASE3? INTERVIEWER: Read categories to respondent. 1 Often 2 Sometimes 3 Rarely 4 Never DK, RF DEP_Q29A Episodes of this sort sometimes occur as a result of a physical illness or injury or the use of medication, drugs or alcohol. Do you think your episodes of feeling ^DT_KEYPHRASE3 ever occurred as the result of physical causes, medication, drugs or alcohol? 1 Yes  2 No (Go to DEP_Q30A) DK, RF (Go to DEP_Q30A) DEP_Q29B Do you think your episodes were always the result of physical causes, medication, drugs or alcohol? 1 Yes  2 No (Go to DEP_Q30A) DK, RF (Go to DEP_Q30A) DEP_Q29C What were the causes?  INTERVIEWER: Mark all that apply. 01 Exhaustion  02 Hyperventilation 03 Hypochondria     217  04 Menstrual cycle  05 Pregnancy / postpartum  06 Thyroid disease  07 Cancer  08 Overweight  09 Medication (excluding illicit drugs) 10 Illicit drugs  11 Alcohol  12 Chemical Imbalance / Serotonin Imbalance 13 Chronic Pain  14 Caffeine  15 No specific diagnosis  16 Accident / Injury 17 Emotional, social or economic reason 18 Other - Specify (Go to DEP_S29C) DK, RF  Go to DEP_Q30A  DEP_S29C What were the causes? INTERVIEWER: Specify. (80 spaces) DK, RF DEP_E29C A response of "Pregnancy / postpartum" or "Menstrual cycle" is invalid for a male respondent. Please return and correct. Rule : Trigger hard edit if SEX_Q01 = 1 and (DEP_Q29C = 4 or 5). DEP_Q30A Did your episodes of feeling ^DT_KEYPHRASE3 ever occur just after someone close to you died? 1 Yes  2 No (Go to DEP_R31) DK, RF (Go to DEP_R31) DEP_Q30B Did your episodes of feeling ^DT_KEYPHRASE3 always occur just after someone close to you died? 1 Yes 2 No DK, RF DEP_R31 In the next questions, the word “episode” means a period lasting 2 weeks or longer when, nearly every day, you were ^DT_KEYPHRASE3 and you also had some of the other problems we just mentioned. The end of an episode is when you no longer have the problems for two weeks in a row. INTERVIEWER: Press <1> to continue. DEP_Q31 During your life, how many episodes of feeling ^DT_KEYPHRASE3 with some other problems lasting two weeks or longer have you ever had? INTERVIEWER: Minimum is 1; maximum is 901. If respondent answers more than 900 episodes, enter “900”. If respondent answers “More than I can remember”, enter “901”. (MIN: 1) (MAX: 901) DK, RF DEP_C31 If DEP_Q31 = 1, go to DEP_Q37. Otherwise, go to DEP_Q38A. DEP_Q37 Was that episode brought on by some stressful experience or did it happen out of the blue? 1 Brought on by stress 2 Out of the blue 3 Don’t remember DK, RF DEP_Q38A At any time in the past 12 months, did you have an episode lasting 2 weeks or longer when you felt ^DT_KEYPHRASE3 and also had some of the other    218  problems already mentioned? INTERVIEWER: If respondent does not remember the problems, press <Ctrl D> to show the list of situations. 1 Yes  2 No (Go to DEP_C38C) DK, RF (Go to DEP_C38C) DEP_Q38A_1 How recently was it? INTERVIEWER: Read categories to respondent. 1 During the past month 2 Between 1 and 6 months ago 3 More than 6 months ago DK, RF DEP_Q38B During the past 12 months, about how many days out of 365 were you in such an episode? (You may use any number between 1 and 365 to answer.) (MIN: 1) (MAX: 365) DK, RF Go to DEP_C39 DEP_C38C If DEP_Q31 = 1, go to DEP_C39. Otherwise, go to DEP_Q38C. DEP_Q38C How old were you the last time you had one of these episodes? INTERVIEWER: Minimum is 0; Maximum is ^(DV_AGE - 1). (MIN: 0) (MAX: 130) DK, RF DEP_E38C The entered age is invalid. Please return and correct. Rule : Trigger hard edit if DEP_Q38C > DV_AGE - 1. DEP_C39 If DEP_Q31 = 1, go to DEP_C62. Otherwise, go to DEP_Q39. DEP_Q39 What is the longest episode you ever had when, most of the day, nearly everyday, you were feeling ^DT_KEYPHRASE3 and you also had some of the other problems we just mentioned? INTERVIEWER: If respondent does not remember the problems, press <Ctrl D> to show the list of problems. If the episode is ongoing, enter how long it has lasted to date. (MIN: 1) (MAX: 900)  DK, RF (Go to DEP_D53) Content type: NOTE TO DATA USERS: At this point respondents with only one episode will be skipped passed this question and some of the following questions, as data on age and duration has already been collected on that one episode in earlier parts of the module (see DEP_Q22A to DEP_N23C). DEP_N39_1 INTERVIEWER: Was that in days, weeks, months or years? 