Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Burden and cardiovascular impact of depression in psoriasis and psoriatic arthritis : a population-based… Burns, Lindsay Claire 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_wall_lindsay.pdf [ 803.55kB ]
Metadata
JSON: 24-1.0074243.json
JSON-LD: 24-1.0074243-ld.json
RDF/XML (Pretty): 24-1.0074243-rdf.xml
RDF/JSON: 24-1.0074243-rdf.json
Turtle: 24-1.0074243-turtle.txt
N-Triples: 24-1.0074243-rdf-ntriples.txt
Original Record: 24-1.0074243-source.json
Full Text
24-1.0074243-fulltext.txt
Citation
24-1.0074243.ris

Full Text

   BURDEN AND CARDIOVASCULAR IMPACT OF DEPRESSION IN PSORIASIS AND PSORIATIC ARTHRITIS: A POPULATION-BASED STUDY  by  LINDSAY CLAIRE BURNS B.Sc. (Honours) St. Francis Xavier University, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August, 2013  ? LINDSAY CLAIRE BURNS 2013   ii ABSTRACT  Introduction: Psoriasis (PsC) and psoriatic arthritis (PsA) represent common, lifelong inflammatory diseases of the skin and joints (respectively) with substantial cardiovascular morbidity.  Considering the important deleterious impact that depression can have on disease complications and long-term health outcomes, we investigated the population burden of physician-diagnosed depression among individuals with PsC and PsA and the potential long-term impact of depression on acute myocardial infarction (AMI) risk in this patient population.    Objectives: 1) To evaluate the incidence and prevalence of depression in the context of PsC and PsA; 2) To compare the odds of depression among individuals with PsC and PsA to healthy controls; 3) To determine whether depression is an independent risk factor for AMI in PsC and PsA; and 4) To determine whether depression modifies the risk of AMI associated with PsC and PsA.   Methods: Epidemiologic methods were used to address Objectives 1-4 based on population-based cohorts of PsC and PsA patients compared to age-, sex-, and index date-matched controls.    Results: 1) The incidence of depression among individuals with PsC and PsA was 3.35 per 1,000 person-years (PY) and the corresponding prevalence during a mean observation period of 4.8 years was 24.63%; 2) There was a 16% increased odds of depression among individuals with PsC and PsA compared to controls; 3) Incident depression increased the risk of incident AMI by 75% in PsC and PsA, and; 4) An increased risk of AMI among individuals with PsC and PsA was observed among individuals with prevalent depression (i.e., depression modified the effect).    Conclusion: These population-based data provide evidence for a substantial burden and increased odds of depression in PsC and PsA compared to controls.  Depression was observed to be an independent risk factor for AMI among patients with PsC and PsA.  Moreover, depression acted as an effect modifier for AMI in the context of PsC and PsA, such that PsC and PsA only led to an increased risk of AMI among individuals with depression.  These data underscore the need to actively screen for depression among PsC and PsA patients and closely monitor cardiovascular health in this high-risk group to improve long-term survival.  iii PREFACE  Chapters 3 and 4 of this thesis represent collaborative work intended for publication in peer-reviewed journals.  Details of authors? contributions for Chapters 3 and 4 are as follows:  Chapter 3: Author Contribution: Lindsay C. Burns was responsible for the concept and design, including literature search, extraction of data, analysis and interpretation of results, preparation of manuscript and all revisions.  Drs. Hyon Choi and Jan Dutz were responsible for study concept and design, analysis plan, results interpretation, and critical review of manuscript.  Chapter 4: Author Contribution: Lindsay C. Burns was responsible for the concept and design, including literature search, extraction of data, analysis and interpretation of results, preparation of manuscript and all revisions.  Drs. Hyon Choi and Jan Dutz were responsible for study concept and design, analysis plan, results interpretation, and critical review of manuscript.  Drs. Eric Sayre, Mushfiqur Rahman, and Mary De Vera contributed to the analysis plan.     Ethical approval for this study was granted by the University of British Columbia?s Behavioural Research Ethics Board (certificate number H06-03851).           iv TABLE OF CONTENTS  ABSTRACT???????????????????????????????????????... ii PREFACE????????????????????????????????.????????. iii TABLE OF CONTENTS???????????????.???????????????????. iv LIST OF TABLES???????????????????...?????????????????.. vi LIST OF FIGURES.????????????????????????????????????.. vii LIST OF ABBREVIATIONS????????????????????????????????? viii ACKNOWLEDGEMENTS?????????????????????????????????? xii DEDICATION???????????????????????????????????.???. xiv   CHAPTER 1  INTRODUCTION AND BACKGROUND???????????????????????.???. 1    1.1 Thesis Overview??????????????????????????????????. 1       1.1.1 Research Statement??????????????????????????????? 1       1.1.2 Overview of Thesis Themes and Chapters?????????????????????. 2    1.2 Psoriasis and Psoriatic Arthritis???????????????????????????? 2       1.2.1 Clinical and Pathological Overview of Psoriasis and Psoriatic Arthritis???.???.??? 3       1.2.2 Epidemiology of Psoriasis and Psoriatic Arthritis????????????.???..??? 5    1.3 High Comorbidity Burden in Psoriasis and Psoriatic Arthritis???????????.??.?? 7    1.4 Cardiovascular Morbidity and Mortality in Psoriasis and Psoriatic Arthritis?????.??.?? 8    1.5 Quality of Life in Psoriasis and Psoriatic Arthritis??????????????????.?? 13    1.6 Epidemiology of Depression????????????????????????????? 17    1.7 The Burden of Depression in Chronic and Psoriatic Disease??????????????? 18    1.8 The Cardiovascular Impact of Depression?????????????????????.?? 20    1.9 Shared Biological Pathways in Psoriatic Disease, Depression, and Cardiovascular Disease?. 22    1.10 Rationale and Overview of Studies??????????????.??????????? 29    1.11 Administrative Data Context????????????????????????????. 30   CHAPTER 2  GENERAL METHODS?????????????????????????????????? 32    2.1 Data Source and Study Population??????????????????????????. 32    2.2 Development of Psoriatic Disease and Control Cohorts?????????????????. 32    2.3 Outcomes of Interest???????????????????????.????????? 33    2.4 Covariate Assessment????????????????????..??????????? 34    2.5 Statistical Analyses????????????????????????????????? 35  v CHAPTER 3  BURDEN OF PHYSICIAN-DIAGNOSED DEPRESSION IN PSORIASIS AND PSORIATIC ARTHRITIS: A POPULATION-BASED STUDY???????????????????.????.  36    3.1 Specific Aims and Corresponding Hypotheses????????????????????.? 36    3.2 Methods?????????????????????????????????????? 36       3.2.1 Development of Psoriatic Disease and Control Cohorts?????????.??????. 36       3.2.2 Outcome Assessment???????????????????????.??????.. 37       3.2.3 Covariate Assessment ??????????????????????.???????. 38       3.2.4 Statistical Analyses????????????????????????.??????? 40    3.3 Results?????????????????????????????.??.??????.. 41   CHAPTER 4  CARDIOVASCULAR IMPACT OF DEPRESSION IN PSORIASIS AND PSORIATIC ARTHRITIS: A POPULATION-BASED STUDY??????????????????????????????  45    4.1 Specific Aims and Corresponding Hypotheses????????????????????? 45    4.2 Methods???????????????????????????????.??????. 46       4.2.1 Development of Psoriatic Disease and Control Cohorts?????????.??????. 46       4.6.2 Assessment of Depression???????????????????????????? 47       4.2.3 Assessment of Outcomes???????????????????????.????? 47       4.2.4 Assessment of Covariates??????????????????????.?????? 48       4.2.5 Statistical Analysis?????????????????????????..?????? 50    4.3 Results????????????????????????????????.?????? 51       4.3.1 Results for Fully-Incident Cohort B1?????????????????.??????? 52       4.3.2 Results for Cohort B2 with Prevalent Depression??????????.????.???? 59   CHAPTER 5  DISCUSSION OF RESULTS????????????????????????.??????? 68    5.1 Overview of Key Findings???????????????????..????.?????? 68   CHAPTER 6  CONCLUSIONS?????????????????????????????.???????.. 74    6.1 Implications and Key Findings?????????????????????.??????? 74    6.2 Future Research Directions??????????????????????..??????? 77    6.3 Conclusions?????????????????????????????.??????? 78   REFERENCES????????????????????????????????..????.. 79     vi LIST OF TABLES  Table 3.1 ICD-9 Definitions of PsD Cohort (Definition A), Covariates, and Outcome???? 40 Table 3.2 Characteristics of Exposed and Unexposed Cohorts (Definition A)??????? 42 Table 3.3 Odds of Prevalent Depression between Cohorts??????????????.. 44 Table 4.1 ICD-9 Definitions of PsD Cohorts (Definitions B1 & B2), Covariates, and Outcome 49 Table 4.2 Characteristics of Exposed and Unexposed Cohorts (Definition B1)??????. 52 Table 4.3 Odds of Incident Depression between Cohorts???????????????.. 54 Table 4.4 Univariate Cox Proportional Hazards Models for Risk of Incident AMI?????.. 55 Table 4.5 Fully-Adjusted Cox Proportional Hazards Model for Risk of Incident AMI????. 56 Table 4.6 Risk Factors for Incident AMI according to Incident Depression Status?????. 57 Table 4.7 Risk Factors for Incident AMI according to Sex???????????????. 58 Table 4.8 Risk Factors for Incident AMI according to Age Category??????????? 58 Table 4.9 Characteristics of Exposed and Unexposed Cohorts (Definition B2)??????. 60 Table 4.10 Odds of Incident Depression between Cohorts??????????????? 62 Table 4.11 Univariate Cox Proportional Hazards Models for Risk of Incident AMI????? 63 Table 4.12 Fully-Adjusted Cox Proportional Hazards Model for Risk of Incident AMI???... 64 Table 4.13 Risk Factors for Incident AMI according to Prevalent Depression Status???? 65 Table 4.14 Risk Factors for Incident AMI according to Sex??????????????? 66 Table 4.15 Risk Factors for Incident AMI according to Age Category??????????.. 67            vii LIST OF FIGURES  Figure 1.1 Main Components of the HPA Axis???????????????????????? 26 Figure 3.1 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition A) at Index Date?? 43 Figure 3.2 Age and Sex Characteristics of Control Cohort (Definition A) at Index Date??????. 43 Figure 4.1 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition B1) at Index Date?. 53 Figure 4.2 Age and Sex Characteristics of Control Cohort (Definition B1) at Index Date?????? 53 Figure 4.3 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition B2) at Index Date?.. 61 Figure 4.4 Age and Sex Characteristics of Control Cohort (Definition B2) at Index Date?????? 61                   viii LIST OF ABBREVIATIONS  5-HT  Serotonin ACTH  Adenocorticotropic hormone AIDS  Acquired immunodeficiency syndrome   AMI  Acute myocardial infarction AS  Ankylosing spondylitis  AVP  Arginine vasopressin BC  British Columbia BCLHD  British Columbia Linked Healthcare Database BMI  Body Mass Index BSA  Body surface area CCHS  Canadian Community Health Survey CCI  Charlson Comorbidity Index CDIC  Canadian Drug Identity Codes CHF  Congestive heart failure CI  Confidence interval CIDI  Composite International Diagnostic Interview COPD  Chronic obstructive pulmonary disease CRA  Canada Revenue Agency CRD  Chronic Renal Disease CRH  Corticotrophin releasing hormone  CRP  C-reactive protein CV  Cardiovascular  ix CVA  Cerebrovascular accident CVD  Cardiovascular disease DALY  Disability-adjusted life years  DLQI  Dermatology Life Quality Index DMARD Disease modifying anti-rheumatic drug EQ-5D  EuroQoL 5 Domains fMRI  Functional magnetic resonance imaging FSS  Fatigue Severity Scale GC  Glucocorticoid GPRD   General Practice Research Database  GWAS  Genome-wide association study HAQ  Health Assessment Questionnaire HPA   Hypothalamic-pituitary-adrenal HR  Hazard ratio HRQOL  Health-related quality of life IBD  Inflammatory bowel disease ICAM  Intracellular adhesion molecule ICD  International Classification of Diseases IDO  Indolamine 2,3-dioxygenase IFN  Interferon IL  Interleukin IMT  Intima-media MCP  Human monocyte chemoattractant protein MDD  Major depressive disorder  x MDE  Major depressive episode MSP  Medical Services Plan NMDA  N-methyl-D-aspartate NCS-R  National Comorbidity Survey Replication NE  Norepinephrine  NPHS  National Population Health Survey O&NS  Oxidative & nitrosative  PopData Population Data BC PVD  Peripheral vascular disease  PVN  Paraventricular nucleus  PsA  Psoriatic arthritis  PsC  Psoriasis PsD  Psoriatic disease PY  Person-years QOL  Quality of life RA  Rheumatoid arthritis  RCT  Randomized controlled trial ROS  Reactive oxygen species SES  Socioeconomic status SF-36  Medical Outcome Study 36-Item Short Form Health Survey SSRI  Selective serotonin reuptake inhibitor TCA  Tricyclic antidepressants TGF  Transforming growth factor TNF  Tumor necrosis factor  xi TRYCATs Tryptophan catabolites (along the indolamine 2,3-dioxygenase pathway) US  United States of America VAS  Visual Analogue Scale VCAM  Vascular cell adhesion molecule WHO  World Health Organization YLD  Years lived with disability                    xii ACKNOWLEDGEMENTS   This body of work would not have been possible without the generous support, guidance, and expertise of a team of remarkable researchers and health care professionals.  I offer my sincere gratitude to my co-supervisors, Drs. Jan Dutz and Hyon Choi, together with my supervisory committee, Drs. Tim Lee and Sunil Kalia, for your expert advice, mentorship, and support through every step of this research project.  To my External Examiner, thank you for the excellent questions and feedback during my defense.  I would like to further thank my manuscript co-authors, Drs. Mary De Vera, Eric Sayre, and Mr. Mushfiqur Rahman for your valuable contributions to the design and analysis plan of these studies.     To the UBC Skin Research Training Centre and Canadian Institutes of Health Research, thank you for your generous funding support for my MSc training.  In addition, I would like to acknowledge the valuable support from the Group for the Assessment of Psoriasis and Psoriatic Arthritis, The Arthritis Society, and the Canadian Arthritis Trainee Association for scientific knowledge exchange and professional development activities.     The Arthritis Research Centre of Canada (ARC) has greatly enriched my training and helped to foster my passion for transforming health research into action.  I am greatly indebted to the entire ARC team for their support and friendship.  I would like to thank Dr. John Esdaile for his leadership and vision in shaping a world without arthritis, Shauneen Kellner, Lisa Singh, Brenda Kapusta, and Lenny Kishi for running such a tight administrative and financial ship, Dave Sale for his IT expertise and support, and Christine Basque for sustainable funds development.  Further, I am grateful for the statistical and methodological advice and mentorship shared by Mr. Mushfiqur  xiii Rahman, Drs. Charlie Goldsmith, Eric Sayre, and Antonio Avi?a Zubieta.  I would also like to acknowledge the important contributions of ARC?s Consumer Advisory Board (CAB) who helped to shape my research question and disseminate key findings to the patient community, with special thanks to Otto Kamensek, Neil White, Sheila Kurr, Marilyn Mulldoon, and Lianne Gulka.  Finally, a number of colleagues have greatly enriched my training, including Nikki Koenig, Pam Rogers, Helen Prlic, Aliya Haji, Sharan Rai, and Sennait Yohannes, with a special thank you to my mentors and lifelong collaborators, Drs. Mary De Vera and Vidula Bhole for their life, career planning, methodological, and statistical advice.      To my husband, Shane Burns, thank you for your unwavering support, love, friendship, and encouragement.  To my oldest friend, Abena Green, thank you for fostering my creativity and skepticism.  To my parents, Robert and Isabel Wall, thank you for a lifetime of love and support, for demonstrating the joy of helping others, and for encouraging me to never stop asking questions.             xiv   To Shane Burns                1 CHAPTER 1 INTRODUCTION AND BACKGROUND  1.1 Thesis Overview  1.1.1 Research Statement  The goals of this thesis are to gain a better understanding of the population burden (i.e., incidence and prevalence) of depression in the context of psoriatic disease (i.e., psoriasis [PsC] and psoriatic arthritis [PsA]) as well as to determine whether depression impacts the risk of acute myocardial infarction (AMI) among individuals with psoriatic disease.  PsC and PsA are chronic inflammatory diseases primarily affecting the skins and joints, respectively.1-5  In recent years, PsC and PsA have been increasingly recognized as multi-system diseases in which excess inflammation can lead to pathology in multiple organs and drive substantial cardiovascular morbidity and mortality.6-34  Preliminary evidence suggests that psychiatric morbidity such as depression is also common among individuals with psoriatic disease,35-37 which may be driven by pain,38 stigma,39 reduced quality of life,40-48 and inflammation.49-59  Given that depression contributes to pain,38,60 obesity,61 fatigue,62,63 poor treatment adherence,64,65 unhealthy behaviours,66,67 reduced health-related quality of life,68 and represents the leading cause of disability worldwide,69 its burden in psoriatic disease is of major clinical interest.  Furthermore, since depression has been shown to be an independent predictor of cardiovascular disease and mortality,70-75 it is possible that depression independently contributes to and/or modifies the excess risk of cardiovascular morbidity observed in psoriatic disease.  This study employs robust epidemiologic methods and leverages the particular strengths of a large, population-based  2 database to examine the potential burden and long-term cardiovascular impact of depression in psoriatic disease, of key interest to researchers, clinicians, patients, and policy makers.    1.1.2 Overview of Thesis Themes and Chapters  This thesis aims to address two major themes concerning depression morbidity in psoriatic disease.  Theme 1 is Burden of Depression in Psoriatic Disease, and addresses the scope of the problem, including population prevalence and odds of depression among individuals with PsC and PsA compared to controls.  Theme 2, Cardiovascular Impact of Depression in Psoriatic Disease focuses on the long-term effects of that burden, and examines whether a comorbid diagnosis of depression is an independent risk factor and effect modifier for cardiovascular disease in PsC and PsA patients.      1.2 Psoriasis and Psoriatic Arthritis   PsC and PsA represent lifelong immune-mediated diseases of the skin and joints, respectively.  In addition to considerable pain, discomfort, and social stigma, individuals with PsC and PsA experience a substantial burden of cardiovascular (CV) morbidity and mortality, and emerging evidence suggests a high rate of psychiatric morbidity as well.  The following sections review the clinical presentation and epidemiology of PsC and PsA, with a focus on the potential underlying mechanisms and impacts of both CV and psychiatric morbidity experienced by individuals with PsC and PsA.      3 1.2.1 Clinical and Pathological Overview of Psoriasis and Psoriatic Arthritis   PsC is a chronic, inflammatory disease primarily affecting the skin that constitutes one of the most common immune-mediated disorders worldwide.76  Although numerous disease variants exist, the most common form of PsC is psoriasis vulgaris, which accounts for 90% of cases and is characterized by a localized and systemic pro-inflammatory state with hyperproliferation of skin cells (called keratinocytes) in the epidermis, the superficial cell layer in the skin.1,76  This rapid turnover of keratinocytes leads to a build-up of skin cells resulting in erythematous papules and well-delineated salmon-coloured plaques with silvery scales.  PsC plaques tend to present on the extremities, especially the elbows and knees, with the trunk, genitals, nails, and scalp also commonly affected.1  The itchy discomfort caused by the plaques in addition to social stigma experienced by psoriatic patients in reaction to their skin lesions have been shown to have a profound impact on health-related quality of life (QOL).44,77,78   PsA was originally characterized by Moll and Wright as an inflammatory arthritis, seronegative for rheumatoid factor, occurring in the presence of PsC.79  Studies have estimated that approximately 30% of individuals with PsC progress to subsequently develop PsA.4,5,79-84  PsA is associated with progressive and destructive inflammation of the joints, entheses, and periarticular connective tissue leading to significant pain and disability.3,4  PsA tends to be an aggressive disease, with the majority of patients developing erosive bone changes and approximately 20% progressing to a highly deforming and disabling condition.4  While there is considerable overlap in the genetic, immunological, clinical, and presentation of PsC and PsA as well as their epidemiology, debate continues over whether PsA represents a severe manifestation of PsC or a genetically and immunologically distinct condition.1  4  Whereas in the past, PsC was primarily understood as a disorder of abnormal epidermal differentiation leading to a hyperproliferation of keratinocytes, evidence over the past two decades suggests a strong immune component driving the pathogenesis of PsC.  Indeed, PsC lesions have been characterized by a proliferation and activation of Th-1, Th-17, and Th-22 lymphocytes and associated pro-inflammatory markers and cytokines such as tumor necrosis factor (TNF)-?, interferon (IFN)-?, C-reactive protein (CRP), intracellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), E-selectin, interleukin (IL)-6, IL-17, and IL-22.16,76,85-89  Further, an expansion of dendritic cells, monocytes/macrophages, and natural killer cells is additionally thought to play a role in the pathogenesis and maintenance of PsC.76,85,86,88  The pro-inflammatory response observed in PsC is more than skin deep: many of the inflammatory markers observed in skin lesions are also found circulating in the bloodstream, with more systemic inflammation occurring in more severe skin disease.90  PsA has been similarly shown to have a major immune component including several pathways shared with PsC, as evidenced by mononuclear cell infiltration into the synovium and entheses, activation of T cell populations (especially natural killer, Th1, and Th17 cells), a corresponding overabundance of pro-inflammatory cytokines, especially TNF-?, IL-1?, IL-6, IFN-?, and IL-17, and a resulting proliferation and activation of synovial and epidermal fibroblasts.4,91-93  PsA is also associated with considerable bone remodeling, with frequent joint erosion in addition to the development of syndesmophyte and bony articulations that can lead to ankylosis.94  Excess TNF-? and other pro-inflammatory cytokines in the synovium increase the differentiation and activity of osteoclasts leading to excess bone resorption, while the altered pathways underlying abnormal bone formation remain poorly understood.91,94  In addition to the discomfort and stigma experienced by PsC patients, the added burden of arthritis contributes to an even lower quality of life among in PsA patients.47   5 1.2.2 Epidemiology of Psoriasis and Psoriatic Arthritis  North American prevalence estimates of PsC generally range from 1 to 3% of the general population.95-98  Studies have shown that the prevalence of PsC varies according to geographical region, with various studies including a recent systematic review of the global burden of psoriasis demonstrating that psoriasis is substantially more common in countries further from the equator.99-101  Overall, global prevalence estimates of PsC have ranged from as low as 0% in Samoa to as high as 8.5% in Norway and 11.8% in Kazach?ye, an Arctic region of the former USSR.99-102  The burden of PsC may also vary according to ethnicity, with higher prevalences having been reported among whites than blacks.95,98,101,103  The prevalence of PsC is also low among ethnic Japanese, Chinese, and Indians, and may be absent in aboriginal Australians, South Americans, and Samoans.45,100,101  South American estimates of PsC prevalence have ranged between 1.3 to 4.2%, and African estimates have varied considerably with an approximately 3% prevalence in dry, rainless countries compared to approximately 0.5% in hot, humid countries.101   Corresponding estimates for the incidence of PsC have ranged from 78.9 to 100.5 per 100,000 person years (PY) in the United States (US)99,104, 230 per 100,000 PY in Italy, 99 130 per 100,000 PY in the Netherlands,105 and 140 per 100,000 PY in the UK.106  The cumulative incidence of PsC is highly variable across African countries, with estimates ranging from 0.1 to 5.0%.103  PsC has a bi-modal peak onset, with the estimated first peak occurring between the ages of 16 to 22 years and the second between the ages of 57 to 60 years,107,108 with early onset psoriasis tending to have more heritability107,109 and a more severe clinical course.109  PsC has been found to occur in equal frequencies among males and females for both early and late onset patients.102,108   The lifetime prevalence of PsA among individuals with PsC is approximately 30%, although estimates have ranged widely from 5-40% depending on study population, length of  6 follow-up, and differences in operational case definitions.4,5,79-84  On average, PsA onset occurs about 10 year following the development of PsC.110  The incidence of PsA among individuals with PsC has been shown to remain constant following the initial diagnosis of PsC at 74 cases per 1,000 PY.84  PsA prevalence and incidence have also been estimated in the general population.  