UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Assessing adolescents with special health care needs : readiness to transition from paediatric to adult… Moynihan, Melissa Mary 2012

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_spring_moynihan_melissa.pdf [ 957.78kB ]
Metadata
JSON: 24-1.0073464.json
JSON-LD: 24-1.0073464-ld.json
RDF/XML (Pretty): 24-1.0073464-rdf.xml
RDF/JSON: 24-1.0073464-rdf.json
Turtle: 24-1.0073464-turtle.txt
N-Triples: 24-1.0073464-rdf-ntriples.txt
Original Record: 24-1.0073464-source.json
Full Text
24-1.0073464-fulltext.txt
Citation
24-1.0073464.ris

Full Text

     ASSESSING ADOLESCENTS WITH SPECIAL HEALTH CARE NEEDS: READINESS TO TRANSITION FROM PAEDIATRIC TO ADULT MEDICAL CARE  by  Melissa Mary Moynihan   B.A., University of Western Ontario, 2004 B.Sc.N., University of Toronto, 2007      THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF     MASTER OF SCIENCE IN NURSING  in  THE FACULTY OF GRADUATE STUDIES     THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)   December 2012    © Melissa Mary Moynihan, 2012    ii ABSTRACT  Successful transition to adult care is critical for adolescents with special health care needs (SHCN), but there are no well-validated measures to assess transition readiness.  This study evaluated the Transition Readiness Assessment Questionnaire (TRAQ) with a younger population and validated the updated Am I ON TRAC for Adult Care questionnaire. During routine clinic appointments 200 youth, ages 12-19 years, from four outpatient clinics at BC Children’s Hospital, Vancouver participated in the study.  The instruments included a demographics information page, the TRAQ, the ON TRAC questionnaire, and two Psychosocial Maturity Inventory (PMI) subscales.  Psychometric properties of the PMI, TRAQ and ON TRAC knowledge items were evaluated using principal components analysis (PCA) with parallel analysis and scree plots.  ON TRAC behaviour items were summed as an index to measure consistent patterns of behaviour; theorized cut-off scores for behaviour items were set at minimally accepted frequency of behaviours required for successful transition to adult care. Relationships between participant age, psychosocial maturity and ON TRAC scores were examined. Youth were from the diabetes (36%), cardiology (27%), gastroenterology (27%), and neurology (10%) clinics, and 57.5% were male.  The PCA of the TRAQ yielded a markedly different factor structure from the original validation study and consequently could not be compared with the ON TRAC for convergent validity.  The PCA of ON TRAC knowledge items identified a 14-item unidimensional scale with a Cronbach’s α=.84.  ON TRAC knowledge and behaviour scores increased with age, with a stronger relationship for knowledge (r=.43, p<.01), while psychosocial maturity correlated with both ON TRAC scores, with a stronger association to behaviour (r=.39, p<.01).  Psychosocial maturity and age had a weak but significant   iii correlation (r=.16, p<.05) suggesting age is a loose proxy for maturity.  Only 27% of 17-year- olds, but 62% 18-year-olds scored above the behaviour cut-off for transition readiness. The ON TRAC questionnaire is a psychometrically sound measure with strong internal consistency, and has potential to be used as a readiness assessment tool in clinical practice. Results of the TRAQ psychometric evaluation raise questions about whether it is a suitable measure to evaluate transition readiness among younger adolescents.                                     iv PREFACE   This thesis was granted ethical approval by the University of British Columbia Behavioural Research Ethics Board, certificate number: H12-00175.                        v TABLE OF CONTENTS ABSTRACT .................................................................................................................................... ii PREFACE ...................................................................................................................................... iv TABLE OF CONTENTS ............................................................................................................... iv LIST OF TABLES ........................................................................................................................ vii LIST OF FIGURES ..................................................................................................................... viii ACKNOWLEDGEMENTS ........................................................................................................... ix DEDICATION ................................................................................................................................ x CHAPTER ONE: SITUATING THE ISSUE ................................................................................. 1 Introduction ................................................................................................................................. 1 Thesis Overview ......................................................................................................................... 6 CHAPTER TWO: REVIEW OF THE LTERATURE ................................................................... 8 Introduction ................................................................................................................................. 8 Transition experiences, expectations, attitudes, needs, and concerns ................................... 14 Transition process and program designs ............................................................................... 21 Transition Readiness ............................................................................................................. 30 Transition outcomes .............................................................................................................. 39 Summary ................................................................................................................................... 41 CHAPTER THREE: RESEARCH DESIGN AND METHODS .................................................. 45 Introduction ............................................................................................................................... 45 Theoretical Frameworks ........................................................................................................... 46 Development Theory ............................................................................................................ 46 Orem’s Self-Care Theory ...................................................................................................... 