1 Days 2 Weeks 3 Months 4 Years (DK, RF not allowed) Go to DEP_D53 DEP_E39_1 An unusual value has been entered. Please confirm or return and change the reporting unit. Rule : Trigger soft edit if (DEP_Q39 > 365 and DEP_N39_1 = 1) or (DEP_Q39 > 52 and DEP_N39_1 = 2) or (DEP_Q39 > 24 and DEP_N39_1 = 3). DEP_E39_2 The reported number of years is invalid, please return and correct. Rule : Trigger hard edit if DEP_N39_1 = 4 and (DEP_Q39 > DV_AGE).    219  DEP_E39_3 Respondent previously reported the episode lasted at least 2 weeks. The minimum length of time is 2 weeks or 14 days, please return and correct. Rule : Trigger hard edit if (DEP_Q39 = 1 and DEP_N39_1 = 2) or (DEP_Q39 < 14 and DEP_N39_1 = 1). DEP_D53 If 1 <= DEP_Q31 < 100, DT_SEVERAL = "^DEP_Q31". Otherwise, DT_SEVERAL = "several". DEP_Q53 Earlier, you mentioned that you had ^DT_SEVERAL episode(s) of feeling ^DT_KEYPHRASE3 with some other problems lasting 2 weeks or longer in your life. How many of these episodes were brought on by some stressful experience? INTERVIEWER: Minimum is 0; maximum is ^DEP_Q31. If respondent answers "All of my episodes", enter ^DEP_Q31. (MIN: 0) (MAX: 901) DK, RF DEP_E53 The respondent reported that the number of episodes brought on by a stressful experience is greater than the total number of episodes. Please return and correct. Rule : Trigger hard edit if DEP_Q53 > DEP_Q31) and (DEP_Q31 <> DK, RF). DEP_C62 If DEP_Q38A = 1 (Yes), go to DEP_R64A. Otherwise, go to DEP_C71. DEP_R64A For the next questions, think about the period of 2 weeks or longer during the past 12 months when your feelings of being ^DT_KEYPHRASE3 were most severe and frequent. INTERVIEWER: Press <1> to continue. DEP_Q64A DEP_Q64A During this period, how often did you feel cheerful? INTERVIEWER: Read categories to respondent. 1 Often 2 Sometimes 3 Occasionally 4 Never DK, RF DEP_Q64B How often did you feel as if you were slowed down? INTERVIEWER: Read categories to respondent. 1 Often 2 Sometimes 3 Occasionally 4 Never DK, RF DEP_Q64C How often could you enjoy a good book or radio or TV program? 1 Often 2 Sometimes 3 Occasionally 4 Never DK, RF DEP_Q65A During this period, how often did you still enjoy the things you used to enjoy? INTERVIEWER: Read categories to respondent. 1 As much as usual 2 Not quite as much as usual 3 Only a little 4 Not at all    220  DK, RF DEP_Q65B How often could you laugh and see the bright side of things? INTERVIEWER: Read categories to respondent. 1 As much as usual 2 Not quite as much as usual 3 Only a little 4 Not at all DK, RF DEP_Q65C How often did you take interest in your physical appearance? 1 As much as usual 2 Not quite as much as usual 3 Only a little 4 Not at all DK, RF DEP_Q65D How often did you look forward to enjoying things? 1 As much as usual 2 Not quite as much as usual 3 Only a little 4 Not at all DK, RF DEP_R66 (Please refer to page 3 of the booklet.) Think about the period of time that lasted one month or longer when your feelings of being ^DT_KEYPHRASE1 were most severe in the past 12 months. Please tell me what number best describes how much these feelings interfered with each of the following activities. For each activity, answer with a number between 0 and 10; 0 means “no interference” while 10 means “very severe interference”. INTERVIEWER: Press <1> to continue. DEP_Q66A In the past 12 months, how much did your feelings of being ^DT_KEYPHRASE1 interfere with your home responsibilities, like cleaning, shopping and taking care of the house or apartment? 