A recent systematic review of international PsA prevalence and incidence found the lowest prevalence of PsA occurring in Japan (0.001% of the population) and the highest in Italy (0.42%), with US estimates ranging from 0.1 to 0.25% of the general population.83  Corresponding estimates for PsA incidence have ranged from a low of 0.1 cases per 100,000 PY in Japan to a high of 23.1 cases per 100,000 PY in Finland, with a median of 6.4 cases per 100,000 PY.83,111  Risk factors for the development of PsA among individuals with PsC include increased disease activity as measured by affected body surface area (BSA)6,84 and obesity/higher body mass index (BMI).84,112   The etiological underpinnings of PsC and PsA are quite complex, with genetic, environmental, and behavioural factors playing a role.  Numerous studies have demonstrated a genetic component to psoriatic disease risk.  A study of 61 monozygotic twin pairs by Faber et al. demonstrated a high concordance rate in PsC presentation, underscoring the importance of genetic factors.113  More recently, a number of large scale genome-wide association studies (GWAS) have identified a many genetic loci as potential susceptibility regions for PsC and PsA, although these genes still fail to account for a large proportion of PsC and PsA heritability.  While the most well-known and widely-replicated susceptibility locus for PsC and PsA is HLA-C, a number of recent GWAS have identified > 20 potential susceptibility loci.114,115   Beyond genetic influences, a number of environmental triggers have been shown to contribute to the development of PsC and PsA.  Low levels of vitamin D exposure is one mechanism thought to underlie the increased prevalence PsC and PsA in areas further from the equator due to limited sun exposure.116,117  Further, studies have shown that UV light stimulates  7 immunosuppressive cytokines and reduces the numbers of mast and T cells in the skin, which is why narrowband UVB therapy is often used as a treatment for moderate to severe psoriasis.118,119  Stressful life events have also been shown to contribute to PsC and PsA risk: one study found that compared to individuals in the lowest quintile of stressful life events, those in the highest quintile of stressful life events had over twice the risk of developing PsC.120  Further, history of trauma and both streptococcal and HIV infections have been shown to increase the future risk of PsC.116  Behavioural factors have also been implicated in epidemiological studies, with cigarette smoking,12,24,121-123 poor diet,117 physical inactivity and obesity,12,112,124-126 and excessive alcohol19,35,122 intake showing associations with the risk of developing PsC and PsA.   1.3 High Comorbidity Burden in Psoriasis and Psoriatic Arthritis  Although previously considered to be limited in pathology to the skin and the joints, PsC and PsA are increasingly being recognized as multi-system diseases, with frequent inflammatory comorbidities and multiple organ involvement.6,86,127-130  It is increasingly recognized that individuals with PsC and PsA have a shortened life expectancy,7,10,11,17,22,26,30,131 markedly increased rates of several adverse health conditions, including cardiovascular disease,6,31,86,127,128,132 metabolic syndrome,86,127 diabetes,6,86,133 gastrointestinal disease,6,19,31,129,132 other autoimmune disease,6,31,130 and neuropsychiatric disorders.19,31,35,37,134  The broad pathology associated with PsC and PsA beyond the skin and joints has led some investigators to propose a new disease classification that better encompasses the scope of their systemic inflammatory nature and substantial comorbidity burden, i.e., psoriatic disease (PsD).129  Below, we describe the literature on a critical outcome in PsD: cardiovascular disease.   8 1.4 Cardiovascular Morbidity and Mortality in Psoriasis and Psoriatic Arthritis   Cardiovascular disease (CVD) represents the leading cause of death in Canadian adults.135  A growing body of epidemiological evidence supports a strong association between PsD and cardiovascular (CV) risk factors (such as smoking, hypertension, and the metabolic syndrome), subclinical CV disease (such as atherosclerosis and intima-media thickening), and CV events (such as AMI, coronary artery disease [CAD], stroke, and peripheral vascular disease [PVD]) and CV mortality.12,29,31,86  CV Risk Factors in PsC and PsA   A number of studies have demonstrated elevated frequencies of CV risk factors in both PsC and PsA settings.  A recent Spanish outpatient case-control study of consecutively-recruited PsC patients (N=72) and healthy controls (N=61) found that PsC patients were more likely to be obese, in addition to showing signs of sub-clinical CV disease.  Notably, over one third of PsC patients had carotid atheroma plaques vs. 8% among controls.  Further, the metabolic syndrome was observed in 40% of PsC patients vs. 13.1% of controls.127  Another single-centre, cross-sectional German study of PsC patients (N=133) found that almost 40% had hypertension and over 20% had hypercholesterolemia.128  Larger-scale studies have also found similar associations with CV risk factors in PsC.  One such study examined heart disease and its risk factors in 3,236 PsC patients vs. 2500 controls using a large US Veterans? Affairs (VA) hospital database, in which the authors found PsC to be associated with an almost threefold increased odds of hypertension, dyslipidemia, and smoking.24  An Italian case-control study compared the rates of CV risk factors among new-onset cases of PsC (N=560) to other recently diagnosed skin conditions (N=690), and  9 found that PsC patients had a 70-90% increased odds of smoking, a 60% increased odds of being overweight, and a 90% increased odds of being obese.120  Prey and colleagues recently conducted a systematic review of case-control studies to summarize the evidence of CV risk factors in PsD in which they found increased odds of metabolic syndrome and obesity, with ORs ranging from 1.3-5.9 and 1.2-5.5, respectively, but less consistent associations with hypertension and dyslipidemia.23    PsA also appears to be associated with an increased burden of CV risk factors compared to healthy controls, as evidenced by a Chinese hospital-based case-control study that compared 102 consecutive PsA cases to 82 hospital staff.  The authors found that body mass index (BMI) was higher in PsA than controls, and after adjustment for BMI, PsA remained associated with increased odds of hypertension, elevated blood pressure, dyslipidemia, insulin resistance, and inflammatory markers.27  Another study found that PsA patients not only had 2.7 times the odds of the metabolic syndrome compared to age- and sex-matched population controls, but that consecutively-recruited PsA patients (N=109) had a higher prevalence of the metabolic syndrome than patients with rheumatoid arthritis (RA, N=699) or ankylosing spondylitis (AS, N=122).136  It is possible that activity limitation or even increased inflammatory activity associated with the additional burden of arthritis in PsA could contribute to an even higher burden of CV risk factors than that observed in PsC, although few studies have directly compared PsC to PsA patients.  One such large cross-sectional study found that obesity was more prevalent among individuals with PsA than PsC, and that both are associated with more obesity than that observed in general population settings.126  These findings were corroborated by a US population-based cohort study that found that obesity increases the future risk of PsA among patients with PsC as well as the general population.112  Finally, a Canadian study that compared CV risk factors among 449 PsC and 611 PsA patients recruited from PsC and PsA clinics found an over twofold increased odds of  10 hypertension in PsA vs. PsC, controlling for relevant clinical covariates.19  These risk factors in PsA translated to an increased risk of intima-media (IMT) wall thickness, an intermediate outcome measure for CVD, in a study of 47 PsA patients compared to 100 healthy controls.21  Further, a large US health claims database found that PsA was associated with a 40% increased prevalence of atherosclerosis, another intermediate CV outcome, compared to healthy controls.18    CV Outcomes in PsC and PsA  The increased prevalence of CV risk factors among patients with PsD (either PsA or PsC) appears to translate to an increased risk of adverse CV outcomes as well.  An analysis of the US health-claims database discussed above demonstrated that PsA was associated with downstream critical CV outcomes as well, with PsA patients showing a 30% increased prevalence of ischaemic heart disease (IHD), 60% increased prevalence of PVD, 50% increased prevalence of congestive heart failure (CHF), and 30% increased prevalence of CVD.18  Cardiovascular morbidity was also assessed using data from an outpatient clinic-based Canadian cohort of PsA patients (N=648).  Compared to general population estimates (from the Canadian Community Health Survey [CCHS]), individuals with PsA had increased prevalences of MI (prevalence ratio [PR] = 2.57), almost double the prevalence of angina and hypertension, but no increased prevalence of CHF or CVA.  Greater disease severity was associated with an increased risk of CV events.17  Several longitudinal cohort studies have evaluated the independent impact of PsC on CV risk.  One such study that employed the General Practice Research Database (GPRD), a UK population-based electronic medical records (EMR) database, found that severe PsC (N=3,603) increased the risk of stroke and AMI (composite outcome) by 53%, independently of known risk factors.131  Another study that employed the GPRD suggested that the risk of AMI in PsC may vary according to disease severity and age.   11 Specifically, 127,139 patients with mild PsC and 3,837 patients with severe PsC were compared to age- and sex-matched population controls for the risk of AMI.  The authors found that mild PsC was associated with a 54% increased risk of AMI and severe PsC was associated with an over seven-fold increased risk of AMI, after adjusting for available risk factors.  Further, this increased risk varied according to age strata, such that younger PsC patients had greater increases in risk compared to older patients.14  Despite these strong observed associations, not all studies have been able to reproduce findings for an increased risk of CVD in PsC.  Data from a Dutch population-based cohort study of 15,820 PsC patients vs. matched controls failed to support an increased risk of heart disease (AMI, angina, and other IHD), with PsC patients showing only a non-significant trend toward a 10% increased risk of IHD, which lost even marginal significance following adjustment for known risk factors.137  However, as noted by Gelfand and colleagues in their accompanying commentary, potential misclassification of exposures and a large proportion of patients with very mild disease could have biased estimates toward the null.138  Nevertheless, Bayesian logic would suggest that such findings should be interpreted in light of converging epidemiological and basic research supporting an increased risk of CVD in PsC.  To that effect, a recent meta-analysis of seven cohort studies found that PsC was significantly associated with a 20% increased risk of stroke and MI (as a composite outcome), independent of conventional CV risk factors.32  In addition to CV outcomes, venous thromboembolism (VTE, including pulmonary embolism [PE] and its precursor, deep vein thrombosis [DVT]) has also been shown to be elevated in PsC.  A recent Swedish hospital-based study found a fivefold increased risk of pulmonary embolism (PE) in the year following hospitalization for PsC, and a 40% increased risk overall.34  However, selection bias could have potentially confounded these results by affecting the co-prevalence of PsC and CV risk factors associated with hospitalization.  Another Danish cohort study extended these findings to the population level, and found that individuals with mild PsC had  12 a 35% increased risk and severe PsC (defined by hospitalization) had two times the future risk of VTE (as a composite outcome) compared to healthy controls, adjusting for known risk factors.9    Mortality Impact of CVD in PsC and PsA   Beyond the increased risk of CVD overall, several cohort studies have demonstrated an increased risk of fatal CVD in both PsC and PsA.  A population study of adults with severe PsC vs. matched controls using the GPRD found that exposed patients had a 57% increased risk of death from CVD, tended to live 6 years less on average, and that CVD was the largest contributed to excess death in this patient population.7  Another study that used the GPRD found that patients with mild PsC did not have an increased risk of all-cause mortality whereas those with severe disease had a 40% increased risk, independent of known risk factors.15  Mehta and colleagues extended these data to specifically evaluate the risk of CV-mortality among patients with severe PsC in the GPRD.  Adjusting for known risk factors, severe PsC increased the risk of CV mortality by 57%, with greater relative risks (RRs) observed in younger age strata.22  One population-based Danish cohort study of PsD patients with mild disease (N=34,371) and severe disease (N=2,621) found that the composite risk of AMI, stroke, and CV death was increased by 20% and 58%, respectively, and that risks were similar between PsC and PsA patients.8  Rates and cause of death were also described in a Canadian cohort of 428 PsA patients followed for 15 years through an outpatient clinic.  Females were found to have a 59% increased risk of all-cause mortality and males, a 65% increased risk, compared to observed rates in the general population.  Circulatory diseases constituted the leading cause of death in PsA patients at 36.2%.30  This increased risk of death in PsD has not been identified in all instances.  A single-centre study of PsA patients in the UK (N=453) found that the observed risk of mortality was similar to general population estimates,  13 although CVD was the leading cause of death.11  Another major cohort study based on 30 year follow-up of a clinical trial of PUVA (UV phototherapy) for severe PsC found an only modestly (10%) increased risk of death compared to general population estimates.26  However, both of these studies lacked a comparable reference group, would have suffered from selection bias (especially in the case of the RCT), and employed differential, non-standard reporting of mortality data between exposed and unexposed cohorts, all of which could have contributed to the null results.  Further, the PUVA RCT-based cohort study failed to replicate well-established findings of an increased prevalence of obesity in PsA, further bringing the external validity of the sample into question.  In light of all epidemiological evidence, it appears highly likely that the PsD is associated with an increased risk of CV risk factors in addition to both nonfatal and fatal CV events.  1.5 Quality of Life in Psoriasis and Psoriatic Arthritis    Given the unpredictable nature of disease activity, the highly visible and itchy skin lesions, the pain of nail and joint disease, and the associated disease complications found in PsC and PsA, it is not surprising that these diseases could greatly impact one?s QOL in important ways.  Indeed, research has shown that the impact of PsD on QOL is quite profound, with PsD affecting many important domains for healthy physical and mental functioning.  The wide-ranging impact of PsC on QOL was highlighted in a systematic review of 17 articles, which found that patients with PsC often reported physical discomfort, emotional distress, a negative body- and self-image, challenges with engaging in their usual activities of daily living, self-consciousness over exposed skin, and limitations at work, with younger age and severe disease predicting more QOL impairment.77  High rates of fatigue among PsD patients was investigated in a telephone survey of 420 PsD patients registered with the US National Psoriasis Foundation (NPF), which found that pain and itchiness in  14 psoriatic lesions in addition to the emotional impact of disease predicted sleep disturbances.  PsA was associated with over three times the odds of sleep disturbances than PsC alone, likely due to increased pain associated with the condition.40  In terms of general disability, another study found that among 17,434 patients who responded to a mailed NPF questionnaire, 79% reported that PsC had a negative impact on their lives and 40% were frustrated with the ineffectiveness of their treatment.44   Self-consciousness associated with skin lesions is another important factor in PsD QOL.  In addition to commonly affected regions like the knees, elbows, and scalp, unsightly and itchy PsC lesions can affect the genital region, and one third of individuals under 55 years old who responded to the large NPF questionnaire reported that the disease negatively affected their sexual activities.44  Other studies have also found that impaired sexual activity is an important aspect of reduced QOL for many patients, with one study reporting impairment in 41% of respondents, particularly among those with joint pain.  Importantly, impaired sexual functioning was associated with increased depression scores.42  Shame regarding physical appearance can also have major impacts on social functioning and activity limitation as well, with studies finding that 81% of PsD patients report feeling embarrassment and shame, 75% reporting feelings of sexual unattractiveness and undesirability, and high rates of social withdrawal including limiting previously enjoyable social activities that involve skin exposure, such as swimming, sports, and sunbathing.46  Furthermore, patients with PsD often report feeling stigmatized due to misperceptions that PsC lesions are communicable, and sometimes face outright discrimination.  In fact, 40% of respondents to the large NPF questionnaire reported that they had received unequal service at hair salons, barbershops, public pools, and health clubs.44  Another study found that over one quarter of PsC patients reported at least one episode in the previous month in which they noticed people making a conscious effort not to touch them, which was uncorrelated with disease activity.  Sadly,  15 those who perceived such stigma reported higher depression scores.139  A clinic-based study of 115 PsC patients also found high rates of perceived stigmatization that strongly predicted psychological distress and disability.  Interestingly, psychological factors were even stronger determinants of disability than disease severity, location of lesions, or disease duration in this sample.140  A recent psychological study by Kleyn et al demonstrated that stigmatization may be experienced so frequently by PsC patients that they show altered cognitive processing as a coping mechanism.  Feelings of disgust, or observations of disgusted faces, are associated with activation of the insular cortex in the brain.  Thirteen male, right handed patients and matched controls were shown faces of disgust vs. neutral faces while under functional magnetic resonance imaging (fMRI) to observe insular cortex activation, and additionally completed a timed decision-making task, the Facial Expression Recognition Task (FERT), to determine whether they had impairment in recognizing faces of disgust.  Not only did patients with PsC demonstrate less bilateral activation of the insular cortex in response to faces of disgust compared to controls, but their ability to differentiate between faces of disgust vs. neutral expressions on FERT was impaired.39  These data suggest a natural coping strategy by which PsC patients have habituated to observing reactions of disgust, which over time may cease to elicit an emotional reaction.  Whether PsD individuals with comorbid depression or anxiety fail to show such habituation to disgusted reactions remains to be seen.   Given the major impact that PsC and PsA can have on physical, mental, and social functioning, it follows that the associated QOL burden would be substantial.  A study by Langley et al. found that patients with PsC had comparable reductions in QOL to that seen in other chronic diseases including ischaemic heart disease (IHD) and diabetes.  Many patients reported that stress was an important trigger or exacerbator of their disease, and that PsC patients were willing to give up an average of 3.8 out of 35 remaining life years to live disease-free.45  Similarly, another study  16 found that impairment in both mental and physical functioning in PsC is comparable to other major diseases such as cancer, arthritis, hypertension, and heart disease, and that PsC patients had the worst scores among all studied conditions on the Short Form-36 (SF-36) QOL questionnaire.141  Further, PsA has been shown to have a comparable142 or even more severe43 impact on disability and QOL as that observed in RA, with PsA demonstrating a larger burden with respect to emotional disturbances and body pain.43  When compared to a gamut of other skin diseases, PsC has been shown to play the largest impact on QOL.48   The additional pain and disability of arthritis in PsA may contribute to even more impairment in QOL than that observed in PsC.  A Swedish study of skin diseases found that PsA was associated with worse QOL on the SF-36 and a visual analogue scale (VAS) than PsC alone.143  In a Canadian cohort study of risk factors for PsA among subjects with PsC (N=402), subjects completed a battery of QOL measures at baseline including the SF-36, Health Assessment Questionnaire (HAQ), EuroQoL 5 domains (EQ-5D), Dermatology Life Quality Index (DLQI), and Fatigue Severity Scale (FSS).  The investigators found that PsA was associated with more QOL impairment on all studied measures with the exception of the skin-specific DLQI.47  A large sample (N=6,497) of dermatologist-confirmed PsD patients from the Nordic Psoriasis Association also completed QOL and disability questionnaires, with PsA patients demonstrating reduced QOL, longer disease durations, and greater PsC severity than those with PsC alone.144  Emotional disturbances and mental health concerns may be a driving force in reduced QOL among PsD patients.  Indeed, a Greek hospital-based study found that anxiety and concern about physical symptoms associated with PsA were independently associated with health-related QOL (HRQOL), in addition to pain and PsC lesion severity,145 and cross-sectional study that screened for depression in PsC found that depression was associated with HRQOL in a graded manner.146   17 1.6 Epidemiology of Depression   The Diagnostic Statistical Manual of Mental Disorders (DSM)-IV defines major depressive disorder (MDD) by at least one core symptom of depressed mood (anhedonia and extreme sadness) or lack of interest, together with at least two of the following symptoms for a minimum of two weeks duration: feelings of worthlessness or guilt, impairments in concentration or decision-making, fatigue, decreased psychomotor activity or agitation, insomnia or hypersomnia, an increase or decrease in weight or appetite, and thoughts of death or suicidality.  