47 Transtheortical/Stages of Change Model .............................................................................. 49 Transition Theory.................................................................................................................. 50 Study Design ............................................................................................................................. 52 Population and Sample ......................................................................................................... 53 Investigation Techniques ...................................................................................................... 54 Instrumentation ..................................................................................................................... 56 Data Collection ..................................................................................................................... 61 Data Analysis Plan ................................................................................................................ 62 Rigor ..................................................................................................................................... 65 Ethical Considerations .............................................................................................................. 66 CHAPTER FOUR: FINDINGS .................................................................................................... 67 Introduction ............................................................................................................................... 67 Sample Characteristics .............................................................................................................. 67 Psychosocial Maturity Inventory .............................................................................................. 68 Psychosocial Maturity Inventory Scores .............................................................................. 72 Transition Readiness Assessment Questionnaire ..................................................................... 74 ON TRAC Questionnaire .......................................................................................................... 82 ON TRAC Knowledge Scale Structure ................................................................................ 82 ON TRAC Knowledge Scale Scores .................................................................................... 89   vi ON TRAC Behaviour Index ................................................................................................. 94 ON TRAC Behaviour Index Scores ...................................................................................... 96 ON TRAC Model ................................................................................................................ 100 Statistical Analysis .................................................................................................................. 102 Relationships Between the Variables.................................................................................. 103 Linear Regression ............................................................................................................... 106 Summary ................................................................................................................................. 107 CHAPTER FIVE: DISSCUSSION AND IMPLICATIONS ...................................................... 110 Introduction ............................................................................................................................. 110 Key Study Findings................................................................................................................. 110 Discussion ............................................................................................................................... 114 Strengths and Limitations ....................................................................................................... 121 Future Research ...................................................................................................................... 122 REFERENCES ........................................................................................................................... 123 APPENDICIES ........................................................................................................................... 137 Appendix A: Youth Study Information Letter ........................................................................ 138 Appendix B: Recruitment Flyer .............................................................................................. 140 Appendix C: Participant Study Package ................................................................................. 141 Appendix D: Principal Component Analysis of the Psychosocial Maturity Inventory .......... 150 Appendix E: Revised Am I ON TRAC for Adult Care Questionnaire ................................... 152 Appendix F: Scoring Guide for Revised Am I ON TRAC for Adult Care Questionnaire ..... 154    vii LIST OF TABLES Table 1    Descriptive Statistics of the Youth Demographic Characteristics ................................ 68 Table 2    Parallel Analysis for 20-items and n = 183................................................................... 69 Table 3    Parallel Analysis for 18-items and n = 183................................................................... 70 Table 4    Reliability of Psychosocial Maturity Inventory (PMI) ................................................. 71 Table 5    Parallel Analysis for 29-items and n = 161................................................................... 75 Table 6    Factor Loadings, Eigenvalues and Percentage of Variance Explained for the Unconstrained Principal Component Analysis of the TRAQ ...................................... 77 Table 7    Factor Loadings for Principal Component Analysis with Forced Two Components and Oblique Rotation of the TRAQ .................................................................................... 80 Table 8    Parallel Analysis 17-items and n = 179 ........................................................................ 84 Table 9    Component Loadings, Eigenvalues, and Percentage of Variance for Principal Component Analysis with No Rotation of the 17-item ON TRAC Knowledge Scale 84 Table 10  Component Loadings, Eigenvalues, and Percentage of Variance for Prinicpal Component Analysis with Oblique Rotation of the 17-item ON TRAC Knowledge Scale ............................................................................................................................. 