0 No interference 01 I 02 I 03 I 04 I 05 I 06 I 07 I 08 I 09 V 10 Very severe interference DK, RF DEP_Q66B_1 How much did your feelings interfere with your ability to attend school? INTERVIEWER: If necessary, enter “11” to indicate “Not applicable”. 0 No interference 01 I 02 I 03 I 04 I 05 I 06 I 07 I 08 I    221  09 V 10 Very severe interference 11 Not applicable DK, RF DEP_Q66B_2 How much did they interfere with your ability to work at a job? INTERVIEWER: If necessary, enter “11” to indicate “Not applicable”. 0 No interference 01 I 02 I 03 I 04 I 05 I 06 I 07 I 08 I 09 V 10 Very severe interference 11 Not applicable DK, RF DEP_Q66C Again thinking about that period of time lasting one month or longer during the past 12 months when your feelings of being ^DT_KEYPHRASE1 were most severe, how much did they interfere with your ability to form and maintain close relationships with other people? (Remember that 0 means “no interference” and 10 “very severe interference”.) 0 No interference 01 I 02 I 03 I 04 I 05 I 06 I 07 I 08 I 09 V 10 Very severe interference DK, RF DEP_Q66D How much did they interfere with your social life? 0 No interference 01 I 02 I 03 I 04 I 05 I 06 I 07 I 08 I 09 V 10 Very severe interference DK, RF DEP_C67 If (DEP_Q66A, DEP_Q66B_1, DEP_Q66B_2, DEP_Q66C and DEP_Q66D) = 0 (no interference) or = 11 (not applicable) or DK, RF, go to DEP_C71. Otherwise, go to DEP_Q68. DEP_Q68 In the past 12 months, about how many days out of 365 were you totally unable to work or carry out your normal activities because of your feelings of being    222  ^DT_KEYPHRASE1? (You may use any number between 0 and 365 to answer.) (MIN: 0) (MAX: 365) DK, RF DEP_C71 If SUI_Q16 = 1 (Yes) or SUI_Q19 = 1 (Yes), go to DEP_C87. Otherwise, go to DEP_Q72. DEP_Q72 Did you ever in your life see, or talk on the telephone to, a medical doctor or other professional about your feelings of being ^DT_KEYPHRASE1? (By other professional, we mean psychologists, psychiatrists, social workers, counsellors, spiritual advisors, homeopaths, acupuncturists, self-help groups or other health professionals.) 1 Yes  2 No (Go to DEP_END) DK, RF (Go to DEP_END) DEP_Q86 During the past 12 months, did you receive professional treatment for your feelings of being ^DT_KEYPHRASE1? 1 Yes 2 No DK, RF DEP_C87 If SUI_Q13 = 1 (Yes), go to DEP_END. Otherwise, go to DEP_Q87. DEP_Q87 During your life, were you ever hospitalized overnight for your feelings of being ^DT_KEYPHRASE1? 1 Yes 2 No DK, RF DEP_END    Pain and discomfort (HUP) 1 - Copied Version - 1001  HUP_BEG Content block External variables required: SEX_Q01: sex of specific respondent (1 = male, 2 = female) from Sex block. DOHUP: do block flag, from the sample file. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space HUP_C1 If DOHUP = 1, go to HUP_R1. Otherwise, go to HUP_END. HUP_R1 The next set of questions asks about the level of pain or discomfort you usually experience. They are not about illnesses like colds that affect people for short periods of time. INTERVIEWER: Press <1> to continue. HUP_Q28 Are you usually free of pain or discomfort? 1 Yes (Go to HUP_END) 2 No  DK, RF (Go to HUP_END) HUP_Q29 How would you describe the usual intensity of your pain or discomfort? INTERVIEWER: Read categories to respondent. 1 Mild    223  2 Moderate 3 Severe DK, RF HUP_Q30 How many activities does your pain or discomfort prevent? INTERVIEWER: Read categories to respondent. 1 None 2 A few 3 Some 4 Most DK, RF HUP_END    Positive Mental Health (PMH) 1 - Copied Version - 1001  PMH_BEG This module is the Mental Health Continuum Short Form© instrument developed by Dr. Corey Keyes (Emory University in Atlanta, Georgia USA). The author granted permission to Statistics Canada for the use of MHC-SF in this survey. Content block External variables required: SEX_Q01: sex of specific respondent (1 = male, 2 = female) from Sex block. DOPMH: do block flag, from the sample file. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space PMH_C01 If DOPMH = 1, go to PMH_R01. Otherwise, go to PMH_END. PMH_R01 (Please refer to page 1 of the booklet.) The following questions are about how you have been feeling during the past month. INTERVIEWER: Press <1> to continue. PMH_Q01 In the past month, how often did you feel: ...happy? INTERVIEWER: Read categories to respondent. 1 Every day  2 Almost every day  3 About 2 or 3 times a week 4 About once a week 5 Once or twice  6 Never  DK, RF (Go to PMH_END) PMH_Q02 (In the past month, how often did you feel:) ...interested in life? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q03 (In the past month, how often did you feel:)    224  ...satisfied with your life? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q04 In the past month, how often did you feel: ...that you had something important to contribute to society? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q05 (In the past month, how often did you feel:) ...that you belonged to a community (like a social group, your neighbourhood, your city, your school)? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q06 (In the past month, how often did you feel:) ...that our society is becoming a better place for people like you? INTERVIEWER: If necessary, explain that "people like you" can refer to any groups to which the respondent feels they belong (i.e. religion, income, ethnicity, age, health status, community, etc.). 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q07 In the past month, how often did you feel: ...that people are basically good? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q08 (In the past month, how often did you feel:) ...that the way our society works makes sense to you? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never    225  DK, RF PMH_Q09 (In the past month, how often did you feel:) ...that you liked most parts of your personality? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q10 In the past month, how often did you feel: ...good at managing the responsibilities of your daily life? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q11 (In the past month, how often did you feel:) ...that you had warm and trusting relationships with others? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q12 (In the past month, how often did you feel:) ...that you had experiences that challenge you to grow and become a better person? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF PMH_Q13 In the past month, how often did you feel: ...confident to think or express your own ideas and opinions? 1 Every day 2 Almost every day 3 About 2 or 3 times a week 4 About once a week 5 Once or twice 6 Never DK, RF  Distress (DIS) 1 - Copied Version - 1001  DIS_BEG Content block External variables required: SEX_Q01: sex of specific respondent (1 = male, 2 = female) from Sex block. DODIS: Do block flag, from the sample file. REFDATE: current date from operating system    226  PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space Display DTE1MOAGOE as Month DD, YYYY, e.g. January 2, 2008. DIS_C01 If DODIS = 1, go to DIS_R01. Otherwise, go to DIS_END. DIS_R01 The following questions deal with feelings you may have had during the past month. INTERVIEWER: Press <1> to continue. DIS_Q01A (Please refer to page 2 of the booklet.) During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel: ...tired out for no good reason? INTERVIEWER: Read categories to respondent. 1 All of the time  2 Most of the time 3 Some of the time  4 A little of the time  5 None of the time  DK, RF (Go to DIS_END) DIS_Q01B During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel: ... nervous? 