Other related depressive disorders include minor depression (lower symptom severity) and dysthymia (low grade depression over an extended period of time).147   The World Health Organization reports that depression constitutes the leading cause of disability worldwide according to years lived with disability (YLD) and the third leading cause of disability as measured by disability-adjusted life years (DALYs) lost.69  The lifetime, one year, and point prevalences of a major depressive episode in Canada have been estimated using the CCHS as 12.2%, 4.8, and 1.8% of the general population, respectively.148  A nationally-representative sample of US adults evaluated the lifetime and one year prevalences of MDD according to DSM-IV criteria to be 13.2% and 5.3%, respectively,149 while another population-based US study found a lifetime MDD prevalence of 16.2%.150  European epidemiological studies have yielded similar estimates for the disease burden of depression, with a representative sample of 6 European countries yielding a lifetime prevalence of mood disorders of 14%.151  A lower prevalence of depression has often been reported in East Asian countries, although such differences may reflect a western cultural bias in symptom classification and diagnostic thresholds in the DSM-IV.152  The etiological underpinnings of depression are complex, with genetics, environmental factors, and life events playing a role in its development.153-155  Risk factors associated with the  18 development of depression include female sex, with females affected at approximately twice the rate of males,69,156 younger age,150,157 low socioeconomic status or poverty,148,149,151,158 chronic illness, 148,155,159,160 and obesity.61,66,125  Given the stigma that often surrounds mental illness, depression may be largely underreported,161,162 with as little as one quarter of major depression patients seeking care for their condition.  Further, differential symptom reporting or screening bias may explain the increased rates of depression among females over males.163,164  While depression has been previously trivialized as a disorder that?s ?all in a person?s head?, it is now recognized that depression is a complex disorder with a strong inflammatory component,50,56,59,165-171 brain remodeling,165,172 neurochemical aberrations,173,174 and a combination of both somatic and psychological symptoms.147,171,175-177  In addition to the tremendous suffering imposed by symptoms of anhedonia, feelings of hopelessness, lethargy, fatigue, loss of concentration, and disruptions in sleeping and eating, depression can have major consequences on general health and is associated with shortened life expectancy through its effects on inflammation, unhealthy behaviours, and in some cases, suicide.73,159  A discussion of the burden, followed by the CV impact on depression on chronic disease, with particular emphasis on the current study population of psoriatic individuals, follows.  1.7 The Burden of Depression in Chronic and Psoriatic Disease   Depression has been shown to have strong associations with several chronic health conditions, and may negatively alter the course and critical outcomes of comorbid diseases.  Depression is common in major health conditions such as CVD, cancer, arthritis, and chronic renal disease (CRD).160,178-183 In general, depression shows bi-directional associations with pain and associated impairment60 in addition to bi-directional associations with obesity.125  Furthermore,  19 depression is characterized by hopelessness, loss of self-efficacy, and impaired concentration, which can undermine one?s ability to successfully manage his disease.  As such, depression leads to unhealthy behaviours such as poor medical adherence, unhealthy dietary patterns, physical inactivity, excessive alcohol consumption, and smoking.65,67,183,184  Comorbid depression has been shown to increase the frequency of hospitalizations, delay return to work, and lead to more disability, more recurrent events, and increased mortality among individuals with CVD.178  In cancer, depression has been shown to speed progression and worsen symptoms178 and in CRD, depression independently predicts progression to dialysis180 and leads to an increased risk of mortality.72    In light of the severe impact PsD can have on quality of life, there is potential for a large burden of depression in psoriatic disease contexts.  Further, the deleterious impact that depression can have on the symptoms, progression, and critical outcomes of other chronic diseases are obviously pertinent in the context of PsC and PsA as well.  Such an increased burden of mental disorders, particularly depression, has recently been appreciated in PsD.  A large-scale Italian study (N=2400) assessed depressive symptomology using the Centre for Epidemiological Studies ? Depression (CES-D) questionnaire and found that 62% of respondents had evidence of depression.36  A small scale Croatian study (N=70) identified mental disorders in 90% of patients with moderate to severe psoriasis, with depressive disorder being the most common with a point prevalence of 20%.35  Another study of 265 prevalent PsC cases found that 32% screened positively for probable depression on the CES-D.146  Using an outpatient database from Germany, the same authors found that PsC was independently associated with the risk of depression (OR = 1.5).37  Another cohort study employing the GPRD found that PsC independently increased the risk of depression compared to age- and sex-matched healthy controls (HR = 1.44).185  A study of hospitalized patients with PsA found a high prevalence of depression (22%), especially among  20 those with polyarthritis (27%).145  An international study of 319 patients with PsA from various clinics found that 11.6% scored positively for current depression on the Hospital Anxiety and Depression Scale (HADS).41  It is possible that as individuals learn to cope with their PsC or PsA condition, their rate of depression decreases.  One small study of newly diagnosed (less than 2 years disease duration, N=57) vs. established PsA patients (N=108) found that new onset cases had higher age- and sex-adjusted prevalences of depression,132 supporting this hypothesis.    1.8 The Cardiovascular Impact of Depression    A growing body of literature has demonstrated that depression and depressive symptomology are associated with an increased risk for CV events and associated fatalities, in addition to progression and poorer outcomes of CVD among prevalent cases.  The current section reviews the epidemiological literature on the CV impact of depression in general as well as its CV impact in the presence of chronic diseases of relevance to PsC and PsA.    A number of investigators have identified an increased prevalence of depression among patients with CVD.  One nationally-representative sample of US adults (N=3032) found that almost one third of stroke patients had comorbid depression, which was 3.5 times the odds of depression among matched controls.  Further, depression predicted poorer functional outcomes among stroke survivors.186  A recent systematic review published in JAMA found that the reported point-prevalence of major depressive disorder among those with CVD ranged from 7-22% across studies,187 although the directionality of risk cannot be inferred from such analyses.  Longitudinal cohort studies have helped to clarify the temporal relationship between depression and CVD.  A quantitative systematic review found that depression increases the risk of developing CAD by 64%, independent of known risk factors.188  Another meta-analysis found that after adjustment for  21 traditional risk factors, depression was associated with a 60% increased future risk of CVD.189  This increased risk for CV events has been shown to extend to increased mortality as well, with studies demonstrating that depression is an independent risk factor for CV- and all-cause mortality in the general population73 as well as in elderly populations.190  To further compound the problem, studies have shown that comorbid depression can increase CV disease progression among survivors of CV events.  A cohort study based on an RCT of 448 patients hospitalized for stroke found that depressive mood symptoms (assessed by questionnaires at baseline) were associated with over twofold the odds of mortality over follow-up, controlling for relevant covariates.71  A meta-analysis of the association of depression following MI with all-cause mortality and CV events found that patients who experienced depression following MI had 2.4 times the risk of all-cause mortality, 2.6 times the risk of CV-mortality, and 1.95 times the risk of new CV events compared to those without post-MI depression.191  To the best of our knowledge, only one study has considered the impact of depression as a potential confounder in the relationship between PsD and CV events.  Schmitt and Ford conducted a case-control study of PsC vs. age- and sex-matched controls using a German administrative database in which they found an increased risk of depression but no evidence of an increased risk of CV events in the sample, with adjustment for depression having no material impact on this null effect.37  However, it is important to note that there were a very small number of CV events (73) in this sample of PsC patients, in addition to a very small proportion of depressed individuals (243) with PsC.  It is highly unlikely that there was enough overlap between depression and CV events in this sample to properly adjust for the potential effect of depression on CV events.  Further, the use of prevalent cases of PsC, depression, and AMI may have further limited statistical power to observe expected relationships.  Indeed, the author?s failure to replicate the now well-established increased risk of CV events in PsC brings the validity and/or statistical power of the  22 study sample into question.  Importantly, we are unaware of any studies that have been designed to assess the independent effect of depression on the risk of CVD.  1.9 Shared Biological Pathways in Psoriatic Disease, Depression, and Cardiovascular Disease   The strong associations observed among PsD, depression, and CV risk factors suggest some shared pathology among these disorders.  Beyond evidence for common environmental and behavioural risk factors in all these disorders as discussed above, including stress, smoking, excessive alcohol consumption, poor diet, physical inactivity, medication non-adherence, low social support, and socioeconomic marginalization, these conditions share many biological features as well.  Indeed, many of the hallmark features of all three conditions include cytokine-mediated activation of pro-inflammatory pathways associated with obesity, increased lipid profile, endothelial dysfunction, insulin resistance, increased thrombotic factors, dysregulation of the stress response system (i.e., the hypothalamic-pituitary-adrenal [HPA] axis), glucocorticoid (GC) resistance, and platelet activation.    As described in Section 1.2.1, the pathogenesis and maintenance of psoriatic disease is largely immune-driven and characterized by increased levels of pro-inflammatory cytokines including IL-1, IL-1?, IL-6, TNF-?, IFN-?, acute phase reactants such as CRP, cellular and endothelial adhesion molecules such as ICAM-1, VCAM-1, L-selectin, and E-selectin,16,76,85-89,192 in addition to an upregulation of dendritic cells, monocytes/macrophages, mast cells, and natural killer cells.16,76,85,86,88  Interestingly, many of these pro-inflammatory changes also occur among depressed individuals who are otherwise healthy.  Numerous studies have revealed that individuals with depression show elevated levels of pro-inflammatory cytokines including IL-1, IL-1?, IL-2, IL-6,  23 IL-12, IFN-?, and TNF-?,166,172,193-195 some evidence for decreased levels of anti-inflammatory cytokines such as IL-4,57 higher levels of adhesion molecules like ICAM-1, MCP-1, and E-selectin,16,89,192,196 upregulated monocytes,166 increased levels of acute phase reactants including CRP,55,166,172,194,195  and high circulating levels of the stress hormone, cortisol.59,170,197  Studies have shown that depressed patients exhibit an exaggerated immune response to psychological stressors.52,58  This pro-inflammatory state in depression, similar to that seen in PsD, may partially explain the increased risk of CVD.  The observed relationship between depression and CVD was attenuated by 13% in one study198 and 17% in another199 upon adjustment for depression-associated inflammatory markers, suggesting that inflammation partially mediated that relationship.  Findings from animal and human studies have shown that depression can be induced by high levels of inflammation.  One study found that administration of IL-1?, IL-6, and TNF-? led to increased corticosterone levels, evidence of sickness behaviour in mice combined with either anorexia or anhedonia as evidenced by failure to consume a highly palatable food source, in addition to evidence of amine turnover in the hypothalamus, locus coeruleus, and amygdala.51  Other studies have shown that IL-1? administration in rats led to similar cytokine-induced sickness behaviour,176 and IFN-? and TNF-? lead to depressive symptomology in mice, mediated by the indolamine 2,3-dioxygenase (IDO) pathway.200  Studies of humans who have been treated with IFN-? have confirmed such findings, with therapy often inducing depressed mood, fatigue, despair, and cognitive impairment.201,202 Some studies have shown that effective treatment of depression can lead to corresponding decreases in systemic inflammation.  Chronic treatment with tricyclic antidepressants (TCAs) can reduce immune activation in rats in mild stress model of depression.203  Further, studies have shown that TCAs and selective serotonin reuptake inhibitors (SSRIs, i.e., third generation antidepressants) can inhibit production of pro-inflammatory cytokines IL-1, IL-6, TNF-?, IFN-?.56  A  24 study by Sutcigil and colleagues found that treatment of depressed patients with the SSRI sertraline resulted in significant decreases in levels of TNF-?, IL-2, IL-12, and MCP-1 in addition to corresponding decreases in anti-inflammatory cytokines including IL-4 and transforming growth factor (TGF)-?1 which correlated with improvement in depressive symptomology.57  Another study found that treatment of depression with the SSRI citalopram inhibited collagen-induced activation and aggregation of platelets, suggesting a potential cardioprotective role of SSRIs.204  However, evidence for an anti-inflammatory impact of antidepressant treatment has been equivocal with some authors failing to observe changes in key cytokine levels.205  If present, it remains unclear whether such anti-inflammatory or anti-thrombotic effects of antidepressants are large enough to have clinical relevance.  Conversely, there has been recent interest in whether anti-TNF biologic drugs for inflammatory disorders have additional anti-depressant benefits.  One study found that administration of the TNF-inhibitor infliximab in a rat model of stress-induced depression resulted in significant decreases in depression like symptoms.206  A limited number of human studies have also suggested a potential anti-depressant effect of anti-TNF therapy, with infliximab, etanercept, and adalimumab showing generally weak but measurable effects.207-211  Recent evidence suggests that anti-TNF therapy may only have antidepressant effects among depressed individuals with a high inflammatory load at baseline.212  Thus, while anti-TNF treatment may be useful as an adjunct therapy for depression in the context of inflammatory disease, therapies that directly target serotonergic transmission are likely to have clinical relevance.  Nonetheless, these data extend the theoretical basis an antidepressant benefit of reduced inflammation, and long-term follow-up of these preliminary RCTs may reveal a clinically-relevant antidepressant benefit over time.  The primary interface for interaction between psychological stressors and immune activity is via the body?s stress response system, the hypothalamic-pituitary-adrenal (HPA) axis.  The HPA axis is the body?s primary regulator of immune function, followed by parasympathetic innervations  25 on immune organs.213  In healthy individuals, activation of the HPA axis results in several physiological reactions that prepare the body to respond appropriately to perceived threats.  The paraventricular nucleus (PVN) of the hypothalamus is regulated by the neurotransmitters serotonin (5-HT; inhibitory) and norepinephrine (NE; excitatory) and can additionally be activated by the presence of pro-inflammatory cytokines (e.g., TNF-?, IL-1, IL-6) to secrete both arginine vasopressin (AVP) and corticotrophin releasing hormone (CRH), which synergistically stimulate the release of adenocorticotropic hormone (ACTH), which in turn stimulates the release of glucocorticoids (GCs; mostly cortisol in humans) from the adrenal cortex.  GCs act on various immune cells to suppress inflammation, thus reducing levels of pro-inflammatory cytokines that may have triggered the original response in a negative feedback loop.  Additional negative feedback loops built into the system include the inhibitory effects of ACTH on the hypothalamus as well as GCs looping back to produce inhibitory control of the anterior pituitary and the hypothalamus.59,176,213  The main components of the HPA axis are illustrated in Figure 1.1.  Short term activation of the HPA axis heightens arousal, improves attention and cognitive function and decreases appetite, increases heart rate, and feeds back to reduce further inflammation, all of which helps the body to prepare potential fight or flight situations.56,170,172,213-215     26 HypothalamusAnterior PituitaryAdrenalCortexNE5-HT-+++/---Immune cellsCytokinesACTHCRHAVP Figure 1.1 Main Components of the HPA Axis   Under normal circumstances, time-limited activation of the HPA axis would naturally lead to decreased inflammation.  In the context of depression, however, chronically decreased levels of serotonin and elevated levels of NE translate to long-term disinhibition and persistent activation of the HPA axis, respectively.165  Conversely, chronic inflammation associated with PsD and CVD can additionally fuel constant activation of the HPA axis.196  The resulting long-term elevations in cortisol levels acting on immune cells, the anterior pituitary, and the PVN can lead to GC resistance, in which cells no longer respond appropriately to elevated levels of cortisol.  In this pathological state, all negative feedback loops in the HPA axis are rendered ineffective, resembling a car with a fully-depressed accelerator and faulty brakes.  In an early landmark study of GC resistance in depression, Lowy and colleagues studied the effects of dexamethasone administration in depressed subjects.  Dexamethasone, a potent synthetic GC, normally  27 suppresses plasma cortisol levels in healthy individuals via its negative feedback effects on the HPA axis coupled with its suppression of pro-inflammatory cytokines that normally trigger the activation of the HPA axis.  However, the authors discovered that in depressed subjects, dexamethasone failed to exert such effects, suggesting the development of GC resistance in immune cells, the anterior pituitary, and the hypothalamus.216  Many studies have since demonstrated the phenomenon of GC resistance in depressed patients.63,196,215,217,218  These aberrations in HPA control are thought to underlie the elevated levels of cortisol, pro-inflammatory cytokines, adhesion molecules, and immune cell infiltration observed in depression.  The increased cycle of inflammation triggered by activation of the HPA axis by depression-related chronic stress or PsD-related pro-inflammatory cytokine profile originating in psoriatic lesions has many effects that could potentially lead to the development of CVD.  As mentioned above, the inflammation observed in PsD and depression corresponds to increasing expression of intercellular and vascular adhesion molecules (ICAMS & VCAMS), endothelial adhesion molecules (selectins), and platelet activation, which are heightened by a sustained acute phase response with elevated CRP and IL-6.196  A large body of research has shown that such adhesion molecules are of critical importance in the development of atherosclerosis, with ICAM-1, VCAM-1, and L-selectin consistently found in atherosclerotic plaques.16,219,220  Further, circulating levels of these adhesion molecules predict the development of CVD.16  Higher levels of CRP observed in both PsD and depression have also been shown to activate fibrinogen deposition in the vessel wall.196  IL-1, IL-6, and TNF-? are also expressed on vascular walls in regions with fatty deposition, and are associated with adipose tissue.  HPA activity on adipocytes leads to vastly upregulated production of pro-inflammatory cytokines and adhesion molecules.221  Moreover, IL-6 inhibits glycogen synthesis, and CRP, IL-6, and TNF-? are strongly correlated with insulin sensitivity, high  28 tryglycerides, low HDL cholesterol, and elevated blood pressure, all markers of the metabolic syndrome that elevates one?s future risk of CVD.196,221  In addition to depression and PsD related inflammation leading to dysregulation of the HPA axis and the resultant increases in systemic inflammation, high levels of inflammation resulting from PsD and CVD can conversely trigger the onset and exacerbate symptoms of depression.  High levels of circulating pro-inflammatory cytokines in psoriatic disease such as TNF-? and IFN-? strongly stimulate the production of IDO.  Tryptophan, the rate-limiting precursor of 5-HT, is catabolized by IDO via the kynurenine pathway, leading to lower plasma bioavailability of tryptophan for the production of 5-HT and an increased synthesis of potentially detrimental tryptophan catabolites along the IDO pathway (TRYCATs).  These TRYCATs include kynurenine, kynurenic acid, xanthurenic acid, and quinolinic acid, which can trigger a cascade of both pro- and anti-inflammatory effects, including antioxidizing and oxidizing, neurotoxic and neuroprotective effects, in addition to increased apoptosis.175,222  5-HT is the key neurotransmitter implicated in depressive symptomology, with decreased levels of central 5-HT leading to depressive symptomology, and conversely, effective treatments for depression involving increasing the quantity or duration of serotonin in the neuronal synapse.214  Thus, the decreased bioavailability of tryptophan for the production of 5-HT is thought to underlie the somatic and mood disturbances in depression such as fatigue, lack of concentration, and anhedonia.56,63,175,223  The activation of reactive oxygen species (ROS) and oxidative and nitrosative (O&NS) pathways by TRYCATs are thought to underlie both increased neurodegeneration and decreased neurogenesis, resulting in volume depletion of the hippocampus, amygdala, prefrontal cortex, anterior cingulate, and basal ganglia.63,169,224  These pathways likely lead to the loss of hippocampal volume due to increased apoptosis and decreased neurogenesis.165,225  For instance, IDO is predominantly activated in microglial cells, which preferentially degrade IDO to the TRYCAT quinolinic acid, an NMDA  29 receptor agonist.  Astrocytes, which normally favor the production of 5-HT and counterbalance the effects of glutamate, are depleted following activation of the IDO pathway by pro-inflammatory cytokines.  The sum effect of these changes is an imbalance between microglia and astrocytes, with corresponding increases in glutamate and decreases in 5-HT transmission.222  Increased activation of the glutamatergic pathway can trigger a cytotoxic cascade that contributes to oxidative stress in addition to apoptosis of neurons, astrocytes, and oligodendrocytes, especially in the hippocampus.175,214    In summary, the increased inflammation in PsD and CVD can trigger the development and maintenance of the physiologic and symptomatic changes in depression, and conversely depression can further fuel the chronic activation of the HPA axis to promote further inflammation, which can in turn exacerbate symptoms of PsD and promote atherosclerosis leading to CVD and mortality following CVD.  