86 Table 11  Parallel Analysis for 14-items and n = 180................................................................... 88 Table 12  Component Loadings, Eigenvalues, and Percentage of Variance for Prinicipal Component Analysis with No Rotation of the 14-item ON TRAC Knowledge Scale 88 Table 13  Descriptive Statistics for ON TRAC Knowledge Scale ............................................... 90 Table 14  Descriptive Statistics for ON TRAC Knowledge Scores by Clinic .............................. 91 Table 15  Descriptive Statistics for ON TRAC Behaviour Index ................................................. 97 Table 16  Descriptive Statistics for ON TRAC Behaviour Index Raw Scores by Age & ON TRAC Stage ................................................................................................................. 98 Table 17  Descriptive Statistics for ON TRAC Behaviour Index Scores by Clinic ................... 100 Table 18  Correlations between Age, PMI, ON TRAC Knowledge Score, ON TRAC Behaviour Index Raw Score, ON TRAC Behaviour Index Cut-off Score, ON TRAC Stage and Years with Condition.................................................................................................. 104 Table 19  ANOVA of ON TRAC and PMI Scores Based on Participant ON TRAC Stage ...... 104 Table 20  Linear Regression of the ON TRAC Knowledge Score ............................................. 106 Table 21  Linear Regression of the ON TRAC Behaviour Index Raw Scores ........................... 107 Table 22  Component Loadings and Eigenvalues for PCA of 20-item PMI .............................. 150 Table 23  Component Loadings and Eigenvalues for PCA of 18-item PMI .............................. 151   viii LIST OF FIGURES Figure 1    Conceptual Map of Literature Review Process ........................................................... 13 Figure 2    Mean PMI Scores by Age............................................................................................ 73 Figure 3    Mean PMI Scores by ON TRAC Stage ....................................................................... 73 Figure 4    Mean ON TRAC Knowledge Scores by Age .............................................................. 92 Figure 5    Mean ON TRAC Knowledge Scores by ON TRAC Stage ......................................... 92 Figure 6    Percentage of Youth that Met the Behaviour Index Cut-off Score > 7 ....................... 99   ix  ACKNOWLEDGEMENTS   I would like to express my gratitude to the people who have helped and supported me throughout this process.  I am grateful to my supervisor, Elizabeth Saewyc, and committee members Sandy Whitehouse and Gladys McPherson.  Elizabeth provided continuous support and guidance throughout all stages of this project.  Without her advice, insights and encouragement this project would not have been possible.  Sandy played a key role in developing the direction of this project and connected me with the clinics and staff at BC Children’s Hospital.  I am thankful to Gladys who provided support in the process and for meticulously editing this work.  I am tremendously appreciative of the time each member took to carefully examine the work and provide feedback.  I wish to acknowledge the significant contributions of Mary Paone.  Mary and Sandy developed the ON TRAC model and original questionnaire, which provided the foundation for this work.  I am thankful for Mary’s enthusiasm for this project and her work with the staff and clinics at BC Children’s Hospital, which made this study possible.  I would also like to thank the clinics and youth who participated in the study; they were integral to this project.  A special thank you goes to Marcia Frank whose mentorship and initial advice was vital to this work.  I am grateful to Wendy Hall and Kathy Gregg for their ongoing support and encouragement.  For my family, my parents Ann and Tom and my brother Scott, who have provided me with their endless support and encouragement not only throughout the duration of this work but throughout the course of my whole life.  It is because of your encouragement that this achievement was possible.   x  EDICATI For my family   1 CHAPTER ONE: SITUATING THE ISSUE Introduction Significant medical advances and improvements to health care in recent decades have led to increases in both the quality of life and life expectancies of children with chronic medical conditions (Viner, 2008).  As a result, greater than 90% of children with special health care needs (SHCN) now survive to adulthood and require transitional services (Betz & Telfair, 2007). In the province of British Columbia approximately 9% of children 10 to 14 years old and 11% of adolescents and young adults 15 to 24 years old have some disability due to a physical or mental health condition (Canadian Paediatric Society, 2011).  Furthermore, it is estimated that 50,000 youth with chronic illnesses and disabilities should transfer annually in Canada from paediatric to adult-oriented medical services (Reid et al., 2004). It is alarming to note that existing survivor estimates suggest that the population of adults with SHCN is growing at a rate which is resulting in service system pressures that may lead to social difficulties and quality of life concerns that should be addressed (Betz & Telfair, 2007). The process of transitioning from paediatric medical care to adult medical care has been recognized as a pertinent health care issue since the 1980s (Stabile et al., 2005).  The complexity of this process is illustrated by the Society for Adolescent Medicine’s frequently cited definition of transition as: “the purposeful, planned process that addresses the medical, psychosocial and educational/vocational needs of adolescents and young adults with chronic physical and medical conditions as they move from child-centred to adult oriented healthcare systems”  (Blum et al., 1993, p.570). The need for well structured and effective transition programs that prepare adolescents for both adulthood and adult medical services has been identified and explored in a variety of medical disciplines including: cystic fibrosis (Flume, 2009), congenital heart defects (Reid et al., 2004), mental health (Singh, 2009), juvenile idiopathic arthritis (McDonagh, Southwood, &   2 Shaw, 2006), asthma (Scal, Davern, Ireland, & Park, 2008), organ transplant (McDonagh & Kelly, 2010), Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome (HIV/AIDS) (Machado, Succi, & Turato, 2010) and diabetes (Nakhla, Daneman, Frank & Guttmann, 2008).  