1 All of the time  2 Most of the time 3 Some of the time  4 A little of the time  5 None of the time (Go to DIS_Q01D) DK, RF (Go to DIS_Q01D) Processing: At the time of the data processing, if DIS_Q01B = 5, then DIS_Q01C will be set to 5 (None of the time). DIS_Q01C (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...so nervous that nothing could calm you down? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_Q01D (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...hopeless? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_Q01E During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel: ...restless or fidgety? 1 All of the time  2 Most of the time    227  3 Some of the time  4 A little of the time  5 None of the time (Go to DIS_Q01G) DK, RF (Go to DIS_Q01G) Processing: At the time of the data processing, if DIS_Q01E = 5, then DIS_Q01F will be set to 5 (None of the time). DIS_Q01F (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...so restless you could not sit still? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_Q01G (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...sad or depressed? 1 All of the time  2 Most of the time 3 Some of the time  4 A little of the time  5 None of the time (Go to DIS_Q01I) DK, RF (Go to DIS_Q01I) Processing: At the time of the data processing, if DIS_Q01G = 5, then DIS_Q01H will be set to 5 (None of the time). DIS_Q01H (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...so depressed that nothing could cheer you up? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_Q01I (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...that everything was an effort? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_Q01J (During the past month, that is, from ^DTE1MOAGOE to yesterday, about how often did you feel:) ...worthless? 1 All of the time 2 Most of the time 3 Some of the time 4 A little of the time 5 None of the time DK, RF DIS_C01K If (DIS_Q01A = 5), and (DIS_Q01B = 5, DK, or RF), and (DIS_Q01D = 5, DK, or RF), and (DIS_Q01E = 5, DK, or RF), and (DIS_Q01G = 5, DK, or RF), and (DIS_Q01I = 5, DK, or RF), and (DIS_Q01J = 5, DK, or RF), go to    228  DIS_END. Otherwise, go to DIS_Q01K. DIS_Q01K We just talked about feelings that occurred to different degrees during the past month. Taking them altogether, did these feelings occur more often in the past month than is usual for you, less often than usual or about the same as usual? 1 More often  2 Less often (Go to DIS_Q01M) 3 About the same (Go to DIS_Q01N) 4 Never have had any (Go to DIS_END) DK, RF (Go to DIS_END) DIS_Q01L Is that a lot more, somewhat more or only a little more often than usual? 1 A lot 2 Somewhat 3 A little DK, RF Go to DIS_Q01N DIS_Q01M Is that a lot less, somewhat less or only a little less often than usual? 1 A lot 2 Somewhat 3 A little DK, RF DIS_Q01N During the past month, how much did these feelings usually interfere with your life or activities? INTERVIEWER: Read categories to respondent. 1 A lot 2 Some 3 A little 4 Not at all DK, RF DIS_END    Socio-demographic characteristics (SDC) 1 - Copied Version - 1001  SDC_BEG Content block External variables required: CURRENTYEAR from Entry BIRTHYEAR from Entry DV_AGE: age of selected respondent from AN3 block SEX_Q01: sex of specific respondent (1 = male, 2 = female) from Sex block. FNAME: first name of respondent from household block. DOSDC: do block flag, from the sample file. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space SDC_C01 If DOSDC = 1, go to SDC_R01.    