Coupled with their multiple shared symptoms and related environmental and behavioural risk factors, the relationships among PsD, depression, and CVD could be described as a vicious cycle.  Despite several mechanisms through which a comorbid diagnosis of depression could increase the risk of CVD among patients with PsC and PsA, no studies have investigated this potential link to date.  To improve cardiovascular outcomes in PsC and PsA, two major knowledge gaps must be filled: 1) quantification of the burden of depression among patients with PsC and PsA (to determine the scope of the problem), and 2) quantification of the impact of depression on AMI outcomes among patients with PsC and PsA.  1.10 Rationale and Overview of Studies   The central objective of this thesis was to leverage the particular strengths of large, population-based administrative datasets to assess the impact of comorbid depression on the risk  30 of acute myocardial infarction (AMI) among individuals with psoriatic disease.  To provide context for these findings, the first objective was to describe the scope of the problem by describing the burden (prevalence) of physician-diagnosed depression among psoriatic patients at the population level.  Therefore, Theme 1 of this thesis quantifies the problem of depression in PsD by estimating the population prevalence of physician-diagnosed depressive disorders and comparing the odds of such disorders between individuals with PsD to age-, sex-, and index date-matched population controls.  Theme 2 expands upon these data by describing the independent impact of depression on the risk of AMI among individuals with PsD, and the potential for depression to modify the effect of AMI among individuals with PsD.   1.11 Administrative Data Context   The use of administrative data to answer key population health questions has expanded rapidly in recent years owing to advances in information technology that allow for large-scale data analysis.  Various sources of administrative data that are routinely collected as part of patient care, government or insurance billing, and resource management can now be linked at the individual level allowing researchers to answer important public health questions.  Further, advances in software and computing power allow for analyses of large-scale datasets that would have previously been unfeasible.  Canada?s publicly-funded, universal health care system has resulted in comprehensive data capture of health services utilization, collected for the purposes of administrative billing.  In British Columbia (BC), a provincially-administered and publically-funded health system called the Medical Services Plan (MSP) covers all hospitalizations, at-home care, prescription drugs for  31 seniors, diagnostic and laboratory tests, and outpatient physician visits for all residents.  Further, all dispensed medications covered by private insurance companies are recorded in provincial records.  Population Data BC (PopData), formerly known as the BC Linked Healthcare Database (BCLHD) is an integrated source of administrative databases with anonymized person-level information from the MSP (including date of service, practitioner, and diagnostic code associated with the visit using International Classification of Disease Version 9 [ICD-9] codes), in addition to comprehensive birth and death records (including the cause and date of death according to ICD-10 codes) from Vital Statistics, all inpatient hospitalizations from the Hospital Separations file (including admission date, multiple diagnostic fields, procedural and intervention codes, and separation dates), as well as province-wide prescription claims data from PharmaNet, regardless of payer.  PharmaNet data includes the date and type of medication dispensed using Canadian Drug Identity Codes [CDIC], and days supplied in the prescription.226-228  The use of large-scale administrative data to answer health questions has many advantages.  As these data are captured at the population level, they reflect the demographic and geographic diversity of the BC population thus reducing the influence of selection bias.  Data on the entire population allows for the development of large-scale cohorts with high statistical power to assess relationships between rare diseases and infrequent outcomes.  Further, many questions that cannot be addressed using the recruitment and randomization of human participants due to feasibility, financial, or ethical concerns can be addressed using administrative data.    32 CHAPTER 2 GENERAL METHODS  2.1 Data Source and Study Population  Our source population was derived from the Population Data BC (PopData) database, an extensive health data resource containing integrated longitudinal information on the entire British Columbia (BC) population covered by the universal healthcare system (4.3 million individuals as of 2005).229  The PopData database links information on all visits to healthcare professionals, laboratory test ordered, hospital visits, interventions, and prescriptions among BC residents.  Because all BC residents except for a small number of federal employees have their medical services covered by the universal Medical Services Plan (MSP), and because all dispensed drugs are recorded regardless of payer, the study population is representative of BC?s adult population.  2.2 Development of Psoriatic Disease and Control Cohorts  All of our analyses employed linked administrative health data from PopData, including all physician visits, hospital admissions, and death records from April 1991 to March 2006, as well as all dispensed medications from April 1996 to March 2006, for the entire adult (?18 years) population of BC, Canada (4.3 million individuals).  For Theme 1, which assessed the burden of depression in psoriatic disease, we sought to employ a cohort definition of PsC and PsA that emphasized sensitivity to better capture the scope of disease burden.  For Theme 2, which assessed the independent impact and potential effect modification of depression on the risk of acute myocardial infarction (AMI) in psoriatic disease, we decided to employ a highly specific case  33 definition for our PsC/PsA cohorts to reduce the likelihood of misclassification of exposures.229  The use of a highly specific case-definition is recommended when exploring relationships between diseases as reduced diagnostic error can improve power to detect disease associations when they are present.230,231  For each cohort, five unexposed, general population controls were matched to each PsC/PsA case by age, sex, and follow-up time (index date).  For cohort entry, individuals were required to have at least one year of registry records available prior to the index date.  For all investigations, we excluded individuals with diagnoses of PsC or PsA prior to the index date (baseline) to yield incident cases.  Expanded details of theme-specific cohort definitions are discussed in Chapters 3 & 4.   All individual-level data contained no personal identifying information and data access procedures complied with BC?s Freedom of Information and Privacy Protection Act.  Ethical approval for this study was granted by the University of British Columbia?s Behavioural Research Ethics Board (certificate number H06-03851).    2.3 Outcomes of Interest   For Theme 1, our outcome of interest was physician-diagnosed depression, as ascertained by ICD-9 codes.  For Theme 2, our outcome of interest was AMI as assessed by ICD-9 codes from the MSP files, ICD-9 or -10 codes from Hospital Separations files, or by cause of death information (ICD-10 codes) from Vital Statistics files.  Specific details of outcome definitions according to study question are fully discussed in theme-specific methods (Chapters 3 & 4).       34 2.4 Covariate Assessment  We investigated potential confounders for the relationships of interest that were available in the administrative databases.  Sociodemographic variables included age, sex, and socioeconomic status (SES) assessed at the index date.  Socioeconomic status was estimated at the index date using income quintiles obtained from the Registry files, ranked from lowest to highest.  PopData calculates income quintiles based on registry records for PharmaCare, BC?s income-based drug insurance program, and Canadian census data.  Registrants for PharmaCare must provide information on household income, which is verified by the Canada Revenue Agency (CRA).232  Almost 80% of BC households have registered for PharmaCare;232 thus, CRA-validated household income was available and likely used to construct income percentiles for most of our sample.  For those individuals not registered in the PharmaCare program, income quintiles are estimated based on average neighbourhood income according to Dissemination Areas (~400-700 individuals per region) from the 2002 Canadian Census.  This method of SES capture has been successfully employed to assess the SES gradient in appropriate AMI care.233  We also considered factors known to influence depression and AMI risk as potential covariates in our multivariate regression models.  We used fixed-in-time binary variables measured in the 12 months prior to the index date to determine the baseline presence of chronic comorbid conditions, including chronic obstructive pulmonary disease (COPD) (ICD-9 = 490.x-496.x), obesity (278.0), alcoholism / liver disease (291.x, 303.x), cerebrovascular accidents (CVA) (430.x-438.x), hypertension (401.x-405.x), sepsis (995.9, 036.x, 038.x), varicose veins (454.x, 671.0), peripheral vascular disease (PVD) (440.x-444.x), congestive heart failure (CHF) (428.x), chronic renal disease (CRD) (585.x), inflammatory bowel disease (IBD) (555.x), malignant neoplasms (140.x-239.x), trauma (860.x-869.x), and fractures (800.x-809.x, 818.x-823.x, 828.x).  The potential impact of other comorbid  35 medical conditions was also assessed using a modified Charlson Comorbidity Index (CCI) score for each subject based on comorbidity data from the 12 months prior to the index date, which has been adapted for use in administrative databases.234,235  The CCI is a weighted variable that captures the burden of diseases that are associated with an increased risk of death, including myocardial infarction (MI), CHF, PVD, dementia, COPD, rheumatic diseases, peptic ulcer disease, mild liver disease, diabetes, diabetes with chronic complications, CRD, moderate to severe liver disease, and AIDS.234  Finally, we adjusted for health resource utilization by assessing the rate of hospitalizations in the year prior to the index date across cohorts.  Use of standard DMARDs, biologic DMARDs, and antidepressants were also assessed over the follow-up.  2.5 Statistical Analyses   For both themes, baseline characteristics were compared between exposed (PsC/PsA cases) and unexposed (control) cohorts.  Both descriptive statistics and regression analyses were employed according to themes, as described in detail in the following chapters.  For all estimates, we calculated 95% confidence intervals (CIs).  All analyses were conducted using SAS, Version 9.2 and 9.3 (SAS Institute, Cary, North Carolina).     36  CHAPTER 3 BURDEN OF PHYSICIAN-DIAGNOSED DEPRESSION IN  PSORIASIS AND PSORIATIC ARTHRITIS: A POPULATION-BASED STUDY  3.1 Specific Aims and Corresponding Hypotheses: Aim 1. To identify the burden (prevalence) of depression among individuals with PsC and PsA in a general population context. Hypothesis 1. There is a substantial burden of depression among individuals with PsC and PsA at the population level. Aim 2. To compare the population rates of depression (adjusted odds ratio) between individuals with PsC/PsA and general population controls. Hypothesis 2. There is an increased prevalence of depression in psoriatic individuals (PsC and PsA) compared to controls.  3.2 Methods  3.2.1 Development of Psoriatic Disease and Control Cohorts As described in Section 2.2 above, we sought to employ a cohort definition in this theme that maximized sensitivity to better capture the burden of psoriatic disease for descriptive purposes, while balancing the need for sensitivity to avoid misclassification of diagnoses (Cohort Definition A).  Individuals were included in our PsD cohort (Definition A) if they received care for their condition between April 1996 and March 2006 as evidenced by one of the following: 1) ?2 ICD-9 specific (four-digit) physician diagnostic codes for PsC or PsA (International Classification of Diseases, Ninth Revision [ICD-9], 696.1 or 696.0, respectively) at least 30 days and no more than  37 2 years apart between Apr 1996 and Mar 2006; or 2) ?1 specific ICD-9 diagnostic code from a hospital.  To ensure a more complete capture of the burden of disease, we additionally included individuals who received: 3) a three-digit (non-specific) code for 696.x (psoriasis and similar disorders) from either a rheumatologist or dermatologist to capture PsA and PsC, respectively (Table 3.1).  Although the code for 696.x also encompasses disorders such as parapsoriasis and pityriasis rosea, the rarity of these diseases would suggest that dermatologists primarily use this code for PsC and rheumatologists primarily use the code for PsA.  Thus, our choice to maximize generalizability (sensitivity) in this cohort likely came at some cost to specificity.  US studies of community-based and managed care electronic databases have shown that ? 2 specific diagnostic codes in 2 years from a GP or ? 1 specialist code for PsC and PsA have positive predictive values (PPVs) from 78-90%.236,237  The index date of diagnosis (baseline) was defined as the date at which all inclusion criteria had been met, i.e., the first specialist / hospital diagnosis or the second general practitioner diagnosis.  Five unexposed, general population controls were matched to each PsC/PsA case by age, sex, and follow-up time (index date).  For entry into either cohort, individuals were required to have at least one year of registry records available prior to the index date.  We excluded all individuals with any specific or non-specific diagnoses of PsC or PsA (696.x) prior to baseline to yield incident cases.    3.2.2 Outcome Assessment    Our outcome of interest was physician-diagnosed depression, as ascertained by ICD-9 codes.  Depression was defined any one inpatient or outpatient ICD-9 code of 296.x (episodic mood disorder), 309.x (adjustment reaction disorder), 300.4 (dysthymic disorder), or 311.x (depressive disorder, NOS), over the potential 10 years of follow-up (Table 3.1).  This definition for  38 depression has been recently employed in a Canadian claims database,238 and a similar definition employed in a US claims database to estimate the prevalence of depression in severe psoriasis.239  Depression definitions using ?1 of the same or similar ICD-9 codes have been validated against chart review with reported validities of ?90%.240,241  3.2.3 Covariate Assessment  We investigated potential confounders for the relationships of interest that were available in the administrative databases.  Sociodemographic variables included age, sex, and socioeconomic status (SES) assessed at the index date.  Socioeconomic status was estimated at the index date using income quintiles obtained from the Registry files, ranked from lowest to highest.  PopData calculates income quintiles based on registry records for PharmaCare, BC?s income-based drug insurance program, and Canadian census data.  Registrants for PharmaCare must provide information on household income, which is verified by the Canada Revenue Agency (CRA).232  Almost 80% of BC households have registered for PharmaCare;232 thus, CRA-validated household income was available and likely used to construct income percentiles for most of our sample.  For those individuals not registered in the PharmaCare program, income quintiles are estimated based on average neighbourhood income according to Dissemination Areas (~400-700 individuals per region) from the 2002 Canadian Census.  We also considered factors known to influence AMI risk as potential covariates in our multivariate regression models.  We used fixed-in-time binary variables measured in the 12 months prior to the index date to determine the baseline presence of chronic comorbid conditions, including chronic obstructive pulmonary disease (COPD) (ICD-9 = 490.x-496.x), obesity (278.0), alcoholism / liver disease (291.x, 303.x), cerebrovascular accidents (CVA) (430.x-438.x), hypertension (401.x-405.x), sepsis (995.9, 036.x, 038.x), varicose  39 veins (454.x, 671.0), peripheral vascular disease (PVD) (440.x-444.x), congestive heart failure (CHF) (428.x), chronic renal disease (CRD) (585.x), inflammatory bowel disease (IBD) (555.x), malignant neoplasms (140.x-239.x), trauma (860.x-869.x), and fractures (800.x-809.x, 818.x-823.x, 828.x).  The potential impact of comorbid medical conditions was also assessed using a modified Charlson Comorbidity Index (CCI) score for each subject based on comorbidity data from the 12 months prior to the index date, which has been adapted for use in administrative databases (Charlson, 1987; Deyo, 1992).  The CCI is a weighted variable that captures the burden of diseases that are associated with an increased risk of death.  Finally, we adjusted for health resource utilization by assessing the rate of hospitalizations in the year prior to the index date across cohorts.  Use of standard DMARDs, biologic DMARDs, and antidepressants were also assessed over the follow-up (Table 3.1).  40 Table 3.1 ICD-9 Definitions of PsD Cohort (Definition A), Covariates, and Outcome Cohort ICD-9 Definition PsD Cohort Definition A a) ?2 GP codes for 696.0, 696.1 (?30 days and ? 2 years apart); or b) ?1 hospital code for 696.0, 696.1; or c) ?1 dermatologist or rheumatologist code for 696.x;  AND ?1 year without 696.x codes prior to index date  Covariate ICD-9 Definition COPD 490.x-496.x Obesity 278.0 Alcoholism/liver disease 291.x, 303.x CVA 430.x-438.x Hypertension 401.x-405.x  Sepsis 995.9, 036.x, 038.x Varicose Veins 454.x, 671.0 PVD 440.x-444.x  CHF 428.x Angina 413.x CRD 585.x IBD 555.x Malignant neoplasms 140.x-239.x Trauma 860.x-869.x Fractures 800.x-809.x, 818.x-823.x, 828.x  CCI  Weight of 1: AMI (410.x), CHF (428.x); PVD (433.9, 441.x, 785.4, V43.4); CVD (430.x - 438.x); Dementia (290.x); CPD (490.x - 496.x, 500.x - 505.x, 506.4); Rheumatic Diseases (710.0, 710.1, 710.4, 714.0 - 714.2, 714.81, 725.x); Peptic Ulcer Disease (531.x - 534.x); Mild Liver Disease (571.2, 571.5, 571.6, 571.4 - 571.49); Diabetes (250.1 - 250.3, 250.7);  Weight of 2: Diabetes with Chronic Complications (250.4 - 250.6); Hemiplegia or Paraplegia (344.1, 342.x); Renal Disease (582.x, 583.1 - 583.7, 585.x, 586.x, 588.x);  Weight of 3: Moderate to Severe Liver Disease (72.2 - 572.8);  Weight of 6: AIDS (042.x - 044.x)  Outcome ICD-9 Definition Depression  296.x, 309.x, 300.4, 311.x   3.2.4 Statistical Analyses   Baseline characteristics were compared between exposed (PsC/PsA cases) and unexposed (control) cohorts.  We calculated the period prevalence of physician-diagnosed  41 depression in both cohorts over total follow-up.  These data were used to estimate crude, age- and sex-adjusted, and fully-adjusted odds ratios (ORs) for depression in PsC/PsA compared to healthy controls, using logistic regression.  Although cohorts were matched based on age and sex, we additionally adjusted for these factors at the analysis stage to account for any differences in the distribution of these variables following data cleaning (such as removing AMI before the index date).  For all ORs, we calculated 95% confidence intervals (CIs).  All analyses were conducted using SAS, Version 9.2 and 9.3 (SAS Institute, Cary, North Carolina).     3.3 Results   Definition A of our PsD exposure cohort included a total of 59,310 incident PsD cases who contributed a total of 282,810 person-years (PY) of follow-up (mean of 4.77 years) and our age-, sex-, and index date-matched control (unexposed) cohort included a total of 286,615 individuals who contributed a total of 1,411,758 PY of follow-up (mean of 4.93 years) between April 1996 and March 2006.  Of PsD cases, 4,709 (8%) had incident PsA at baseline and 54,601 (92%) had incident PsC.  Based on a total population of 4.3 million residents of BC, Canada, this cohort yielded a prevalence of PsD of 1.4%, which is in keeping with previous population estimates of the disease burden.82,95-98  Characteristics of the study sample are summarized in Table 3.2.  In both cohorts, the mean age was 49 (SD=17) years there were equal percentages of males and females.  The age of PsD incidence appeared to be normally distributed about the mean for both males and females, with no evidence for a bi-modal peak onset in this sample (Figures 3.1 and 3.2).       42 Table 3.2 Characteristics of Exposed and Unexposed Cohorts (Definition A)  Covariate PsD Exposure Cohort  (N =59,310) N (%) Age- & Sex-Matched Control Cohort (N = 286,615) N (%) Age, Mean (SD) 49.18 (17.13) 49.08 (16.98) Male Sex 29467 (49.68) 142480 (49.71) COPD  4771 (8.04) 18982 (6.62) Obesity  94 (0.16) 324 (0.11) Alcoholism/liver disease 408 (0.69) 1529 (0.53) CVA  644 (1.09) 2738 (0.96) Hypertension 9098 (15.34) 40622 (14.17) Sepsis  235 (0.40) 756 (0.26) Varicose Veins  561 (0.95) 2197 (0.77) PVD  868 (1.46) 3107 (1.08) CHF  763 (1.29) 2924 (1.02) Angina 1911 (3.22) 7567 (2.64) CRD  244 (0.41) 697 (0.24) IBD  223 (0.38) 667 (0.23) Malignant Neoplasms 6569 (11.08) 21941 (7.66) Trauma  42 (0.07) 204 (0.07) Fractures  657 (1.11) 204 (0.07) Standard DMARDs (over follow-up) 5653 (9.53) 6746 (2.35) Biologic DMARDS (over follow-up) 434 (0.73) 316 (0.11) *Age assessed at baseline, comorbidities assessed in the year prior to the index date, and medication use assessed over follow-up  43 0200040006000800010000120001400018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryPsoriatic Cohort (Def A)FemaleMaleTotal  Figure 3.1 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition A) at Index Date  01000020000300004000050000600007000018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryControl Cohort N (Def A)FemaleMaleTotal Figure 3.2 Age and Sex Characteristics of Control Cohort (Definition A) at Index Date  44  Over a mean follow-up of 4.77 PY, a total of 14,610 individuals in the PsD cohort (Definition A) had at least one diagnosis of depression, corresponding to a period prevalence of 24.63%.  Compared to age-, sex-, and index date-matched controls in the crude logistic regression analysis, individuals with psoriatic disease had a 25% increased odds of having depression over the follow-up period.  Statistical adjustment for age and sex did not change the OR materially, whereas after final adjustment for all covariates in Section 3.2.3, the increased odds were attenuated to 16%, but remained statistically significant (Table 3.3).  Table 3.3 Odds of Prevalent Depression between Cohorts  Prevalence of Depression  N (%)  Unadjusted OR Depression Age- and Sex- Adjusted OR Depression Fully-Adjusted* OR Depression PsD Cohort Def A (N=59,310) 14610 (24.63) 1.25 (1.22-1.27) 1.25 (1.22-1.28) 1.16 (1.13-1.18) *Adjusted for all baseline covariates and socioeconomic status  45 CHAPTER 4 CARDIOVASCULAR IMPACT OF DEPRESSION IN PSORIASIS AND PSORIATIC ARTHRITIS: A POPULATION-BASED STUDY  4.1 Specific Aims and Corresponding Hypotheses:  Primary Aims Aim 1. To evaluate the independent impact of depression on the risk of incident acute myocardial infarction (AMI) in patients with incident PsC and PsA. Hypothesis 1. Depression is an independent predictor of AMI risk among individuals with PsC and PsA. Aim 2. To determine whether depression modifies the risk of incident AMI in patients with incident PsC and PsA.   Hypothesis 2. The risk of AMI associated with PsC and PsA varies by depression strata.  Secondary Aims Aim 3. To determine whether sex modifies the risk of incident AMI in patients with incident PsC and PsA. Hypothesis 3. The risk of AMI associated with PsC and PsA varies by sex strata. Aim 4. To determine whether age modifies the risk of incident AMI in patients with incident PsC and PsA. Hypothesis 4. The risk of AMI associated with PsC and PsA varies by age strata.    46 4.2 Methods  4.2.1 Development of Psoriatic Disease and Control Cohorts  As the central objective in this study was to assess the longitudinal relationship between diseases, we sought to employ a highly specific cohort definition for psoriatic disease (Cohort Definition B), as described in Section 2.2 above, to avoid misclassification error that could bias estimates toward the null.  