229  Otherwise, go to SDC_END. SDC_R01 Now some general background questions which will help us compare the health of people in Canada. INTERVIEWER: Press <1> to continue. SDC_D01 DV_CNTRYTEXT = (STRING 80) = SDC_Q01 DV_CNTRYCODE = (0..99990) = SDC_Q01 SDC_Q01 In what country were you born? INTERVIEWER: Ask the respondent to specify country of birth according to current boundaries. Start typing the name of the country to activate the search function. Enter (CAN) to select Canada. Enter "Other - Specify" if the country is not part of the list. DK, RF Programmer: Call Trigram Search. Null is not allowed. Don't know and Refusal are allowed. The Search File to be used corresponds to the Excel file "Country_Pays_LookUpList_With_StdrdCodeFinal". The DV_CNTRYCODE and the DV_CNTRYTEXT are the two fields that should be displayed on the pop-up screen when the Search File is called. However, the corresponding DV_CNTRYCODE also needs to be saved and used as the key to indicate exactly which unique entry in the Search File was selected (i.e., it is the code that differentiates between the English, French and other spelling variations of country names). SDC_C02A If DV_CNTRYCODE = 90000 (Other-Specify), go to SDC_S01. Otherwise, go to SDC_C02B. SDC_C02B If DV_CNTRYCODE = 11124 (Born in Canada), go to SDC_Q04. Otherwise, go to SDC_Q02. SDC_S01 In what country were you born? INTERVIEWER: Specify. (80 spaces) (DK, RF not allowed) SDC_Q02 Were you born a Canadian citizen? 1 Yes (Go to SDC_Q04) 2 No  DK, RF (Go to SDC_Q04) SDC_Q03 In what year did you first come to Canada to live? INTERVIEWER: The respondent may have first come to live in Canada on a work or study permit or by claiming refugee status. If the respondent moved to Canada more than once, enter the first year they arrived in Canada (excluding vacation time spent in Canada). If the respondent cannot give the exact year of arrival in Canada, ask for a best estimate of the year. (MIN: 1870) (MAX: 2100) DK, RF SDC_E03A The year that the respondent first came to Canada is in the future. Please return and correct. Rule : Trigger hard edit if SDC_Q03 > CURRENTYEAR. SDC_E03B The year that the respondent first came to Canada is before the year of birth. Please return and correct. Rule : Trigger hard edit if SDC_Q03 < ((CURRENTYEAR - DV_AGE) - 1). SDC_Q04 To which ethnic or cultural groups did your ancestors belong? (For example: French, Scottish, Chinese, East Indian) INTERVIEWER: Mark all that apply. 01 Canadian  02 French  03 English     230  04 German  05 Scottish 06 Irish  07 Italian  08 Ukrainian  09 Dutch (Netherlands)  10 Chinese  11 Jewish  12 Polish  13 Portuguese  14 South Asian (e.g. East Indian, Pakistani, Sri Lankan) 15 Norwegian  16 Welsh  17 Swedish  18 First Nations (North American Indian) 19 Métis  20 Inuit  21 Other - Specify (Go to SDC_S04) DK, RF  Go to SDC_C05  SDC_S04 To which ethnic or cultural groups did your ancestors belong? (For example: French, Scottish, Chinese, East Indian) INTERVIEWER: Specify. (80 spaces) (DK, RF not allowed) SDC_C05 If SDC_Q01 or DV_CNTRYCODE = Canada, United States, Germany or Greenland, go to SDC_Q05. Otherwise, go to SDC_Q07. SDC_Q05 Are you an Aboriginal person that is, First Nations, Métis or Inuit? First Nations includes Status and Non-Status Indians. INTERVIEWER: The terms "First Nations" and "North American Indian" can be interchanged. Some respondents may prefer one term over the other. 1 Yes  (Go to SDC_Q06) 2 No  DK, RF  Go to SDC_Q07  Content type: Tag: Aboriginal group This question should be answered regardless of whether or not this person is an Aboriginal person of North America. Aboriginal people are usually those with ancestors who resided in North America prior to European contact and who identify with one of the three Aboriginal groups listed on the questionnaire:First Nations (North American Indian), Métis and Inuit. Persons who consider themselves to be East Indian or Asian Indian, or who have ethnic roots on the subcontinent of India, should respond "No, not an Aboriginal person" to this question. Individuals who refer to themselves as Métis in the context of mixed ancestry, but who do not have North American Aboriginal ancestry-for example, those from Africa, the Caribbean and South America-should respond "No, not an Aboriginal person". SDC_Q06 INTERVIEWER: If the respondent has already specified the Aboriginal group(s), select the group(s) from the list below; if not, ask: Are you First Nations, Métis or Inuit? INTERVIEWER: Mark all that apply. First Nations (North American Indian) includes Status and Non-Status    231  Indians. The terms "First Nations" and "North American Indian" can be interchanged. respondents may prefer one term over the other.  Some 1 First Nations Indian) (North American 2 Métis 3 Inuit DK, RF Go to SDC_Q08 SDC_E06 You have entered Don't know or Refusal for SDC_Q06. Respondents sometimes get confused with the terminology used to describe different aboriginal groups. If you wish to change the entry, return to SDC_Q06 and enter the appropriate answer. Otherwise, please confirm. Rule : Trigger soft edit if SDC_Q06 = DK, RF. SDC_D07 Not Applicable SDC_Q07 You may belong to one or more racial or cultural groups on the following list. Are you...? INTERVIEWER: Read categories to respondent and mark up to 4 responses that apply. If respondent answers "mixed", "bi-racial" or "multi-racial", etc, probe for specific groups and mark each one separately (e.g. White, Black, Chinese). 01 White  02 South Asian (e.g., East Indian, Pakistani, Sri Lankan) 03 Chinese  04 Black  05 Filipino  06 Latin American  07 Arab  08 Southeast Asian (e.g., Vietnamese, Cambodian, Malaysian, Laotian)  09 West Asian (e.g., Iranian, Afghan) 10 Korean  11 Japanese  12 Other - Specify (Go to SDC_S07) DK, RF  Go to SDC_Q08  Help text: Tag: Racial or cultural group All response categories and examples must be read aloud, even if the respondent has already given the interviewer one response. DO NOT code responses that do not appear on the list of response categories. For example, do not mark "White", if the respondent says "Caucasian". Record "Caucasian" in the "Other-specify" category. SDC_S07 You may belong to one or more racial or cultural groups on the following list. Are you...? INTERVIEWER: Specify. (80 spaces) DK, RF SDC_R13 Now a question about the dwelling in which you live. INTERVIEWER: Press <1> to continue. SDC_Q13 Is this dwelling...? INTERVIEWER: Read categories to respondent. If the respondent’s household contains both owners and renters, such as a boarder, the dwelling should be considered owned. 1 Owned by you or a member of this household, even if it is still being paid for 2 Rented, even if no cash rent is paid    232  DK, RF Help text: Tag: Owned or rented Choose "Owned" if the respondent and/or another member of this household own the dwelling in which they live, even if the dwelling is on rented or leased land, or if it is part of a condominium, or if it is still being paid for by the respondent or another member of this household. Choose "Rented" in all other cases, even if the dwelling occupied by the respondent is provided without cash rent or at a reduced rent (for example, a clergy's residence or a superintendent's dwelling in an apartment building), or the dwelling is part of a co-operative. SDC_END    Education (EDU) 1 - Copied Version - 1001  EDU_BEG Content block External variables required: FNAME[20]: first name(s) of household member(s) from USU block. DOEDU: do block flag, from the sample file. AGE[20]: Age of household member(s) from DM block. SEX[20]: sex of household member(s) (1 = male, 2 = female) from DM block. PE_Q01: first name of specific respondent from USU block PE_Q02: last name of specific respondent from USU block Screen display: Display on header bar PE_Q01 and PE_Q02 separated by a space EDU_C01 If DOEDU = 1, go to EDU_R01. Otherwise, go to EDU_END. EDU_R01 The following questions are about education. INTERVIEWER: Press <1> to continue. EDU_B01 Call sub block "Education of the respondent" (EDU1) EDU_C02 If there is at least one household member who is >= 15 years of age other than the selected respondent, go to EDU_R02. Otherwise, go to EDU_END. EDU_R07 Now I would like you to think about the rest of your household. INTERVIEWER: Press <1> to continue. EDU_B02 Call sub block "Education of other household members" (EDU2) Content type: Ask this block for each household member aged 15 and older other than selected respondent. Maximum of 19 times. Begin with the first person rostered and continue in the order the household was rostered. EDU_END    Education of the respondent (EDU1) 1 - Copied Version - 1001  EDU1_BEG   EDU1_Q01 What is the highest grade of elementary or high school you have ever completed? 