Individuals met inclusion criteria for Definition B of our PsD cohort if they received care for PsC or PsA between April 1996 and March 2006 as evidenced by one of the following: 1) ?2 ICD-9 specific (four-digit) physician diagnostic codes for PsC or PsA (696.1 or 696.0, respectively) at least 30 days and no more than 2 years apart; or 2) ?1 specific ICD-9 diagnostic code from a dermatologist (for PsC), a rheumatologist (for PsA), or from hospital.  The key difference from Definition A was that this cohort required specialists to employ 4-digit, specific codes for PsC or PsA (696.1 and 696.0, respectively) rather than 3-digit, non-specific codes for psoriasis and similar disorders (696.x).  This definition was employed to gain further specificity by ruling out disorders such as parapsoriasis and pityriasis rosea, which are also encompassed in 696.x.  The index date of diagnosis (baseline) was defined as the date at which all inclusion criteria had been met, i.e., the first specialist / hospital diagnosis or the second general physician diagnosis.  Five unexposed, general population controls were matched to each PsC/PsA case by age, sex, and follow-up time (index date).  For entry into either cohort, individuals were required to have at least one year of registry records available prior to the index date.  To better capture the temporal nature of the studied relationships, we chose to employ new-onset cases of exposures and outcomes wherever possible.  Thus, individuals with diagnoses of PsC, PsA, or MI prior to baseline were excluded to yield incident cases.  Ideally, we sought to employ incident cases of  47 depression in all analyses.  However, due to concerns over potential power limitations with using incident cases of depression to predict infrequent AMI outcomes, we ran all analyses first using incident depression (by removing all cases of depression prior to the index date), and then employing prevalent depression cases to improve power and stabilize estimates.  Thus, the cohort definition was further broken down into Definition B1 (fully incident cohort definition, with prevalent depression cases removed) and Definition B2 (incident PsD, incident AMI, prevalent depression).  Both cohort definitions are described in Table 4.1.  4.2.2 Assessment of Depression  Depression was defined as in Chapter 3 by the following inpatient or outpatient ICD-9 codes: 296.2, 296.3, 300.4, or 311.  Definition B1 employed incident cases of depression whereas Definition B2 employed prevalent cases (Table 4.1).    4.2.3 Assessment of Outcomes   Our outcome in this study was first non-fatal or fatal acute myocardial infarction (AMI) event during the follow-up period, as ascertained by ICD-9 or ICD-10 codes.  Specifically, non-fatal events were ascertained by any one ICD-9 code for AMI (410.x) in the Hospital Separations data, corresponding to AMI events that occurred as the indication for or as a complication during hospitalization.  Canadian validation studies of administrative hospital discharge records have established a high degree of accuracy in ICD-9 coding for AMI, with PPVs ranging from 89% to 96%.242,243  The event date for non-fatal AMIs was the date of hospital admission.  To capture fatal AMI events, we identified all fatal AMIs that occurred during hospitalization in addition to deaths  48 that occurred outside of hospitals with AMI listed as the cause of death on the death certificate, based on ICD-10 codes (121.x) in Vital Statistics records.  In these cases, the death date was employed as the event date.  The use of ICD-10 codes for fatal AMI has been previously validated in Canadian administrative data, with a PPV of 93.5%.244  4.2.4 Assessment of Covariates   We employed the same covariate definitions in this study as those described in Section 3.2.3 above.  Covariates defined by ICD-9 codes are summarized in Table 4.1.  49 Table 4.1 ICD-9 Definitions of PsD Cohorts (Definitions B1 & B2), Covariates, and Outcome Cohort ICD-9 Definition PsD Cohort Definition B1 a) ?2 GP codes for 696.0, 696.1 (?30 days and ? 2 years apart); or b) ?1 hospital, dermatologist, or rheumatologist code for 696.0, 696.1; AND ?1 year without 696.x, 296.x, 309.x, 300.4, 311.x, or 410.x codes prior to index date  PsD Cohort Definition B2 a) ?2 GP codes for 696.0, 696.1 (?30 days and ? 2 years apart); or b) ?1 hospital, dermatologist, or rheumatologist code for 696.0, 696.1; AND ?1 year without 696.x or 410.x codes prior to index date  Covariate ICD-9 Definition Depression  296.x, 309.x, 300.4, 311.x COPD 490.x-496.x Obesity 278.0 Alcoholism/liver disease 291.x, 303.x CVA 430.x-438.x Hypertension 401.x-405.x  Sepsis 995.9, 036.x, 038.x Varicose Veins 454.x, 671.0 PVD 440.x-444.x  CHF 428.x Angina 413.x CRD 585.x IBD 555.x Malignant neoplasms 140.x-239.x Trauma 860.x-869.x Fractures 800.x-809.x, 818.x-823.x, 828.x  CCI  Weight of 1: AMI (410.x), CHF (428.x); PVD (433.9, 441.x, 785.4, V43.4); CVD (430.x - 438.x); Dementia (290.x); CPD (490.x - 496.x, 500.x - 505.x, 506.4); Rheumatic Diseases (710.0, 710.1, 710.4, 714.0 - 714.2, 714.81, 725.x); Peptic Ulcer Disease (531.x - 534.x); Mild Liver Disease (571.2, 571.5, 571.6, 571.4 - 571.49); Diabetes (250.1 - 250.3, 250.7);  Weight of 2: Diabetes with Chronic Complications (250.4 - 250.6); Hemiplegia or Paraplegia (344.1, 342.x); Renal Disease (582.x, 583.1 - 583.7, 585.x, 586.x, 588.x);  Weight of 3: Moderate to Severe Liver Disease (72.2 - 572.8);  Weight of 6: AIDS (042.x - 044.x)  Outcome ICD-9 Definition ICD-10 Definition AMI 410.x  121.x      50 4.2.5 Statistical Analysis   We first compared the baseline characteristics and risk factors for CVD between our PsD and control cohorts.  We identified incident cases of PsC and PsA and calculated person-time of follow-up from the index date (date at which criteria for PsC or PsA had been met, or matched physician visit date in unexposed controls) to the date of AMI, last health care service use, death, or the end of the study period (March 31st, 2006), whichever came first.  Incidence rates of AMI according to person-years were calculated for both cohorts.  We first attempted our regression analyses using incident depression (Definition B1) by excluding all prevalent cases that occurred prior to the index date in both cohorts.  Because of the power limitations associated with diminished sample size in this scenario, we additional re-ran our regression analyses using prevalent depression cases.  We calculated the incidence rate of depression in our incident cohort according to person years of follow-up, and the period prevalence and odds of depression compared to controls in our prevalent depression cohort.  Cox proportional hazards models were then used to determine the independent impact of depression (first incident, then prevalent) on the risks of AMI in PsD, adjusting for all covariates described in Section 3.2.3, above.  A backward stepwise selection procedure was employed to select covariates with statistically significant (p<0.05) independent associations with AMI for inclusion into the final model, to optimize model fit.  We decided a priori that key covariates, including age, sex, exposure status, and depression status, would remain in the final model, regardless of significance level.  To test for effect modification by depression, we repeated our Cox analysis first limiting to individuals without depression and then to only those individuals with depression over follow-up to examine the impact that depression status had on hazard ratios (HRs) associated with PsC/PsA exposure status.  As secondary analyses, we further assessed for effect modification by sex and age.  An age cut-off of  51 50 was employed in our stratified analyses as AMI is a disease associated with advanced age, and CV risk has been shown to follow a quadratic curve in which the competing risk of age, particularly among those over 50, outweighs the impact of accelerating risk factors for atherosclerosis including metabolic syndrome.245,246  For all Cox models, we tested the proportionality assumption using log-log plots and by testing the significance of interactions with PsC/PsA and depression exposures with follow-up time.    4.3 Results   Our exposure cohort definition B2, which included prevalent depression cases, included a total N of 10,041 incident cases of PsD, yielding a total of 46,012 PY of follow-up between April 1996 and March 2006 (mean follow-up of 4.58 years).  Of PsD cases, 5,854 were diagnosed with PsA at baseline (58.3%) whereas 4,187 had PsC (41.7%). The corresponding age-, sex-, and index date-matched control cohort included a total of 47,415 individuals who contributed a total of 232,619 PY of follow-up (mean follow-up of 4.91 years).   When we excluded cases and controls with diagnoses of depression prior to the index date (incident depression definition, B1), our fully-incident PsD exposure cohort was reduced to an N of 7,992 cases (of whom, 4669 (58.4%) had PsA at baseline)who contributed a total of 34,288 PY of follow-up (mean of 4.54 PY) and our age-, sex-, and index date-matched control (unexposed) cohort to a total of 39,436 individuals who contributed a total of 185,198 PY of follow-up (mean of 4.70 PY) between April 1996 and March 2006.  Below, we describe key summary statistics and results according first to the incident depression cohort definition (B1), followed by the prevalent depression cohort definition (B2).      52 4.3.1 Results for Fully-Incident Cohort B1  Characteristics of the study sample are summarized in Table 4.2.  The mean ages in the PsD and control cohort were 49 (SD=17) and 48 (SD=17) years, respectively, and both cohorts were 51% male.  The age of PsD incidence was approximately normally distributed about the mean for both males and females, although females had a relatively lower rate of PsD incidence between the ages of 28 and 57 compared to males, with some suggestion of a bi-model peak onset among females only (Figures 4.1 and 4.2).    Table 4.2 Characteristics of Exposed and Unexposed Cohorts (Definition B1)  Covariate PsD Exposure Cohort (N=7,992) N (%) Age- & Sex-Matched Control Cohort (N = 39,436) N (%) Age, Mean (SD) 48.93 (17.16) 48.37 (16.85) Male Sex 4289 (53.67) 21112 (53.53) COPD  611 (7.65) 2460 (6.24) Obesity  26 (0.33) 34 (0.09) Alcoholism/liver disease 62 (0.78) 158 (0.40) CVA  99 (1.24) 307 (0.78) Hypertension 1156 (14.46) 5265 (13.35) Sepsis  37 (0.46) 88 (0.22) Varicose Veins  55 (0.69) 272 (0.69) PVD  99 (1.24) 346 (0.88) CHF  129 (1.61) 306 (0.78) Angina 207 (2.60) 786 (2.00) CRD  33 (0.41) 78 (0.20) IBD  27 (0.34) 102 (0.26) Malignant Neoplasms 752 (9.41) 2755 (6.99) Trauma  6 (0.05) 19 (0.05) Fractures  119 (1.49) 19 (0.05) Standard DMARDs (over follow-up) 1563 (19.56) 879 (2.23) Biologic DMARDS (over follow-up) 138 (1.73) 43 (0.11) *Age assessed at baseline, comorbidities assessed in the year prior to the index date, and medication use assessed over follow-up  53 02004006008001000120014001600180018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryPsoriatic Cohort (Def B1)FemaleMaleTotal Figure 4.1 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition B1) at Index Date  010002000300040005000600070008000900018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryControl Cohort (Def B1)FemaleMaleTotal Figure 4.2 Age and Sex Characteristics of Control Cohort (Definition B1) at Index Date     54 Over the 34,288 PY of follow-up in the PsD cohort (Definition B1), a total of 115 new cases of depression occurred corresponding to an incidence rate of 3.35 per 1,000 PY.  Correspondingly, the control cohort had a total of 364 new depression cases over 185,198 PY of follow-up, yielding a depression incidence rate of 1.97 per 1,000 PY.  We compared the crude, age- and sex-adjusted, and fully-adjusted odds of incident depression between cohorts over the follow-up as shown in Table 4.3.  The magnitude of effect estimates in the context of incident depression were larger than those in Theme 1, with unadjusted, age- and sex-adjusted, and fully-adjusted OR estimates of 1.57, 1.55, and 1.42, respectively.  Table 4.3 Odds of Incident Depression between Cohorts  Incidence of Depression  per 1,000 PY  Unadjusted OR Depression Age- and Sex- Adjusted OR Depression Fully-Adjusted* OR Depression PsD Cohort Def B1 (N=7,992) 3.35 1.57 (1.27-1.94) 1.55 (1.26-1.92) 1.42 (1.14-1.75) *Adjusted for all baseline covariates and socioeconomic status  Over the course of follow-up, 180 AMI events were observed in the PsD cohort and 794 AMI events were observed in the matched control cohort, corresponding to incidence rates of 5.2 and 4.3 events per 1,000 PY respectively.  We employed Cox proportional hazards models to calculate the univariate hazard ratios (HRs) of AMI according to PsD status, depression, and all covariates described in Section 3.2.3.  Univariate HRs for all variables of interest are displayed in Table 4.4.  Of interest, in these unadjusted analyses, PsD increased the risk of AMI by 23% and incident depression increased by risk by 2.8 times.       55 Table 4.4 Univariate Cox Proportional Hazards Models for Risk of Incident AMI Parameter Hazard Ratio (95% CI) PsD Cohort B1 1.23 (1.04-1.44) Incident depression 2.81 (2.03-3.91) Age (per one year increase) 1.07 (1.07-1.08) Male sex 1.58 (1.38-1.80) CCI* 1.40 (1.35-1.46) Biologics? 1.25 (0.52-3.00) Standard DMARDs? 1.09 (0.84-1.42) Antidepressants? 1.50 (1.30-1.73) History of Hospitalizations* 1.44 (1.36-1.52) SES (per one quintile increase) 0.93 (0.89-0.98) COPD* 1.75 (1.42-2.14) Obesity* 3.22 (1.04-10.00) Alcoholism/Liver Disease* 1.70 (0.81-3.58) Cerebrovascular Accidents* 6.02 (4.34-8.37) Hypertension* 3.36 (2.94-3.85) Sepsis* 2.26 (1.01-5.03) Varicose Veins* 1.32 (0.66-2.65) Peripheral Vascular Disease* 7.03 (5.27-9.36) Congestive Heart Failure* 7.25 (5.30-9.92) Angina* 4.88 (3.92-6.07) Chronic Renal Disease* 5.62 (2.80-11.26) Inflammatory Bowel Disease* 0.00 (0.00-5.66e^132) Malignant Neoplasms* 1.83 (1.51-2.22) Trauma* 2.20 (0.31-15.63) Fractures* 2.51 (1.14-5.68)         ? assessed over entire follow-up period         * assessed in year prior to index date   Results of the univariate models in Table 4.4 above were used to inform selection of covariates into our fully-adjusted Cox models.  We employed backwards stepwise regression to optimize model fit, with our initial model including all covariates with P values ? 0.05.  We then iteratively removed additional covariates with P values >0.01, starting with the largest P value, until all remaining covariates in the model were significant predictors of AMI risk.  As noted in Section 4.2.5, we decided a priori that PsD status, depression, age, and sex would remain in the final  56 model regardless of significance level.  We also excluded drug use over follow-up in our models owing to co-linearity (although exploratory analyses demonstrated that their inclusion in the model did not materially change effect estimates).  Our final, fully-adjusted model is summarized in Table 4.5.  After full adjustment, PsD showed a trend toward an increased risk of AMI although significance was lost.  Significant predictors included depression (75% increased risk), comorbidity score, increasing age, male sex, low socioeconomic status, hypertension, PVD, and angina.    Table 4.5 Fully-Adjusted Cox Proportional Hazards Model for Risk of Incident AMI Parameter HR (95% CI) PsD Cohort B1 1.12 (0.95-1.32) Incident Depression 1.75 (1.26-2.45) CCI 1.14 (1.09-1.20) Age (per one year increase) 1.07 (1.07-1.08) Male Sex 2.10 (1.84-2.40) SES (per one quintile increase) 0.94 (0.90-0.99) Hypertension* 1.38 (1.20-1.59) Peripheral Vascular Disease* 1.89 (1.40-2.55) Angina* 1.71 (1.36-2.14)        * assessed in year prior to index date  To investigate whether depression acts as an effect modifier whereby the risk of AMI in PsD varies according to depression status (Aim 2) in the context of cohort definition B1 (incident depression), we first tested the significance of an interaction term for exposure status x incident depression status.  We then stratified our cohorts by depression status and re-ran our final Cox models according to each stratum.  Among those with incident depression, the frequencies of AMI were 8 out of 115 individuals in the PsD cohort and 29 out of 364 individuals in the matched control cohort.  Among those with no depression, the frequencies of AMI were 172 / 7,877 in the PS cohort and 765 / 39,072 in the matched control cohort.  The results of the effect modification models according to incident depression status are displayed in Table 4.6 below.  Likely owing to small  57 sample sizes and large CIs, the results of these analyses failed to reach significance (P value for interaction = 0.09). Table 4.6 Risk Factors for Incident AMI according to Incident Depression Status Parameter HRs among Individuals with Incident Depression (95% CI) HRs among Individuals without Depression (95% CI) PsD Cohort B1 0.79 (0.34-1.85) 1.13 (0.95-1.33) CCI 1.08 (0.81-1.44) 1.15 (1.09-1.20) Age (per one year increase) 1.04 (1.02-1.06) 1.07 (1.07-1.08) Male Sex 1.61 (0.81-3.17) 2.10 (1.83-2.41) SES (per one quintile increase) 1.01 (0.81-1.27) 0.94 (0.90-0.99) Hypertension* 1.52 (0.71-3.24) 1.36 (1.18-1.57) Peripheral Vascular Disease* 0.63 (0.07-5.54) 2.02 (1.49-2.72) Angina* 1.26 (0.35-4.48) 2.76 (1.40-2.21) * assessed in year prior to index date    We further investigated whether sex modifies the risk of incident AMI in patients with incident PsC (Aim 3) in the context of cohort definition B1.  We began by testing the significance of an interaction term between exposure status x sex, yielding a significant P value (P<0.001).  Among 25,401 males, AMI events occurred in 109 of 4,289 in the PsD cohort and 529 of 21,112 in the matched control cohort.  Among 22,027 females, AMI events occurred in 71 of 3,703 individuals in the PsD cohort and 265 of 18,324 in the matched control cohort.  The results of the effect modification analyses by sex are displayed in Table 4.7 below.  Sex modified the effect of PsD status on AMI risk, such that that a diagnosis of PsD only increased the risk of AMI among females.         58 Table 4.7 Risk Factors for Incident AMI according to Sex Parameter HRs among Males (95% CI) HRs among Females (95% CI) PsD Cohort B1 1.02 (0.83-1.26) 1.33 (1.02-1.74) Incident Depression 1.79 (1.10-2.90) 1.63 (1.02-2.61) CCI 1.13 (1.06-1.20) 1.16 (1.08-1.26) Age (per one year increase) 1.07 (1.06-1.07) 1.08 (1.07-1.09) SES (per one quintile increase) 0.93 (0.88-0.98) 0.97 (0.90-1.05) Hypertension* 1.36 (1.14-1.63) 1.43 (1.14-1.79) Peripheral Vascular Disease* 1.83 (1.26-2.65) 2.08 (1.24-3.50) Angina* 1.71 (2.30-2.26) 1.77 (1.19-2.61) * assessed in year prior to index date    Finally, we investigated whether age acts as an effect modifier on the risk of AMI (Aim 4) in the context of cohort definition B1, whereby the risk imposed by PsD status differs among individuals aged ? 50 vs. < 50.  The results of Cox models stratified by age are revealed in Table 4.8, below.  Although age did not modify the impact of PsD cohort status on AMI risk (P for interaction = 0.17), it did appear to modify the effect of depression on AMI risk, with depression contributing to a six-fold increased risk of AMI among individuals under 50 vs. a twofold increased risk among individuals aged 50 and above (P for interaction <0.01).  Table 4.8 Risk Factors for Incident AMI according to Age Category Parameter HRs among Individuals ?50 (95% CI) HRs among Individuals <50 (95% CI) PsD Cohort B1 1.08 (0.91-1.28) 1.05 (0.63-1.74) Incident Depression 1.88 (1.29-3.73) 6.09 (2.93-12.64) CCI 1.22 (1.17-1.28) 1.10 (0.86-1.41) Male Sex 1.72 (1.49-1.97) 4.09 (2.44-6.87) SES (per one quintile increase) 0.93 (0.88-0.97) 0.92 (0.80-1.06) Hypertension* 1.64 (1.42-1.89) 2.26 (1.17-4.38) Peripheral Vascular Disease* 2.34 (1.73-3.17) 3.03 (0.40-22.90) Angina* 1.94 (1.54-2.44) 2.07 (0.89-8.82) * assessed in year prior to index date    59 4.3.2 Results for Cohort B2 with Prevalent Depression Characteristics of the study sample are summarized in Table 4.9.  The mean ages in the PsD and control cohort were 49 (SD=17) and 48 (SD=17) years, respectively, and both cohorts were composed of 51% males.  A total of 10,041 individuals in the PsD cohort (Definition B2) contributed a total of 46,012 PY of follow-up (mean follow-up = 4.58 years), and controls contributed a total of 232,619 PY of follow-up (mean of 4.91 years).  The age of PsD incidence was approximately normally distributed about the mean for both males and females, although females had a relatively lower rate of PsD incidence between the ages of 28 and 57 compared to males, and some evidence of a bi-modal peak onset limited to females (Figures 4.3 and 4.4).                   60 Table 4.9 Characteristics of Exposed and Unexposed Cohorts (Definition B2)  Covariate PsD Exposure Cohort (N=10,041) N (%) Age- & Sex-Matched Control Cohort (N = 47,415) N (%) Age, Mean (SD) 48.66 (17.01) 48.05 (16.72) Male Sex 5135 (51.14) 24170 (50.98) COPD  873 (8.69) 3132 (6.61) Obesity  39 (0.39) 46 (0.10) Alcoholism/liver disease 105 (1.05) 285 (0.60) CVA  132 (1.31) 377 (0.80) Hypertension 1418 (14.12) 6163 (13.00) Sepsis  48 (0.48) 110 (0.23) Varicose Veins  80 (0.80) 336 (0.71) PVD  129 (1.28) 409 (0.86) CHF  159 (1.58) 371 (0.78) Angina 253 (2.52) 960 (2.02) CRD  40 (0.40) 91 (0.19) IBD  40 (0.40) 125 (0.26) Malignant Neoplasms 996 (9.92) 3374 (7.12) Trauma  8 (0.08) 27 (0.06) Fractures  154 (1.53) 27 (0.06) Standard DMARDs (over follow-up) 2042 (20.34) 1121 (2.36) Biologic DMARDS (over follow-up) 194 (1.93) 53 (0.11) *Age assessed at baseline, comorbidities assessed in the year prior to the index date, and medication use assessed over follow-up         61 0500100015002000250018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryPsoriatic Cohort (Def B2)FemaleMaleTotal Figure 4.3 Age and Sex Characteristics of Psoriatic Disease Cohort (Definition B2) at Index Date  02000400060008000100001200018-27 28-37 38-47 48-57 58-67 68-77 78-87 88 +Age CategoryControl Cohort N (Def B2)FemaleMaleTotal Figure 4.4 Age and Sex Characteristics of Control Cohort (Definition B2) at Index Date     62 Of the 10,041 individuals in the PsD cohort (Definition B2), a total of 2,164 had a code for depression over follow-up, for a prevalence of 21.55%.  Correspondingly, of 47,415 matched controls, 8,349 had depression over follow-up for a prevalence of 17.60%.  We compared the crude, age- and sex-adjusted, and fully-adjusted odds of prevalent depression between cohorts as shown in Table 4.10.  The magnitude of effect estimates in the context of incident depression were larger than those in PsD Definition A and smaller than those in Definition B2, as would be expected in a definition with intermediate precision.  The unadjusted, age- and sex-adjusted, and fully-adjusted OR estimates of depression between the PsD and unexposed control cohort were 1.29, 1.30, and 1.26, respectively.  Table 4.10 Odds of Incident Depression between Cohorts  Prevalence of Depression     N (%)  Unadjusted OR Depression Age- and Sex- Adjusted OR Depression Fully-Adjusted* OR Depression PsD Cohort Def B2 (N=10,041) 2164 (21.55) 1.29 (1.22-1.36) 1.30 (1.23-1.37) 1.26 (1.19-1.33) *Adjusted for all baseline covariates and socioeconomic status    Over the entire follow-up period, 268 incident AMI events occurred in the PsD cohort vs. 983 events in the matched control cohort, corresponding to incidence rates of 5.8 and 4.2 events per 1,000 PY, respectively.  Cox proportional hazards models were used to calculate the univariate hazard ratios (HRs) of AMI according to PsD status, depression, and all covariates described in Section 3.2.3.  Univariate HRs for all variables of interest are displayed in Table 4.11.  In these unadjusted analyses, PsD increased the risk of AMI by 38% and prevalent depression increased by risk by 16%.   63 Table 4.11 Univariate Cox Proportional Hazards Models for Risk of Incident AMI Parameter Hazard Ratio (95% CI) PsD Cohort Def B2 1.38 (1.21-1.58) Prevalent depression 1.16 (1.02-1.32) Age (per one year increase) 1.07 (1.07-1.08) Male sex 1.56 (1.39-1.75) CCI* 1.40 (1.36-1.45) Biologics? 1.37 (0.68-2.74) Standard DMARDs? 1.13 (0.90-1.40) Antidepressants? 