1 Grade 8 or Secondary lower (Québec: II or lower)     233  (Go to EDU1_Q03) 2 Grade 9 - 10 (Québec: Secondary III or IV, Newfoundland and Labrador: 1st year secondary)  (Go to  EDU1_Q03) 3 Grade 11 - 13 (Québec: Secondary V, Newfoundland and Labrador: 2nd to 3rd year of secondary)  DK, RF  Help text: Tag: Educational Attainment The attainment of a certificate, diploma or degree is considered to be at a higher level than some post secondary education without a certificate, diploma or degree. EDU1_Q02 Did you complete a high school diploma or its equivalent? 1 Yes 2 No DK, RF EDU1_Q03 Have you received any other education that could be counted towards a certificate, diploma or degree from an educational institution? 1 Yes (Go to EDU1_Q04) 2 No  DK, RF  Go to EDU1_Q05  EDU1_Q04 What is the highest certificate, diploma or degree that you have completed? 1 Less than high school diploma or its equivalent 2 High school diploma or a high school equivalency certificate 3 Trade Certificate or Diploma 4 College, cegep or other non- university certificate or diploma (other than trades certificates or diplomas) 5 University certificate or diploma below the bachelor’s level 6 Bachelor’s degree (e.g. B.A., B.Sc., LL.B.) 7 University certificate, diploma or degree above the bachelor’s level DK, RF EDU1_Q05 Are you currently attending a school, college, cegep or university? INTERVIEWER: Ask respondent to include attendance only for courses that can be used as credit towards a certificate, diploma or degree. 1 Yes (Go to EDU1_Q06) 2 No  DK, RF  Go to EDU1_END  EDU1_Q06 Are you enrolled as…? INTERVIEWER: Read categories to respondent. 1 A full-time student 2 A part-time student 3 Both full-time and part-time student DK, RF EDU1_END    Education of other household members (EDU2) 1 - Copied Version - 1001  EDU2_BEG     234  EDU2_D07 Not Applicable EDU2_Q07 What is the highest grade of elementary or high school ^DT_NAME ever completed? 1 Grade 8 or lower (Québec: Secondary II or lower) 2 Grade 9 - 10 (Québec: Secondary III or IV, Newfoundland and Labrador: 1st year secondary) 3 Grade 11 - 13 (Québec: Secondary V, Newfoundland and Labrador: 2nd to 3rd year of secondary) DK, RF Help text: Tag: Educational Attainment The attainment of a certificate, diploma or degree is considered to be at a higher level than some post secondary education without a certificate, diploma or degree. EDU2_D08 If SEXn = 1, DT_SEX1 = "he". Otherwise, DT_SEX1 = "she". EDU2_C08 If EDU2_Q07 = 1, 2, go to EDU2_Q09. Otherwise, go to EDU2_Q08. EDU2_Q08 Did ^DT_SEX1 complete high school or its equivalent? 1 Yes 2 No DK, RF EDU2_Q09 Has ^DT_SEX1 received any other education that could be counted towards a certificate, diploma or degree from an educational institution? 1 Yes (Go to EDU2_D10) 2 No  DK, RF  Go to EDU2_END  EDU2_D10 Not Applicable EDU2_Q10 What is the highest certificate, diploma or degree ^DT_SEX1 has completed? 1 Less than high school diploma or its equivalent 2 High school diploma or a high school equivalency certificate 3 Trade Certificate or Diploma 4 College, cegep or other non- university certificate or diploma (other than trades certificates or diplomas) 5 University certificate or diploma below the bachelor’s level 6 Bachelor’s degree (e.g. B.A., B.Sc., LL.B.) 7 University certificate, diploma or degree above the bachelor’s level DK, RF EDU2_END    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0377727/manifest

Comment

Related Items