1.29 (1.15-1.45) History of Hospitalizations* 1.31 (1.27-1.36) SES (per one quintile increase) 0.93 (0.90-0.97) COPD* 1.70 (1.43-2.03) Obesity* 1.98 (0.64-6.16) Alcoholism/Liver Disease* 1.45 (0.82-2.55) Cerebrovascular Accidents* 6.85 (5.25-8.94) Hypertension* 3.27 (2.90-3.69) Sepsis* 2.30 (1.15-4.61) Varicose Veins* 1.18 (0.63-2.20) Peripheral Vascular Disease* 6.61 (5.10-5.57) Congestive Heart Failure* 6.57 (4.95-8.72) Angina* 4.66 (3.83-5.67) Chronic Renal Disease* 6.18 (3.41- 11.18) Inflammatory Bowel Disease* 0.61 (0.15-2.43) Malignant Neoplasms* 1.80 (1.52-2.13) Trauma* 1.43 (0.20-10.16) Fractures* 3.03 (1.63-5.65)                 ? assessed over entire follow-up period                 * assessed in year prior to index date   We then employed a backward stepwise Cox regression modeling procedure, with the results of the univariate models in Table 4.11 used to inform selection using the same methods as described for PsD cohort definition B1.  Our final, fully-adjusted model is shown in Table 4.12.  After full adjustment, PsD was found to increase the risk of AMI by 26% and depression by 29%.  Other significant predictors that remained in the final model included comorbidity score, increasing age, male sex, low socioeconomic status, and histories of CVA, hypertension, PVD, and angina.    64 Table 4.12 Fully-Adjusted Cox Proportional Hazards Model for Risk of Incident AMI Parameter HR (95% CI) PsD Cohort Def B2 1.25 (1.09-1.44) Prevalent Depression 1.28 (1.13-1.46) CCI 1.15 (1.10-1.20) Age (per one year increase) 1.07 (1.06-1.07) Male Sex 1.96 (1.74-2.20) SES (per one quintile increase) 0.94 (0.91-0.98) Cerebrovascular Accidents* 1.45 (1.09-1.92) Hypertension* 1.35 (1.19-1.53) Peripheral Vascular Disease* 1.69 (1.29-2.22) Angina* 1.65 (1.34-2.01)             * assessed in year prior to index date  We then employed this cohort definition to investigate whether depression acts as an effect modifier on the risk AMI in PsD (Aim 2).  We found a significant interaction between exposure status and depression status in our final model (P=0.01), indicating effect modification.  We thus stratified our cohorts by depression status and re-ran our final Cox models according to each stratum.  Among those with prevalent depression, 96 of 2,124 in the PsD cohort experienced an AMI and 218 of 8,343 in the control cohort experienced an AMI over follow-up.  Among those without any diagnoses of depression, the frequencies of AMI were 172 of 7,877 individuals in the PsD cohort and 765 out of 39,072 in the matched control cohort.  These higher frequencies of depression provided improved power for the planned analyses of effect modification.  The results of these analyses according to prevalent depression status are displayed in Table 4.13 below.  Interestingly, depression modified the effect of PsD status on the risk of AMI, such that an increased risk was only observed among individuals with prevalent depression over follow-up.      65 Table 4.13 Risk Factors for Incident AMI according to Prevalent Depression Status Parameter HRs among Individuals with Prevalent Depression (95% CI) HRs among Individuals without Depression (95% CI) PsD Cohort Def B2 1.59 (1.25-2.04) 1.13 (0.95-1.33) CCI 1.16 (1.06-1.27) 1.14 (1.08-1.20) Age (per one year increase) 1.05 (1.05-1.06) 1.07 (1.07-1.08) Male Sex 1.65 (1.32-2.06) 2.10 (1.83-2.41) SES (per one quintile increase) 0.94 (0.87-1.02) 0.94 (0.90-0.99) Cerebrovascular Accidents* 2.01 (1.26-3.21) 1.29 (0.90-1.83) Hypertension* 1.28 (0.99-1.68) 1.35 (1.17-1.57) Peripheral Vascular Disease* 1.10 (0.60-2.03) 1.97 (1.45-2.67) Angina* 1.40 (0.91-2.46) 1.76 (1.40-2.21) * assessed in year prior to index date    We next investigated whether sex modifies the risk of incident AMI in patients with incident PsC (Aim 3) in the context of cohort definition B2.  We tested the significance of an interaction term between exposure status and sex, yielding a P value = 0.01 suggesting effect modification.  Among 29,304 males, AMI events occurred in 150 of 5,135 in the PsD cohort and 629 of 24,170 in the matched control cohort.  Among 28,151 females, AMI events occurred in 118 of 4,906 individuals in the PsD cohort and 354 of 23,245 individuals in the matched control cohort.  The results of the effect modification analyses by sex are displayed in Table 4.14 below.  Sex modified the effect of PsD status on AMI risk, such that a diagnosis of PsD only increased the risk of AMI among females, similar to the results of the Definition B1 analysis.        66 Table 4.14 Risk Factors for Incident AMI according to Sex Parameter HRs among Males (95% CI) HRs among Females (95% CI) PsD Cohort Def B2 1.10 (0.92-1.32) 1.55 (1.26-1.92) Prevalent Depression 1.21 (1.01-1.44) 1.40 (1.56-1.70) CCI 1.13 (1.07-1.20) 1.16 (1.09-1.24) Age 1.07 (1.06-1.07) 1.07 (1.06-1.08) SES (per one quintile increase) 0.94 (0.89-0.98) 0.96 (0.90-1.02) Cerebrovascular Accidents* 1.18 (0.81-1.74) 2.06 (1.37-3.10) Hypertension* 1.28 (1.09-1.52) 1.42 (1.17-1.73) Peripheral Vascular Disease* 1.60 (1.13-2.27) 2.05 (1.32-3.18) Angina* 1.63 (1.26-2.11) 1.77 (1.28-2.44) * assessed in year prior to index date    Finally, we investigated whether age acts as an effect modifier on the risk of AMI (Aim 4) in the context of cohort definition B2, whereby the risk imposed by PsD status differs among individuals aged ? 50 vs. <50.  We tested the need to stratify by age by testing the significance of an interaction term of exposure status x age (dichotomized at age 50) in the final model, which yielded significance (P<0.001).  Table 4.15 displays the results of the final Cox models stratified by age.  Among those aged ? 50, 223 AMI events occurred out of 4,655 individuals in the PsD cohort and 855 events occurred among 21,358 individuals in the unexposed cohort.  Conversely, among those aged <50, 45 AMI events occurred among 5,386 individuals in the PsD cohort and 128 events occurred among 26,057 individuals in the unexposed cohort.   As shown in Table 4.15, age modified the effects of PsD cohort status and prevalent depression on the risk of AMI, with an elevated 63% increased risk of AMI found among PsD patients aged < 50 vs. no increased risk found in those ? 50, and furthermore, no increased risk of AMI was observed among depressed individuals aged ? 50 whereas depression led to 2.3 times the risk of AMI among those aged < 50.       67 Table 4.15 Risk Factors for Incident AMI according to Age Category Parameter HRs among Individuals ?50 (95% CI) HRs among Individuals <50 (95% CI) PsD Cohort Def B2 1.14 (0.98-1.32) 1.63 (1.15-2.30) Prevalent Depression 1.14 (0.98-1.31) 2.34 (1.71-3.18) CCI 1.21 (1.16-1.26) 1.20 (1.04-1.38) Male Sex 1.66 (1.47-1.88) 2.86 (2.02-4.06) SES (per one quintile increase) 0.92 (0.89-0.96) 0.95 (0.85-1.06) Cerebrovascular Accidents* 1.85 (1.39-2.47) 8.41 (2.65-26.58) Hypertension* 1.54 (1.35-1.76) 2.99 (1.86-4.78) Peripheral Vascular Disease* 2.09 (1.58-2.76) 3.50 (0.86-14.23) Angina* 1.90 (1.55-2.34) 1.72 (0.54-5.54) * assessed in year prior to index date                   68 CHAPTER 5 DISCUSSION OF RESULTS  5.1 Overview of Key Findings   We conducted population-based studies of the burden and cardiovascular impact of depression in the context of psoriatic disease (PsD).  Although previous research has estimated the burden of depression in PsD using various methods, to the best of our knowledge, no Canadian studies have estimated the burden of physician-diagnosed depression in PsD, an important measure of health care utilization for the purposes of resource allocation.  Further, we are unaware of any studies that have specifically assessed the impact (independent or otherwise) of depression on the risk of CV events among individuals with PsC or PsA, who represent a highly-relevant patient population given their well-established increased risk of CV morbidity and mortality.32  Thus, we believe that this thesis work represents the first-ever study of the population burden and CV impact of depression in PsD, filling major knowledge gaps in the field.   In our most sensitive PsD cohort (Definition A), we found that the population burden of depression was substantial with a prevalence of almost 25%.  Data from our more specific cohorts (Definitions B1&2) confirmed the high burden of depression with an incidence rate of 3.4 per 1,000 PY and a prevalence of 21.6% over a mean of 4.6 years follow-up.  Across cohorts, the burden of depression was higher than that observed in age-, sex-, and index date-matched controls, with stronger effects observed with the use of increasingly specific definitions of PsD.  We further demonstrated that depression is an independent risk factor for incident AMI among individuals with incident PsD, with incident depression imposing a 75% increased risk after adjusting for all available risk factors.  The corresponding risk estimate for incident AMI associated with prevalent  69 depression in the context of incident PsD was attenuated slightly at 28%, suggesting a period of high risk associated with new-onset depression, perhaps owing to uncontrolled inflammation or other behavioural factors.  Interestingly, the addition of depression in our multivariate models for AMI attenuated the observed risk associated with PsD, suggesting colinearity (and thus, shared pathological pathways) between depression and PsD.  Other established risk factors for AMI that were confirmed in our Cox models of PsD patients included male sex and a history of CV risk factors including cerebrovascular accidents, hypertension, angina, and peripheral vascular disease.  Interestingly, baseline socioeconomic status (SES) also significantly predicted the risk of AMI in all models, with each quintile increase in SES being associated with an approximately 7% decreased risk of developing an AMI event.   Our analyses of effect modification, particularly in the case of depression, also yielded some remarkable results.  While our incident depression analysis was insufficiently powered to measure the impact of PsD status according to depression strata (e.g., only 8 AMI events among PsD individuals with depression over follow-up resulting in large confidence intervals and model instability), our prevalent depression analysis showed clear effect modification.  Specifically, PsD patients without a co-diagnosis of depression had no increased risk of CV events in our final model whereas those with a co-diagnosis of depression had a 59% increased risk, independent of age, sex, SES, comorbid medical conditions, and history of hospitalizations.  As increasing age and male sex are well-known risk factors for AMI, we additionally stratified our risk estimates for AMI by these variables to assess their impact on the effect of PsD status.  In both cohorts, sex was found to modify the effect of PsD status on AMI risk, with an increased risk only found among females.  Finally, our effect modification analysis in our prevalent depression cohort (Def B2) for sex corroborated previous findings that AMI risk in PsC interacts with age,14 with no increased risk found among those aged 50 and above compared to a 63% increased risk found among those  70 younger than 50.  These data suggest that any increased risk of AMI associated with PsD becomes negligible in older populations when the risk associated with advanced age outweighs the competing risk imposed by PsD.  A related but unexpected finding was that age also appeared to modify the effect of depression on the risk of AMI, such that depression did not independently predict the risk of AMI in older populations but imposed a 2.3 fold increased risk of AMI among individuals under the age of 50, who would naturally have lower baseline risk factors for AMI.  Collectively, these data help to carve out key high-risk target populations for whom psoriatic disease may impart a particularly large relative risk of CV.    Given the emerging appreciation of the importance of CV risk factor screening in the context of PsD, these data move the field forward by identifying key subpopulations that are particularly sensitive to the effects of PsD on AMI risk.  Rather than a blanket approach that treats all psoriatic patients as high risk, these data help to refine screening protocols to target CV risk factors among high-risk patients.    While the high rate of psychiatric morbidity in PsD has garnered considerable attention in recent years, the effects of such morbidity are generally viewed exclusively in terms of their psychological, behavioural, and quality of life impacts on patients.  Correspondingly, considerable research has assessed the burden of depression in PsD35-38,122,134,145,146,185,247 but no studies (of which we are aware) have attempted to assess the independent impact of depression on CVD risk in the high risk population of PsD.  In light of strong evidence that depression is itself an independent risk factor for CV morbidity and mortality,72,74,75,189,191 the lack of data on its effect in PsD represents an important evidence gap.  Results of the current analyses help to bridge this gap, and support our hypotheses for an important impact of depression on CV risk in this patient population.      71  Strengths and limitations of our study deserve comment.  The use of population-based cohorts of PsD patients reduces selection-bias and increases external validity of our findings by representing the geographic and demographic diversity of PsD patients in BC.  For our descriptive study of the burden of depression, our algorithm for PsD yielded prevalence estimates that converged with previous population estimates of PsD prevalence, suggesting reasonable generalizability.  Further, large sample sizes afforded through the use of administrative databases allow reasonable power for analyses of the relationships between relatively infrequent exposures and outcomes.  Such large sample sizes allow for the utilization of highly specific disease definitions to ensure high PPVs of exposures and outcomes of interest to reduce error while maintaining sufficient power to capture infrequent events.  The universal nature of Canada?s health care system makes PopData a rich data source with comprehensive information on all physician visits, hospitalizations, medications, procedures, sociodemographics, and vital statistics on almost all of BC?s 4.3 million residents, with up to 10 years of follow-up data to study the temporal relationship between diseases.  However, the use of administrative data for the purpose of research is not without its limitations.  As these data have not been collected for the purpose of research, they often fail to capture key variables of interest such as lifestyle factors or clinical and laboratory features of disease.  Although some data on factors such as obesity and alcoholism were included in our analyses, data capture would be largely incomplete leading to marginal confounding.  While we observed an effect with SES, it is important to note that these data are largely based on imputation of income by neighbourhood, and individuals may frequently move without updating their MSP status.  Thus, some degree of marginal confounding would be expected for this variable.  Additional lifestyle factors known to be associated with our exposures and outcomes of interest including smoking were also unavailable in the database.  Finally, as is the case with all administrative data, some misclassification of exposures and outcomes will occur.   72 For our cohort definitions, we were cognizant of the importance of balancing sensitivity (generalizability) and specificity (ability to rule out misclassified cases) according to the intent of the analysis.  In our Definition A, which was intended to describe the population scope of the problem, we allowed inclusion of non-specific codes for psoriasis and similar disorders (ICD-9 = 696.x) from specialists, under the assumption that dermatologist coding would generally correspond to PsC and rheumatologist coding would correspond to PsA.  This led to a population prevalence of 1.4% for PsD, which converged with previous estimates.  To reduce misclassification error in our Cox analyses, we restricted inclusion to specific codes for PsC or PsA, which substantially reduced our sample size and the ratio of PsC to PsA patients.  Thus, while this definition was likely to be more specific to PsD, many true PsC patients were likely excluded, decreasing the representativeness of the sample to the spectrum of PsD in the general population.  These data suggest a coding bias in which dermatologists tend to employ non-specific coding for PsC whereas rheumatologists may tend to employ more specific coding.  This is likely because PsC represents the primary condition in the 696.x classification, leading rheumatologists to explicitly differentiate PsA diagnoses from PsC with the additional digit.  Thus, it should be noted that the results of our survival analyses should be interpreted in light of a high proportion of individuals with more severe disease (PsA) on the PsD spectrum.  The slightly higher prevalences of depression observed in our more specific cohorts may suggest reduced coding error in these contexts or a stronger relationship between depression and PsA than PsC.  It is worth noting that we employed previously validated exposure and outcome definitions with reasonable PPVs and proven utility for comparing disease relationships in administrative data contexts to avoid the impact of misclassification error236,237,242-244,248 (especially in our Cox analysis cohorts).  Although covariate and outcome data have been largely validated in Canadian contexts, our definitions for PsD have only been validated in US settings to date.  Thus, future validation studies on the positive predictive value of PsD diagnostic  73 codes would be valuable in Canadian administrative data contexts.  Nonetheless, any non-differential misclassification of exposures or outcomes would likely bias estimates toward the null by introducing error and essentially ?watering down? the effect.  Thus, the strength of relationships between diseases is more likely to be underestimated than overestimated in administrative data contexts.  74 CHAPTER 6 CONCLUSIONS  The body of work presented in this thesis underscores a major burden and important independent impact of depression in shaping cardiovascular risk among individuals with psoriasis and psoriatic arthritis.  In this concluding section, we highlight key research findings within the contexts of current practice and policy, and discuss recommendations for future research on this important topic.  6.1 Implications of Key Findings   A number of clinically important findings emerge from this work.  The population-burden of depression was found to be substantial among patients with PsD, corresponding to a higher incidence and prevalence than that seen in healthy controls, independent of age, sex, SES, comorbidities, and health care utilization.  Depression, which alone can have devastating effects on a person?s well-being, has been shown to have many adverse consequences in the context of PsD and other chronic diseases, including increased levels of fatigue, increased perceptions of pain, reduced quality of life, decreased ability to manage one?s disease through healthy behaviours and medication adherence (due to decreased self-efficacy and concentration), disruptions in eating and sleeping, and increases in unhealthy behaviours like smoking and excessive alcohol consumption.  Essentially all of these symptoms are common to patients with PsC and PsA, suggesting that depression could compound the clinical burden of disease in these patients, increasing the complexity of effective treatment.  Further, we have described evidence of a strong pro-inflammatory component to depression pathogenesis and maintenance, including several shared  75 etiological underpinnings with PsD.  Finally, we have reviewed the evidence on both depression and PsD leading to an atherosclerotic state and ultimately increasing the risk of CV events.  In light of these many shared risk factors, disease pathways, symptoms, and critical outcomes, we hypothesized that depression could compound the problem of CVD in PsD, and thus investigated the long term consequences of a co-diagnosis of depression in the context of PsD.  Results from our longitudinal cohort studies on the risk of CV events associated with depression in PsD revealed several key findings with the potential to improve patient care and critical outcomes.  First, in our fully incident analysis, depression increased the risk of AMI by 75% in the context of PsD.  Furthermore, prevalent depression modified the effect of PsD on the risk of AMI such that an increased risk was only observed among patients with depression over the follow-up.  Moreover, we identified effect modification for PsD status on the risk of AMI associated with 1) sex, with females demonstrating an increased risk while males did not; and 2) age, with individuals under 50 showing an increased risk while those above 50 did not.  Finally, we found that both age and sex modified the effect of depression status on the risk of AMI, such that depression only increased the risk of AMI among females and those under the age of 50.  Identification of this key subset of patients who are: 1) depressed; 2) female; 3) younger than 50 allows for targeted management of CV risk factors among PsD patients with these particular traits.  As CVD obviously represents an extremely costly health outcome in terms of both health resource utilization and loss of life, preventive measures should be taken to circumvent such outcomes.  However, with limited health resources and heavy clinical loads among practitioners, a targeted strategy to identify high-risk individuals represents the most cost-effective and feasible approach to improve long-term health outcomes in psoriatic individuals.    The complex interrelationship and high comorbidity load between PsD and depression suggests that treating physicians should consider all the ways in which skin, joint, and depressive  76 symptoms can interact to affect general health rather than focusing strictly on skin and joint disease.  Considering that multiple symptoms associated with comorbidities of PsD can further fuel disease progression and long-term adverse health outcomes, it is important to view PsC and PsA as systemic diseases with a need to comprehensively address comorbidity load and the interrelated symptoms and risk factors of such comorbidity.  This may include consideration of the mental and CV health benefits of therapeutic approaches in addition to the primary focus of skin and joint improvement.  Ideal care of complex PsC and PsA patients would involve an integrated and preventive treatment strategy with collaboration with affiliated health care professionals as required, ranging from social workers and psychologists to clinical dietitians and cardiologists.117,247,249  Barriers to overcome in screening for depression among patients with PsD include patients? reluctance to report mental health symptoms due to fear of stigmatization.  One telephone survey of 1054 US individuals found that almost half of respondents endorsed reluctance to disclose symptoms of depression due to factors including belief that such discussions fall outside the realm of primary care, fear of antidepressant medication, and concerns about stigmatization resulting from being labeled as a ?psychiatric patient?.161  Moreover, there may be ?a disconnect? between physicians? and patients? perceptions of depression symptoms, or barriers to treatment or referral from a physician perspective as well.  One study of 43 consultations between 5 dermatologists and their patients in a UK clinic revealed that participating dermatologists were generally unable to identify psychological distress among their patients, regardless of level of physician empathy.  Further, this study revealed that even when a mental disorder such as depression was identified, the caring physician failed to take any action (treatment or referral) in the majority of cases.247  Further research is needed on the potential reasons for non-referral in  77 order to overcome treatment barriers and improve screening of depression in dermatology, rheumatology, and general practice settings.  6.2 Future Research Directions   The administrative data context of this thesis has made it impossible to assess for several relevant covariates of interest that could confound the relationships between exposures and outcomes.  Thus, these data call for confirmation in other longitudinal cohorts enriched with lifestyle and behavioural variables.  Additionally, cohort studies with larger sample sizes or longer follow-up time would allow for fully incident analyses of effect modification.  It would additionally be useful to assess the impact of depression on AMI risk in other PsD populations to better generalize findings.  Further validation studies of the positive predictive values of PsD codes in Canadian administrative data contexts would also be valuable.  Finally, it would be useful to assess the potential effect of depression on other critical outcomes in PsD, such as PVD, VTE, stroke, and CV-mortality.  While preliminary research has investigated the potential anti-inflammatory effects of antidepressant treatment, further research is needed in the field to assess potential long-term benefits using intermediate and clinical outcomes in terms of both PsD and AMI.  To more comprehensively address the full-spectrum of disease burden in PsD, head-to-head trials or longitudinal cohort studies of biologic and other therapies could assess the efficacy/effectiveness of such therapies on the burden of depressive and CV comorbidity in addition to skin and joint disease.    Finally, further research is needed to identify barriers to patient reporting and physician screening / treatment of depression and CV risk factors in dermatology, rheumatology, and general  78 practice settings.  Closing these gaps in care may yield important benefits for quality of life, disease management, critical outcomes, and overall survival of PsD patients.  6.3 Conclusions   These population-based data provide evidence that depression is a prevalent and independent risk factor for AMI among individuals with PsD.  Moreover, our findings suggest that depression, in addition to age and sex, acts as a major effect modifier in the context of PsD, such that PsD leads to an increased risk of AMI only among 1) individuals with depression, 2) individuals under 50 years of age, and 3) females.  Thus, we have identified three subsets of patients for whom PsD imposes a particularly high risk for CV outcomes.  These data underscore the need to actively screen for depression among PsC and PsA patients and closely monitor CV health in these high-risk patient populations to reduce critical outcomes and improve long-term survival.   79 References  1. Griffiths CE, Barker JN. Pathogenesis and clinical features of psoriasis. Lancet. Jul 21 2007;370(9583):263-271. 2. Raychaudhuri SP. A Cutting Edge Overview: Psoriatic Disease. Clin Rev Allergy Immunol. Feb 23 2012;23:23. 3. Dhir V, Aggarwal A. Psoriatic Arthritis: a Critical Review. Clin Rev Allergy Immunol. Feb 1 2012;1:1. 4. Gladman DD. Psoriatic arthritis. Dermatol Ther. Jan-Feb 2009;22(1):40-55. 5. Wollina U, Unger L, Heinig B, Kittner T. Psoriatic arthritis. Dermatol Ther. Mar-Apr 2010;23(2):123-136. 6. Armstrong AW, Schupp C, Bebo B. Psoriasis comorbidities: results from the National Psoriasis Foundation surveys 2003 to 2011. Dermatology. 2012;225(2):121-126. 7. Abuabara K, Azfar RS, Shin DB, Neimann AL, Troxel AB, Gelfand JM. Cause-specific mortality in patients with severe psoriasis: a population-based cohort study in the U.K. Br J Dermatol. Sep 2010;163(3):586-592. 8. Ahlehoff O, Gislason GH, Charlot M, et al. Psoriasis is associated with clinically significant cardiovascular risk: a Danish nationwide cohort study. J Intern Med. Aug 2011;270(2):147-157. 9. Ahlehoff O, Gislason GH, Lindhardsen J, et al. Psoriasis carries an increased risk of venous thromboembolism: a Danish nationwide cohort study. PLoS ONE. 2011;6(3):e18125. 10. Arumugam R, McHugh NJ. Mortality and causes of death in psoriatic arthritis. J Rheumatol Suppl. Jul 2012;89:32-35.  80 11. Buckley C, Cavill C, Taylor G, et al. Mortality in psoriatic arthritis - a single-center study from the UK. J Rheumatol. Oct 2010;37(10):2141-2144. 12. Federman DG, Shelling M, Prodanovich S, Gunderson CG, Kirsner RS. Psoriasis: an opportunity to identify cardiovascular risk. Br J Dermatol. Jan 2009;160(1):1-7. 13. Gelfand JM, Mehta NN, Langan SM. Psoriasis and cardiovascular risk: strength in numbers, part II. J Invest Dermatol. May 2011;131(5):1007-1010. 14. Gelfand JM, Neimann AL, Shin DB, Wang X, Margolis DJ, Troxel AB. Risk of myocardial infarction in patients with psoriasis. Jama. Oct 11 2006;296(14):1735-1741. 15. Gelfand JM, Troxel AB, Lewis JD, et al. The risk of mortality in patients with psoriasis: results from a population-based study. Arch Dermatol. Dec 2007;143(12):1493-1499. 16. Ghazizadeh R, Shimizu H, Tosa M, Ghazizadeh M. Pathogenic mechanisms shared between psoriasis and cardiovascular disease. Int J Med Sci. 2010;7(5):284-289. 17. Gladman DD, Ang M, Su L, Tom BD, Schentag CT, Farewell VT. Cardiovascular morbidity in psoriatic arthritis. Ann Rheum Dis. Jul 2009;68(7):1131-1135. 18. Han C, Robinson DW, Jr., Hackett MV, Paramore LC, Fraeman KH, Bala MV. Cardiovascular disease and risk factors in patients with rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis. J Rheumatol. Nov 2006;33(11):2167-2172. 19. Husted JA, Thavaneswaran A, Chandran V, et al. Cardiovascular and other comorbidities in patients with psoriatic arthritis: a comparison with patients with psoriasis. Arthritis Care Res (Hoboken). Dec 2011;63(12):1729-1735. 20. Kimball AB, Wu Y. Cardiovascular disease and classic cardiovascular risk factors in patients with psoriasis. Int J Dermatol. Nov 2009;48(11):1147-1156. 21. Kimhi O, Caspi D, Bornstein NM, et al. Prevalence and risk factors of atherosclerosis in patients with psoriatic arthritis. Semin Arthritis Rheum. Feb 2007;36(4):203-209.  81 22. Mehta NN, Azfar RS, Shin DB, Neimann AL, Troxel AB, Gelfand JM. Patients with severe psoriasis are at increased risk of cardiovascular mortality: cohort study using the General Practice Research Database. Eur Heart J. Apr 2010;31(8):1000-1006. 23. Prey S, Paul C, Bronsard V, et al. Cardiovascular risk factors in patients with plaque psoriasis: a systematic review of epidemiological studies. J Eur Acad Dermatol Venereol. Apr 2010;24 Suppl 2:23-30. 24. Prodanovich S, Kirsner RS, Kravetz JD, Ma F, Martinez L, Federman DG. Association of psoriasis with coronary artery, cerebrovascular, and peripheral vascular diseases and mortality. Arch Dermatol. Jun 2009;145(6):700-703. 25. Ramonda R, Lo Nigro A, Modesti V, et al. Atherosclerosis in psoriatic arthritis. Autoimmun Rev. Oct 2011;10(12):773-778. 26. Stern RS, Huibregtse A. Very severe psoriasis is associated with increased noncardiovascular mortality but not with increased cardiovascular risk. J Invest Dermatol. May 2011;131(5):1159-1166. 27. Tam LS, Tomlinson B, Chu TT, et al. Cardiovascular risk profile of patients with psoriatic arthritis compared to controls--the role of inflammation. Rheumatology (Oxford). May 2008;47(5):718-723. 28. Tobin AM, Veale DJ, Fitzgerald O, et al. Cardiovascular disease and risk factors in patients with psoriasis and psoriatic arthritis. J Rheumatol. Jul 2010;37(7):1386-1394. 29. Vena GA, Vestita M, Cassano N. Psoriasis and cardiovascular disease. Dermatol Ther. Mar-Apr 2010;23(2):144-151. 30. Wong K, Gladman DD, Husted J, Long JA, Farewell VT. Mortality studies in psoriatic arthritis: results from a single outpatient clinic. I. Causes and risk of death. Arthritis Rheum. Oct 1997;40(10):1868-1872.  82 31. Wu Y, Mills D, Bala M. Psoriasis: cardiovascular risk factors and other disease comorbidities. J Drugs Dermatol. Apr 2008;7(4):373-377. 32. Xu T, Zhang YH. Association of psoriasis with stroke and myocardial infarction: meta-analysis of cohort studies. Br J Dermatol. Dec 2012;167(6):1345-1350. 33. Zhu TY, Li EK, Tam LS. Cardiovascular risk in patients with psoriatic arthritis. Int J Rheumatol. 2012;2012(10):714321. 34. Zoller B, Li X, Sundquist J, Sundquist K. Risk of pulmonary embolism in patients with autoimmune disorders: a nationwide follow-up study from Sweden. Lancet. Jan 21 2012;379(9812):244-249. 35. Biljan D, Laufer D, Filakovic P, Situm M, Brataljenovic T. Psoriasis, mental disorders and stress. Coll Antropol. Sep 2009;33(3):889-892. 36. Esposito M, Saraceno R, Giunta A, Maccarone M, Chimenti S. An Italian study on psoriasis and depression. Dermatology. 2006;212(2):123-127. 37. Schmitt J, Ford DE. Psoriasis is independently associated with psychiatric morbidity and adverse cardiovascular risk factors, but not with cardiovascular events in a population-based sample. J Eur Acad Dermatol Venereol. Aug 2010;24(8):885-892. 38. Husted JA, Tom BD, Farewell VT, Gladman DD. Longitudinal study of the bidirectional association between pain and depressive symptoms in patients with psoriatic arthritis. Arthritis Care Res (Hoboken). May 2012;64(5):758-765. 39. Kleyn CE, McKie S, Ross AR, et al. Diminished neural and cognitive responses to facial expressions of disgust in patients with psoriasis: a functional magnetic resonance imaging study. J Invest Dermatol. Nov 2009;129(11):2613-2619.  83 40. Callis Duffin K, Wong B, Horn EJ, Krueger GG. Psoriatic arthritis is a strong predictor of sleep interference in patients with psoriasis. J Am Acad Dermatol. Apr 2009;60(4):604-608. 41. Cauli A, Gladman DD, Mathieu A, et al. Patient global assessment in psoriatic arthritis: a multicenter GRAPPA and OMERACT study. J Rheumatol. May 2011;38(5):898-903. 42. Gupta MA, Gupta AK. Psoriasis and sex: a study of moderately to severely affected patients. Int J Dermatol. Apr 1997;36(4):259-262. 43. Husted JA, Gladman DD, Farewell VT, Cook RJ. Health-related quality of life of patients with psoriatic arthritis: a comparison with patients with rheumatoid arthritis. Arthritis Rheum. Apr 2001;45(2):151-158. 44. Krueger G, Koo J, Lebwohl M, Menter A, Stern RS, Rolstad T. The impact of psoriasis on quality of life: results of a 1998 National Psoriasis Foundation patient-membership survey. Arch Dermatol. Mar 2001;137(3):280-284. 45. Langley RG, Krueger GG, Griffiths CE. Psoriasis: epidemiology, clinical features, and quality of life. Ann Rheum Dis. Mar 2005;64 Suppl 2(2):ii18-23; discussion ii24-15. 46. Mease PJ, Menter MA. Quality-of-life issues in psoriasis and psoriatic arthritis: outcome measures and therapies from a dermatological perspective. J Am Acad Dermatol. Apr 2006;54(4):685-704. 47. Rosen CF, Mussani F, Chandran V, Eder L, Thavaneswaran A, Gladman DD. Patients with psoriatic arthritis have worse quality of life than those with psoriasis alone. Rheumatology (Oxford). Mar 2012;51(3):571-576. 48. Tejada Cdos S, Mendoza-Sassi RA, Almeida HL, Jr., Figueiredo PN, Tejada VF. Impact on the quality of life of dermatological patients in southern Brazil. An Bras Dermatol. Nov-Dec 2011;86(6):1113-1121.  84 49. Kannan S, Heller MM, Lee ES, Koo JY. The role of tumor necrosis factor-alpha and other cytokines in depression: what dermatologists should know. J Dermatolog Treat. Sep 30 2011;30:30. 50. Yu B, Becnel J, Zerfaoui M, Rohatgi R, Boulares AH, Nichols CD. Serotonin 5-hydroxytryptamine(2A) receptor activation suppresses tumor necrosis factor-alpha-induced inflammation with extraordinary potency. J Pharmacol Exp Ther. Nov 2008;327(2):316-323. 51. Brebner K, Hayley S, Zacharko R, Merali Z, Anisman H. Synergistic effects of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha: central monoamine, corticosterone, and behavioral variations. Neuropsychopharmacology. Jun 2000;22(6):566-580. 52. Christian LM, Franco A, Iams JD, Sheridan J, Glaser R. Depressive symptoms predict exaggerated inflammatory responses to an in vivo immune challenge among pregnant women. Brain Behav Immun. Jan 2010;24(1):49-53. 53. Deverts DJ, Cohen S, DiLillo VG, et al. Depressive symptoms, race, and circulating C-reactive protein: the Coronary Artery Risk Development in Young Adults (CARDIA) study. Psychosom Med. Oct 2010;72(8):734-741. 54. Halder I, Marsland AL, Cheong J, Muldoon MF, Ferrell RE, Manuck SB. Polymorphisms in the CRP gene moderate an association between depressive symptoms and circulating levels of C-reactive protein. Brain Behav Immun. Jan 2010;24(1):160-167. 55. Matthews KA, Schott LL, Bromberger JT, Cyranowski JM, Everson-Rose SA, Sowers M. Are there bi-directional associations between depressive symptoms and C-reactive protein in mid-life women? Brain Behav Immun. Jan 2010;24(1):96-101. 56. Schiepers OJ, Wichers MC, Maes M. Cytokines and major depression. Prog Neuropsychopharmacol Biol Psychiatry. Feb 2005;29(2):201-217.  85 57. Sutcigil L, Oktenli C, Musabak U, et al. Pro- and anti-inflammatory cytokine balance in major depression: effect of sertraline therapy. Clin Dev Immunol. 2007;2007(10):76396. 58. Weinstein AA, Deuster PA, Francis JL, Bonsall RW, Tracy RP, Kop WJ. Neurohormonal and inflammatory hyper-responsiveness to acute mental stress in depression. Biol Psychol. May 2010;84(2):228-234. 59. Wichers M, Maes M. The psychoneuroimmuno-pathophysiology of cytokine-induced depression in humans. Int J Neuropsychopharmacol. Dec 2002;5(4):375-388. 60. Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch Intern Med. Nov 10 2003;163(20):2433-2445. 61. Zhao G, Ford ES, Li C, et al. Serious psychological distress and its associations with body mass index: findings from the 2007 Behavioral Risk Factor Surveillance System. Int J Public Health. Jun 2009;54 Suppl 1:30-36. 62. Chou KL. Chronic fatigue and affective disorders in older adults: evidence from the 2007 British National Psychiatric Morbidity Survey. J Affect Disord. Mar 5 2013;145(3):331-335. 63. Maes M. An intriguing and hitherto unexplained co-occurrence: Depression and chronic fatigue syndrome are manifestations of shared inflammatory, oxidative and nitrosative (IO&NS) pathways. Prog Neuropsychopharmacol Biol Psychiatry. Apr 29 2011;35(3):784-794. 64. Dempe C, Junger J, Hoppe S, et al. Association of anxious and depressive symptoms with medication nonadherence in patients with stable coronary artery disease. J Psychosom Res. Feb 2013;74(2):122-127. 65. DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for noncompliance with medical treatment: meta-analysis of the effects of anxiety and depression on patient adherence. Arch Intern Med. Jul 24 2000;160(14):2101-2107.  86 66. Patten SB, Williams JV, Lavorato DH, Eliasziw M. A longitudinal community study of major depression and physical activity. Gen Hosp Psychiatry. Nov-Dec 2009;31(6):571-575. 67. van Gool CH, Kempen GI, Penninx BW, Deeg DJ, Beekman AT, van Eijk JT. Relationship between changes in depressive symptoms and unhealthy lifestyles in late middle aged and older persons: results from the Longitudinal Aging Study Amsterdam. Age Ageing. Jan 2003;32(1):81-87. 68. Daly EJ, Trivedi MH, Wisniewski SR, et al. Health-related quality of life in depression: a STAR*D report. Ann Clin Psychiatry. Feb 2010;22(1):43-55. 69. World Health Organization. The global burden of disease: 2004 update. 2008. 70. Alboni P, Favaron E, Paparella N, Sciammarella M, Pedaci M. Is there an association between depression and cardiovascular mortality or sudden death? J Cardiovasc Med (Hagerstown). Apr 2008;9(4):356-362. 71. House A, Knapp P, Bamford J, Vail A. Mortality at 12 and 24 months after stroke may be associated with depressive symptoms at 1 month. Stroke. Mar 2001;32(3):696-701. 72. Kellerman QD, Christensen AJ, Baldwin AS, Lawton WJ. Association between depressive symptoms and mortality risk in chronic kidney disease. Health Psychol. Nov 2010;29(6):594-600. 73. Mykletun A, Bjerkeset O, Dewey M, Prince M, Overland S, Stewart R. Anxiety, depression, and cause-specific mortality: the HUNT study. Psychosom Med. May 2007;69(4):323-331. 74. Mykletun A, Bjerkeset O, Overland S, Prince M, Dewey M, Stewart R. Levels of anxiety and depression as predictors of mortality: the HUNT study. Br J Psychiatry. Aug 2009;195(2):118-125.  87 75. Patten SB, Williams JV, Lavorato D, Li Wang J, Khaled S, Bulloch AG. Mortality associated with major depression in a Canadian community cohort. Can J Psychiatry. Nov 2011;56(11):658-666. 76. Lowes MA, Bowcock AM, Krueger JG. Pathogenesis and therapy of psoriasis. Nature. Feb 22 2007;445(7130):866-873. 77. de Korte J, Sprangers MA, Mombers FM, Bos JD. Quality of life in patients with psoriasis: a systematic literature review. J Investig Dermatol Symp Proc. Mar 2004;9(2):140-147. 78. Mukhtar R, Choi J, Koo JY. Quality-of-life issues in psoriasis. Dermatol Clin. Oct 2004;22(4):389-395, viii. 79. Moll JM, Wright V. Psoriatic arthritis. Semin Arthritis Rheum. 1973;3(1):55-78. 80. Zachariae H. Prevalence of joint disease in patients with psoriasis: implications for therapy. Am J Clin Dermatol. 2003;4(7):441-447. 81. Catanoso M, Pipitone N, Salvarani C. Epidemiology of psoriatic arthritis. Reumatismo. 2012;64(2):66-70. 82. Gelfand JM, Gladman DD, Mease PJ, et al. Epidemiology of psoriatic arthritis in the population of the United States. J Am Acad Dermatol. Oct 2005;53(4):573. 83. Alamanos Y, Voulgari PV, Drosos AA. Incidence and prevalence of psoriatic arthritis: a systematic review. J Rheumatol. Jul 2008;35(7):1354-1358. 84. Christophers E, Barker JN, Griffiths CE, et al. The risk of psoriatic arthritis remains constant following initial diagnosis of psoriasis among patients seen in European dermatology clinics. J Eur Acad Dermatol Venereol. May 2010;24(5):548-554. 85. Nograles KE, Brasington RD, Bowcock AM. New insights into the pathogenesis and genetics of psoriatic arthritis. Nature Clinical Practice Rheumatology. 2009;5(2):83.  88 86. Davidovici BB, Sattar N, Prinz J, et al. Psoriasis and systemic inflammatory diseases: potential mechanistic links between skin disease and co-morbid conditions. J Invest Dermatol. Jul 2010;130(7):1785-1796. 87. Lowes MA, Kikuchi T, Fuentes-Duculan J, et al. Psoriasis vulgaris lesions contain discrete populations of Th1 and Th17 T cells. J Invest Dermatol. May 2008;128(5):1207-1211. 88. Lowes MA, Russell CB, Martin DA, Towne JE, Krueger JG. The IL-23/T17 pathogenic axis in psoriasis is amplified by keratinocyte responses. Trends Immunol. Apr 2013;34(4):174-181. 89. Dowlatshahi EA, van der Voort EA, Arends LR, Nijsten T. Markers of systemic inflammation in psoriasis: a systematic review and meta-analysis. Br J Dermatol. Apr 1 2013. 90. Liu Y, Krueger JG, Bowcock AM. Psoriasis: genetic associations and immune system changes. Genes and immunity. Jan 2007;8(1):1-12. 91. Robinson H, Kelly S, Pitzalis C. Basic synovial biology and immunopathology in psoriatic arthritis. J Rheumatol Suppl. Aug 2009;83:14-16. 92. Yoo IS, Lee JH, Song ST, Kim JH, Lee HJ, Kang SW. T-helper 17 cells: the driving force of psoriasis and psoriatic arthritis. Int J Rheum Dis. Dec 2012;15(6):531-537. 93. Raychaudhuri SP. Role of IL-17 in psoriasis and psoriatic arthritis. Clin Rev Allergy Immunol. Apr 2013;44(2):183-193. 94. Mensah KA, Schwarz EM, Ritchlin CT. Altered bone remodeling in psoriatic arthritis. Curr Rheumatol Rep. Aug 2008;10(4):311-317. 95. Gelfand JM, Stern RS, Nijsten T, et al. The prevalence of psoriasis in African Americans: results from a population-based study. J Am Acad Dermatol. Jan 2005;52(1):23-26.  89 96. Kurd SK, Gelfand JM. The prevalence of previously diagnosed and undiagnosed psoriasis in US adults: results from NHANES 2003-2004. J Am Acad Dermatol. Feb 2009;60(2):218-224. 97. Robinson D, Jr., Hackett M, Wong J, Kimball AB, Cohen R, Bala M. Co-occurrence and comorbidities in patients with immune-mediated inflammatory disorders: an exploration using US healthcare claims data, 2001-2002. Curr Med Res Opin. May 2006;22(5):989-1000. 98. Stern RS, Nijsten T, Feldman SR, Margolis DJ, Rolstad T. Psoriasis is common, carries a substantial burden even when not extensive, and is associated with widespread treatment dissatisfaction. J Investig Dermatol Symp Proc. Mar 2004;9(2):136-139. 99. Parisi R, Symmons DP, Griffiths CE, Ashcroft DM. Global epidemiology of psoriasis: a systematic review of incidence and prevalence. J Invest Dermatol. Feb 2013;133(2):377-385. 100. Jacobson CC, Kumar S, Kimball AB. Latitude and psoriasis prevalence. J Am Acad Dermatol. Oct 2011;65(4):870-873. 101. Raychaudhuri SP, Farber EM. The prevalence of psoriasis in the world. Journal of the European Academy of Dermatology and Venereology. 2001;15(1):16-17. 102. Ferr?ndiz C, Bordas, Garc?a P, Puig S, Pujol R, Smand?a A. Prevalence of psoriasis in Spain (Epiderma Project: phase I). Journal of the European Academy of Dermatology and Venereology. 2001;15(1):20-23. 103. Leder RO, Farber EM. The variable incidence of psoriasis in sub-Saharan Africa. Int J Dermatol. Dec 1997;36(12):911-919.  90 104. Icen M, Crowson CS, McEvoy MT, Dann FJ, Gabriel SE, Maradit Kremers H. Trends in incidence of adult-onset psoriasis over three decades: a population-based study. J Am Acad Dermatol. Mar 2009;60(3):394-401. 105. Donker GA, Foets M, Spreeuwenberg P, van der Werf GT. [Management of psoriasis in family practice is now in closer agreement with the guidelines of the Netherlands Society of Family Physicians]. Nederlands tijdschrift voor geneeskunde. Jun 13 1998;142(24):1379-1383. 106. Huerta C, Rivero E, Rodriguez LA. Incidence and risk factors for psoriasis in the general population. Arch Dermatol. Dec 2007;143(12):1559-1565. 107. Henseler T, Christophers E. Psoriasis of early and late onset: characterization of two types of psoriasis vulgaris. J Am Acad Dermatol. Sep 1985;13(3):450-456. 108. Smith AE, Kassab JY, Rowland Payne CM, Beer WE. Bimodality in age of onset of psoriasis, in both patients and their relatives. Dermatology. 1993;186(3):181-186. 109. Ferr?ndiz C, Pujol RM, Garc?a-Patos V, Bordas X, Smand?a JA. Psoriasis of early and late onset: A clinical and epidemiologic study from Spain. Journal of the American Academy of Dermatology. 2002;46(6):867-873. 110. Gladman DD, Shuckett R, Russell ML, Thorne JC, Schachter RK. Psoriatic arthritis (PSA)--an analysis of 220 patients. The Quarterly journal of medicine. Feb 1987;62(238):127-141. 111. Soriano ER, Rosa J, Velozo E, et al. Incidence and prevalence of psoriatic arthritis in Buenos Aires, Argentina: a 6-year health management organization-based study. Rheumatology (Oxford). Apr 2011;50(4):729-734. 112. Love TJ, Zhu Y, Zhang Y, et al. Obesity and the risk of psoriatic arthritis: a population-based study. Ann Rheum Dis. Aug 2012;71(8):1273-1277.  91 113. Farber EM, Nall ML, Watson W. Natural history of psoriasis in 61 twin pairs. Arch Dermatol. Feb 1974;109(2):207-211. 114. Oka A, Mabuchi T, Ozawa A, Inoko H. Current understanding of human genetics and genetic analysis of psoriasis. The Journal of dermatology. Mar 2012;39(3):231-241. 115. Hebert HL, Ali FR, Bowes J, Griffiths CE, Barton A, Warren RB. Genetic susceptibility to psoriasis and psoriatic arthritis: implications for therapy. Br J Dermatol. Mar 2012;166(3):474-482. 116. Chandran V, Raychaudhuri SP. Geoepidemiology and environmental factors of psoriasis and psoriatic arthritis. J Autoimmun. May 2010;34(3):J314-321. 117. Wolters M. Diet and psoriasis: experimental data and clinical evidence. Br J Dermatol. Oct 2005;153(4):706-714. 118. Bulat V, Situm M, Dediol I, Ljubicic I, Bradic L. The mechanisms of action of phototherapy in the treatment of the most common dermatoses. Coll Antropol. Sep 2011;35 Suppl 2:147-151. 119. Zakarian K, Nguyen A, Letsinger J, Koo J. Excimer laser for psoriasis: a review of theories regarding enhanced efficacy over traditional UVB phototherapy. J Drugs Dermatol. Aug 2007;6(8):794-798. 120. Naldi L, Chatenoud L, Linder D, et al. Cigarette smoking, body mass index, and stressful life events as risk factors for psoriasis: results from an Italian case-control study. J Invest Dermatol. Jul 2005;125(1):61-67. 121. Danielsen K, Olsen AO, Wilsgaard T, Furberg AS. Is the prevalence of psoriasis increasing? A 30 year follow-up of a population-based cohort. British Journal of Dermatology. 2013:n/a-n/a.  92 122. Hayes J, Koo J. Psoriasis: depression, anxiety, smoking, and drinking habits. Dermatol Ther. Mar-Apr 2010;23(2):174-180. 123. Mattey DL, Brownfield A, Dawes PT. Relationship between pack-year history of smoking and response to tumor necrosis factor antagonists in patients with rheumatoid arthritis. Journal of Rheumatology. June 2009;36(6):1180. 124. Frankel HC, Han J, Li T, Qureshi AA. The association between physical activity and the risk of incident psoriasis. Arch Dermatol. Aug 2012;148(8):918-924. 125. Luppino FS, de Wit LM, Bouvy PF, et al. Overweight, obesity, and depression: a systematic review and meta-analysis of longitudinal studies. Arch Gen Psychiatry. Mar 2010;67(3):220-229. 126. Bhole VM, Choi HK, Burns LC, et al. Differences in body mass index among individuals with PsA, psoriasis, RA and the general population. Rheumatology (Oxford). Mar 2012;51(3):552-556. 127. Arias-Santiago S, Orgaz-Molina J, Castellote-Caballero L, et al. Atheroma plaque, metabolic syndrome and inflammation in patients with psoriasis. Eur J Dermatol. May-Jun 2012;22(3):337-344. 128. Jacobi A, Kupke C, Behzad M, Hertl M. Comorbidities, metabolic risk profile and health-related quality of life in German patients with plaque-type psoriasis: a cross-sectional prospective study. Int J Dermatol. Feb 22 2013;22(10):1365-4632. 129. Scarpa R. New insights into the concept of psoriatic disease. J Rheumatol Suppl. Jul 2012;89:4-6. 130. Wu JJ, Nguyen TU, Poon KY, Herrinton LJ. The association of psoriasis with autoimmune diseases. J Am Acad Dermatol. Nov 2012;67(5):924-930.  93 131. Mehta NN, Yu Y, Pinnelas R, et al. Attributable risk estimate of severe psoriasis on major cardiovascular events. Am J Med. Aug 2011;124(8):775 e771-776. 132. Khraishi M, MacDonald D, Rampakakis E, Vaillancourt J, Sampalis JS. Prevalence of patient-reported comorbidities in early and established psoriatic arthritis cohorts. Clin Rheumatol. Jul 2011;30(7):877-885. 133. Khalid U, Hansen PR, Gislason GH, et al. Psoriasis and New-Onset Diabetes Mellitus: A Danish nationwide cohort study. Diabetes Care. Mar 14 2013;14:14. 134. Rabin F, Bhuiyan SI, Islam T, Haque MA, Islam MA. Psychiatric and psychological comorbidities in patients with psoriasis- a review. Mymensingh Med J. Oct 2012;21(4):780-786. 135. Setayeshgar S, Whiting SJ, Vatanparast H. Prevalence of 10-year risk of cardiovascular diseases and associated risks in canadian adults: the contribution of cardiometabolic risk assessment introduction. International journal of hypertension. 2013;2013:276564. 136. Mok CC, Ko GT, Ho LY, Yu KL, Chan PT, To CH. Prevalence of atherosclerotic risk factors and the metabolic syndrome in patients with chronic inflammatory arthritis. Arthritis Care Res (Hoboken). Feb 2011;63(2):195-202. 137. Wakkee M, Herings RM, Nijsten T. Psoriasis may not be an independent risk factor for acute ischemic heart disease hospitalizations: results of a large population-based Dutch cohort. J Invest Dermatol. Apr 2010;130(4):962-967. 138. Gelfand JM, Azfar RS, Mehta NN. Psoriasis and cardiovascular risk: strength in numbers. J Invest Dermatol. Apr 2010;130(4):919-922. 139. Gupta MA, Gupta AK, Watteel GN. Perceived deprivation of social touch in psoriasis is associated with greater psychologic morbidity: an index of the stigma experience in  94 dermatologic disorders. Cutis; cutaneous medicine for the practitioner. Jun 1998;61(6):339-342. 140. Richards HL, Fortune DG, Griffiths CE, Main CJ. The contribution of perceptions of stigmatisation to disability in patients with psoriasis. J Psychosom Res. Jan 2001;50(1):11-15. 141. Rapp SR, Feldman SR, Exum ML, Fleischer AB, Jr., Reboussin DM. Psoriasis causes as much disability as other major medical diseases. J Am Acad Dermatol. Sep 1999;41(3 Pt 1):401-407. 142. Sokoll KB, Helliwell PS. Comparison of disability and quality of life in rheumatoid and psoriatic arthritis. J Rheumatol. Aug 2001;28(8):1842-1846. 143. Lundberg L, Johannesson M, Silverdahl M, Hermansson C, Lindberg M. Health-related quality of life in patients with psoriasis and atopic dermatitis measured with SF-36, DLQI and a subjective measure of disease activity. Acta Derm Venereol. Nov-Dec 2000;80(6):430-434. 144. Zachariae H, Zachariae R, Blomqvist K, et al. Quality of life and prevalence of arthritis reported by 5,795 members of the Nordic Psoriasis Associations. Data from the Nordic Quality of Life Study. Acta Derm Venereol. 2002;82(2):108-113. 145. Kotsis K, Voulgari PV, Tsifetaki N, et al. Anxiety and depressive symptoms and illness perceptions in psoriatic arthritis and associations with physical health-related quality of life. Arthritis Care Res (Hoboken). Oct 2012;64(10):1593-1601. 146. Schmitt JM, Ford DE. Role of depression in quality of life for patients with psoriasis. Dermatology. 2007;215(1):17-27.  95 147. American Psychiatric Association. Task Force on DSM-IV. Diagnostic and statistical manual of mental disorders: DSM-IV. 4th ed. Washington, DC: American Psychiatric Association; 1994. 148. Patten SB, Wang JL, Williams JV, et al. Descriptive epidemiology of major depression in Canada. Can J Psychiatry. Feb 2006;51(2):84-90. 149. Hasin DS, Goodwin RD, Stinson FS, Grant BF. Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on Alcoholism and Related Conditions. Arch Gen Psychiatry. Oct 2005;62(10):1097-1106. 150. Kessler RC, Birnbaum H, Bromet E, Hwang I, Sampson N, Shahly V. Age differences in major depression: results from the National Comorbidity Survey Replication (NCS-R). Psychol Med. Feb 2010;40(2):225-237. 151. Alonso J, Angermeyer MC, Bernert S, et al. Prevalence of mental disorders in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr Scand Suppl. 2004;420(420):21-27. 152. Chang SM, Hahm BJ, Lee JY, et al. Cross-national difference in the prevalence of depression caused by the diagnostic threshold. J Affect Disord. Feb 2008;106(1-2):159-167. 153. Kendler KS, Aggen SH, Neale MC. Evidence for Multiple Genetic Factors Underlying DSM-IV Criteria for Major Depression. JAMA psychiatry (Chicago, Ill.). Jun 1 2013;70(6):599-607. 154. de Graaf R, Bijl RV, Smit F, Vollebergh WA, Spijker J. Risk factors for 12-month comorbidity of mood, anxiety, and substance use disorders: findings from the Netherlands Mental Health Survey and Incidence Study. Am J Psychiatry. Apr 2002;159(4):620-629.  96 155. Jordanova V, Stewart R, Goldberg D, et al. Age variation in life events and their relationship with common mental disorders in a national survey population. Soc Psychiatry Psychiatr Epidemiol. Aug 2007;42(8):611-616. 156. Marcus SM, Kerber KB, Rush AJ, et al. Sex differences in depression symptoms in treatment-seeking adults: confirmatory analyses from the Sequenced Treatment Alternatives to Relieve Depression study. Compr Psychiatry. May-Jun 2008;49(3):238-246. 157. Scott KM, Von Korff M, Alonso J, et al. Age patterns in the prevalence of DSM-IV depressive/anxiety disorders with and without physical co-morbidity. Psychol Med. Nov 2008;38(11):1659-1669. 158. Rubio JM, Markowitz JC, Alegria A, et al. Epidemiology of chronic and nonchronic major depressive disorder: results from the national epidemiologic survey on alcohol and related conditions. Depress Anxiety. Aug 2011;28(8):622-631. 159. Satyanarayana S, Enns MW, Cox BJ, Sareen J. Prevalence and correlates of chronic depression in the canadian community health survey: mental health and well-being. Can J Psychiatry. Jun 2009;54(6):389-398. 160. Dunlop DD, Lyons JS, Manheim LM, Song J, Chang RW. Arthritis and heart disease as risk factors for major depression: the role of functional limitation. Med Care. Jun 2004;42(6):502-511. 161. Bell RA, Franks P, Duberstein PR, et al. Suffering in silence: reasons for not disclosing depression in primary care. Annals of family medicine. Sep-Oct 2011;9(5):439-446. 162. Lee S, Tsang A, Huang YQ, et al. The epidemiology of depression in metropolitan China. Psychol Med. May 2009;39(5):735-747. 163. Bogner HR, Gallo JJ. Are higher rates of depression in women accounted for by differential symptom reporting? Soc Psychiatry Psychiatr Epidemiol. Feb 2004;39(2):126-132.  97 164. Silverstein B. Gender difference in the prevalence of clinical depression: the role played by depression associated with somatic symptoms. Am J Psychiatry. Mar 1999;156(3):480-482. 165. Kubera M, Obuchowicz E, Goehler L, Brzeszcz J, Maes M. In animal models, psychosocial stress-induced (neuro)inflammation, apoptosis and reduced neurogenesis are associated to the onset of depression. Prog Neuropsychopharmacol Biol Psychiatry. Apr 29 2011;35(3):744-759. 166. Maes M. Depression is an inflammatory disease, but cell-mediated immune activation is the key component of depression. Prog Neuropsychopharmacol Biol Psychiatry. Apr 29 2011;35(3):664-675. 167. Maes M, Fisar Z, Medina M, Scapagnini G, Nowak G, Berk M. New drug targets in depression: inflammatory, cell-mediated immune, oxidative and nitrosative stress, mitochondrial, antioxidant, and neuroprogressive pathways. And new drug candidates--Nrf2 activators and GSK-3 inhibitors. Inflammopharmacology. Jun 2012;20(3):127-150. 168. Maes M, Kubera M, Leunis JC. The gut-brain barrier in major depression: intestinal mucosal dysfunction with an increased translocation of LPS from gram negative enterobacteria (leaky gut) plays a role in the inflammatory pathophysiology of depression. Neuro Endocrinol Lett. Feb 2008;29(1):117-124. 169. Maes M, Kubera M, Obuchowiczwa E, Goehler L, Brzeszcz J. Depression's multiple comorbidities explained by (neuro)inflammatory and oxidative & nitrosative stress pathways. Neuro Endocrinol Lett. 2011;32(1):7-24. 170. Leonard BE. The concept of depression as a dysfunction of the immune system. Curr Immunol Rev. Aug 2010;6(3):205-212.  98 171. Maes M, Leonard BE, Myint AM, Kubera M, Verkerk R. The new '5-HT' hypothesis of depression: cell-mediated immune activation induces indoleamine 2,3-dioxygenase, which leads to lower plasma tryptophan and an increased synthesis of detrimental tryptophan catabolites (TRYCATs), both of which contribute to the onset of depression. 172. Miller AH. Depression and immunity: a role for T cells? Brain Behav Immun. Jan 2010;24(1):1-8. 173. Murphy DL, Moya PR, Fox MA, Rubenstein LM, Wendland JR, Timpano KR. Anxiety and affective disorder comorbidity related to serotonin and other neurotransmitter systems: obsessive-compulsive disorder as an example of overlapping clinical and genetic heterogeneity. Philosophical transactions of the Royal Society of London. Series B, Biological sciences. 2013;368(1615):20120435. 174. Marazziti D, Baroni S, Picchetti M, Piccinni A, Silvestri S, Dell'Osso L. [New developments on the serotonin hypothesis of depression: shunt of tryptophan]. Rivista di psichiatria. Jan-Feb 2013;48(1):23-34. 175. Maes M, Mihaylova I, Ruyter MD, Kubera M, Bosmans E. The immune effects of TRYCATs (tryptophan catabolites along the IDO pathway): relevance for depression - and other conditions characterized by tryptophan depletion induced by inflammation. Neuro Endocrinol Lett. Dec 2007;28(6):826-831. 176. Dantzer R, O'Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci. Jan 2008;9(1):46-56. 177. Trivedi MH, Rush AJ, Wisniewski SR, et al. Factors associated with health-related quality of life among outpatients with major depressive disorder: a STAR*D report. J Clin Psychiatry. Feb 2006;67(2):185-195.  99 178. Krishnan KR, Delong M, Kraemer H, et al. Comorbidity of depression with other medical diseases in the elderly. Biol Psychiatry. Sep 15 2002;52(6):559-588. 179. Bachen EA, Chesney MA, Criswell LA. Prevalence of mood and anxiety disorders in women with systemic lupus erythematosus. Arthritis Rheum. Jun 15 2009;61(6):822-829. 180. Hedayati SS, Minhajuddin AT, Afshar M, Toto RD, Trivedi MH, Rush AJ. Association between major depressive episodes in patients with chronic kidney disease and initiation of dialysis, hospitalization, or death. Jama. May 19 2010;303(19):1946-1953. 181. Hootman JM, Cheng WY. Psychological distress and fair/poor health among adults with arthritis: state-specific prevalence and correlates of general health status, United States, 2007. Int J Public Health. Jun 2009;54 Suppl 1:75-83. 182. Seavey WG, Kurata JH, Cohen RD. Risk factors for incident self-reported arthritis in a 20 year followup of the Alameda County Study Cohort. J Rheumatol. Oct 2003;30(10):2103-2111. 183. Lustman PJ, Clouse RE. Depression in diabetic patients: the relationship between mood and glycemic control. J Diabetes Complications. Mar-Apr 2005;19(2):113-122. 184. Lin EH, Katon W, Von Korff M, et al. Relationship of depression and diabetes self-care, medication adherence, and preventive care. Diabetes Care. Sep 2004;27(9):2154-2160. 185. Kurd SK, Troxel AB, Crits-Christoph P, Gelfand JM. The risk of depression, anxiety, and suicidality in patients with psoriasis: a population-based cohort study. Arch Dermatol. Aug 2010;146(8):891-895. 186. Goodwin RD, Devanand DP. Stroke, depression, and functional health outcomes among adults in the community. J Geriatr Psychiatry Neurol. Mar 2008;21(1):41-46. 187. Thombs BD, de Jonge P, Coyne JC, et al. Depression screening and patient outcomes in cardiovascular care: a systematic review. JAMA. Nov 12 2008;300(18):2161-2171.  100 188. Wulsin LR, Singal BM. Do depressive symptoms increase the risk for the onset of coronary disease? A systematic quantitative review. Psychosom Med. Mar-Apr 2003;65(2):201-210. 189. Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. International journal of geriatric psychiatry. Jul 2007;22(7):613-626. 190. Schulz R, Beach SR, Ives DG, Martire LM, Ariyo AA, Kop WJ. Association between depression and mortality in older adults: the Cardiovascular Health Study. Arch Intern Med. Jun 26 2000;160(12):1761-1768. 191. van Melle JP, de Jonge P, Spijkerman TA, et al. Prognostic association of depression following myocardial infarction with mortality and cardiovascular events: a meta-analysis. Psychosom Med. Nov-Dec 2004;66(6):814-822. 192. Do DP, Dowd JB, Ranjit N, House JS, Kaplan GA. Hopelessness, depression, and early markers of endothelial dysfunction in U.S. adults. Psychosom Med. Sep 2010;72(7):613-619. 193. Dowlati Y, Herrmann N, Swardfager W, et al. A meta-analysis of cytokines in major depression. Biol Psychiatry. Mar 1 2010;67(5):446-457. 194. Pizzi C, Manzoli L, Mancini S, Bedetti G, Fontana F, Costa GM. Autonomic nervous system, inflammation and preclinical carotid atherosclerosis in depressed subjects with coronary risk factors. Atherosclerosis. Sep 2010;212(1):292-298. 195. Howren MB, Lamkin DM, Suls J. Associations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med. Feb 2009;71(2):171-186. 196. Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis. Feb 2000;148(2):209-214.  101 197. Pace TW, Miller AH. Cytokines and glucocorticoid receptor signaling. Relevance to major depression. Ann N Y Acad Sci. Oct 2009;1179:86-105. 198. Kop WJ, Stein PK, Tracy RP, Barzilay JI, Schulz R, Gottdiener JS. Autonomic nervous system dysfunction and inflammation contribute to the increased cardiovascular mortality risk associated with depression. Psychosom Med. Sep 2010;72(7):626-635. 199. Vaccarino V, Johnson BD, Sheps DS, et al. Depression, inflammation, and incident cardiovascular disease in women with suspected coronary ischemia: the National Heart, Lung, and Blood Institute-sponsored WISE study. Journal of the American College of Cardiology. Nov 20 2007;50(21):2044-2050. 200. O'Connor JC, Andre C, Wang Y, et al. Interferon-gamma and tumor necrosis factor-alpha mediate the upregulation of indoleamine 2,3-dioxygenase and the induction of depressive-like behavior in mice in response to bacillus Calmette-Guerin. The Journal of neuroscience : the official journal of the Society for Neuroscience. Apr 1 2009;29(13):4200-4209. 201. Bonaccorso S, Marino V, Biondi M, Grimaldi F, Ippoliti F, Maes M. Depression induced by treatment with interferon-alpha in patients affected by hepatitis C virus. J Affect Disord. Dec 2002;72(3):237-241. 202. Prather AA, Rabinovitz M, Pollock BG, Lotrich FE. Cytokine-induced depression during IFN-alpha treatment: the role of IL-6 and sleep quality. Brain Behav Immun. Nov 2009;23(8):1109-1116. 203. Kubera M, Basta-Kaim A, Papp M. The effect of chronic treatment with imipramine on the immunoreactivity of animals subjected to a chronic mild stress model of depression. Immunopharmacology. Sep 1995;30(3):225-230.  102 204. Tseng YL, Chiang ML, Huang TF, Su KP, Lane HY, Lai YC. A selective serotonin reuptake inhibitor, citalopram, inhibits collagen-induced platelet aggregation and activation. Thrombosis research. Dec 2010;126(6):517-523. 205. Maes M, Meltzer HY, Bosmans E, et al. Increased plasma concentrations of interleukin-6, soluble interleukin-6, soluble interleukin-2 and transferrin receptor in major depression. J Affect Disord. Aug 18 1995;34(4):301-309. 206. Karson A, Demirtas T, Bayramgurler D, Balci F, Utkan T. Chronic administration of infliximab (TNF-alpha inhibitor) decreases depression and anxiety-like behaviour in rat model of chronic mild stress. Basic & clinical pharmacology & toxicology. May 2013;112(5):335-340. 207. Menter A, Augustin M, Signorovitch J, et al. The effect of adalimumab on reducing depression symptoms in patients with moderate to severe psoriasis: a randomized clinical trial. J Am Acad Dermatol. May 2010;62(5):812-818. 208. Yip L, Harrison S, Foley P. Infliximab rescue of efalizumab withdrawal flare and psoriasis-precipitated depression. The Australasian journal of dermatology. Nov 2008;49(4):250-251. 209. Bassukas ID, Hyphantis T, Gamvroulia C, Gaitanis G, Mavreas V. Infliximab for patients with plaque psoriasis and severe psychiatric comorbidity. J Eur Acad Dermatol Venereol. Feb 2008;22(2):257-258. 210. Tyring S, Gottlieb A, Papp K, et al. Etanercept and clinical outcomes, fatigue, and depression in psoriasis: double-blind placebo-controlled randomised phase III trial. Lancet. Jan 7 2006;367(9504):29-35.  103 211. Feldman SR, Gottlieb AB, Bala M, et al. Infliximab improves health-related quality of life in the presence of comorbidities among patients with moderate-to-severe psoriasis. Br J Dermatol. Sep 2008;159(3):704-710. 212. Raison CL, Rutherford RE, Woolwine BJ, et al. A randomized controlled trial of the tumor necrosis factor antagonist infliximab for treatment-resistant depression: the role of baseline inflammatory biomarkers. JAMA psychiatry (Chicago, Ill.). Jan 2013;70(1):31-41. 213. Chrousos GP. The hypothalamic-pituitary-adrenal axis and immune-mediated inflammation. N Engl J Med. May 18 1995;332(20):1351-1362. 214. Leonard BE. The immune system, depression and the action of antidepressants. Prog Neuropsychopharmacol Biol Psychiatry. May 2001;25(4):767-780. 215. Pace TW, Hu F, Miller AH. Cytokine-effects on glucocorticoid receptor function: relevance to glucocorticoid resistance and the pathophysiology and treatment of major depression. Brain Behav Immun. Jan 2007;21(1):9-19. 216. Lowy MT, Reder AT, Antel JP, Meltzer HY. Glucocorticoid resistance in depression: the dexamethasone suppression test and lymphocyte sensitivity to dexamethasone. Am J Psychiatry. Nov 1984;141(11):1365-1370. 217. Juruena MF, Pariante CM, Papadopoulos AS, Poon L, Lightman S, Cleare AJ. Prednisolone suppression test in depression: prospective study of the role of HPA axis dysfunction in treatment resistance. Br J Psychiatry. Apr 2009;194(4):342-349. 218. Bauer ME, Papadopoulos A, Poon L, et al. Dexamethasone-induced effects on lymphocyte distribution and expression of adhesion molecules in treatment-resistant depression. Psychiatry research. Dec 15 2002;113(1-2):1-15. 219. Blankenberg S, Barbaux S, Tiret L. Adhesion molecules and atherosclerosis. Atherosclerosis. Oct 2003;170(2):191-203.  104 220. Galkina E, Ley K. Vascular adhesion molecules in atherosclerosis. Arteriosclerosis, thrombosis, and vascular biology. Nov 2007;27(11):2292-2301. 221. Iwen KA, Senyaman O, Schwartz A, et al. Melanocortin crosstalk with adipose functions: ACTH directly induces insulin resistance, promotes a pro-inflammatory adipokine profile and stimulates UCP-1 in adipocytes. The Journal of endocrinology. Mar 2008;196(3):465-472. 222. Muller N, Schwarz MJ. The immune-mediated alteration of serotonin and glutamate: towards an integrated view of depression. Mol Psychiatry. Nov 2007;12(11):988-1000. 223. Maes M. The cytokine hypothesis of depression: inflammation, oxidative & nitrosative stress (IO&NS) and leaky gut as new targets for adjunctive treatments in depression. Neuro Endocrinol Lett. Jun 2008;29(3):287-291. 224. Campbell S, MacQueen G. An update on regional brain volume differences associated with mood disorders. Current opinion in psychiatry. Jan 2006;19(1):25-33. 225. Sapolsky RM. Is impaired neurogenesis relevant to the affective symptoms of depression? Biol Psychiatry. Aug 1 2004;56(3):137-139. 226. Morgan S, Evans RG, Hanley GE, Caetano PA, Black C. Income-Based Drug Coverage in British Columbia: Lessons for BC and the Rest of Canada. Healthcare policy = Politiques de sante. Nov 2006;2(2):115-127. 227. Chamberlayne R, Green B, Barer ML, Hertzman C, Lawrence WJ, Sheps SB. Creating a population-based linked health database: a new resource for health services research. Canadian journal of public health = Revue canadienne de sante publique. Jul-Aug 1998;89(4):270-273. 228. Schnatter AR, Acquavella JF, Thompson FS, Donaleski D, Theriault G. An analysis of death ascertainment and follow-up through statistics Canada's Mortality Data Base  105 system. Canadian journal of public health = Revue canadienne de sante publique. Jan-Feb 1990;81(1):60-65. 229. Solomon DH, Love TJ, Canning C, Schneeweiss S. Risk of diabetes among patients with rheumatoid arthritis, psoriatic arthritis and psoriasis. Ann Rheum Dis. Dec 2010;69(12):2114-2117. 230. Gordis L. Epidemiology. 4th ed. Philadelphia: Saunders; 2009. 231. Dicker RC. Field Epidemiology. In: Gregg MB, ed. Vol 1. 3rd ed: Oxford University Press; 2008. 232. Hanley GE, Morgan S. On the validity of area-based income measures to proxy household income. BMC health services research. 2008;8:79. 233. Hanley GE, Morgan S, Reid RJ. Income-related inequity in initiation of evidence-based therapies among patients with acute myocardial infarction. J Gen Intern Med. Nov 2011;26(11):1329-1335. 234. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. Jun 1992;45(6):613-619. 235. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of chronic diseases. 1987;40(5):373-383. 236. Icen M, Crowson CS, McEvoy MT, Gabriel SE, Maradit Kremers H. Potential misclassification of patients with psoriasis in electronic databases. J Am Acad Dermatol. Dec 2008;59(6):981-985. 237. Asgari MM, Wu JJ, Gelfand JM, et al. Validity of diagnostic codes and prevalence of psoriasis and psoriatic arthritis in a managed care population, 1996-2009. Pharmacoepidemiology and drug safety. May 2 2013.  106 238. De Vera MA, Berard A. Antidepressant use during pregnancy and the risk of pregnancy-induced hypertension. British journal of clinical pharmacology. Aug 2012;74(2):362-369. 239. Han C, Lofland JH, Zhao N, Schenkel B. Increased prevalence of psychiatric disorders and health care-associated costs among patients with moderate-to-severe psoriasis. J Drugs Dermatol. Aug 2011;10(8):843-850. 240. Solberg LI, Engebretson KI, Sperl-Hillen JM, Hroscikoski MC, O'Connor PJ. Are claims data accurate enough to identify patients for performance measures or quality improvement? The case of diabetes, heart disease, and depression. Am J Med Qual. Jul-Aug 2006;21(4):238-245. 241. Martens P, Fransoo R, McKeen N, et al. Patterns of Regional Mental Illness Disorder Diagnoses and Service Use in Manitoba: A Population-Based Study. Winnipeg: Manitoba Centre for Health Policy; November 18, 2004. 242. Austin PC, Daly PA, Tu JV. A multicenter study of the coding accuracy of hospital discharge administrative data for patients admitted to cardiac care units in Ontario. American heart journal. Aug 2002;144(2):290-296. 243. Cox JL, Melady MP, Chen E, Naylor CD. Towards improved coding of acute myocardial infarction in hospital discharge abstracts: a pilot project. Can J Cardiol. Apr 1997;13(4):351-358. 244. Quan H, Li B, Saunders LD, et al. Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health services research. Aug 2008;43(4):1424-1441. 245. Gelber RP, Gaziano JM, Orav EJ, Manson JE, Buring JE, Kurth T. Measures of obesity and cardiovascular risk among men and women. Journal of the American College of Cardiology. Aug 19 2008;52(8):605-615.  107 246. Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. May 12 1998;97(18):1837-1847. 247. Richards HL, Fortune DG, Weidmann A, Sweeney SK, Griffiths CE. Detection of psychological distress in patients with psoriasis: low consensus between dermatologist and patient. Br J Dermatol. Dec 2004;151(6):1227-1233. 248. West SL, Richter A, Melfi CA, McNutt M, Nennstiel ME, Mauskopf JA. Assessing the Saskatchewan database for outcomes research studies of depression and its treatment. J Clin Epidemiol. Aug 2000;53(8):823-831. 249. de Zoysa P. Psychological interventions in dermatology. Indian J Dermatol. Jan 2013;58(1):56-60.              

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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074243/manifest

Comment

Related Items