According to the Canadian Paediatric Society (2011) the purpose of transition is “…to provide health care that is uninterrupted, coordinated, developmentally appropriate and psychologically sound before and throughout the transfer of youth into the adult system” (p.3). Ultimately, the objective of successful health care transition is to optimize health and to facilitate individuals with SHCN to achieve their maximum potential (Blum et al., 1993).  Several expert panels, conferences and policy statements have focused on enhancing transition through recommendations, guidelines and models developed to promote a more fluid transition process (Rosen, Blum, Britto, Sawyer, & Seigel, 2003).  However, despite being a prevalent topic of concern in the literature, transition has been studied rarely (Nakhla et al., 2008; Rosen, Blum, Britto, Sawyer, & Seigel, 2003; Singh, 2009) and current evidence suggests that youth with chronic medical conditions continue to confront many difficulties at the time of transition (Tuchman, Slap, & Britto, 2008). Generally, any period of transition is identified as a time when persons are vulnerable and at risk of harmful health outcomes (Meleis & Trangenstein, 1994).  In addition, adolescence is recognized as a period of significant changes and increased risk-taking behaviours (Flume, 2009; Singh, 2009). Health care transition from paediatric to adult services occurs simultaneously with the transition to adulthood during which time youth are becoming more independent and responsible in all aspects of their lives, including managing their chronic conditions (Machado et al., 2010; Pai & Ostendorf, 2011).  Throughout this transition period youth are at risk of being lost between paediatric and adult services and at increased risk of withdrawing from health care   3 services (Singh, 2009). A Canadian study on the prevalence of successful transition from paediatric to adult cardiac health care found that only 47% of patients successfully transferred and 27% of the young adults reported not having had a cardiac appointment since leaving paediatric services (Reid et al., 2004).  Similar findings from a study of transition care of young adults with type 1 diabetes (T1D) indicate a health care dropout rate of 48% between paediatric and adult services, which the researchers suggest could be a conservative estimate of the problem (Scott, Vallis, Charett, Murray, & Latta, 2005). High rates of unsuccessful transition are a serious concern for all youth with SCHN as inadequate transition has been associated with greater risk of treatment non-compliance, lack of medical follow-up, and increased morbidity and mortality (Viner, 2008).  Adolescents with SHCN are faced with a variety of challenges upon transition to adult care including: acquiring self-care knowledge and skills, adjusting to the culture of adult services, ending long-term relationships with trusted healthcare providers, and independently managing their conditions (Stabile et al., 2005; Tuchman et al., 2008).  Consequently, the need for transitional care that is developmentally appropriate and focused on both the knowledge and skill development required for negotiating adult care services has been emphasized (Paone, Wigle, & Saewyc, 2006).  Health care professionals (HCP) in both disease-specific and general paediatric health services have concentrated on identifying transition barriers and developing effective transitional models and programs in order to make transition an efficient process striving to optimize the health and well-being of young people with chronic medical conditions (Rosen et al., 2003).  Despite these research efforts health care transition continues to be a challenging issue and there remains many unanswered questions regarding this process.   4 There is ample information regarding the challenges and barriers that impede successful transition but there continues to be a deficit of empirically based research and little conclusive evidence to support a definitive transition model (Rosen et al., 2003).  Best practice guidelines and expert opinions frequently recommend that transfer of care between paediatric and adult services should ideally occur when adolescents are physically, medically, and socially prepared for transition (Lotstein et al., 2005).  The Society for Adolescent Medicine has outlined the following fundamental principles of transition: services that are appropriate for chronological age and developmental attainment, addresses common concerns of youth, increases the patient’s autonomy, personal responsibility and promotes self-reliance, uses flexible and individualized plans, and an identified HCP who works with the adolescent and their family to oversee the transition process (Rosen et al., 2003).  Even with consistent agreement on the essential components of health care transition, there remains a dearth of information regarding how to determine when adolescents are ready and adequately prepared to successfully transition to adult medical care. In a review of health care transition studies Betz (2004) found chronological age to be the most frequently reported primary criteria for transitioning youth with SHCN to adult care services.  As the primary determinate for transition, age is used as an estimate of the adolescents’ developmental level, which is then employed as a guide for coordinating transitional care (Pai & Ostendorf, 2011). The most commonly reported age range for health care transition is between 16 and 22 (Betz, 2004).  At BC’s Children’s Hospital, the policy is that youth are seen in outpatient clinics up to their 19th birthday, after which they are to be seen by adult health care providers (S. Whitehouse, personal communication, May 10, 2011).  Although it is the most frequently used criteria for transition, the literature regards age alone as insufficient criteria for   5 the timing of transition (Fleming, Carter, & Gillibrand, 2002).  HCP need a valid and reliable means of assessing adolescent maturity and readiness for health care transition planning, preparation and implementation of transfer to adult care (Jalkut & Allen, 2009).  As a result researchers have explored the concept of transition readiness, and individual transition programs have developed their own assessment tools and checklists that include readiness measures (Betz, 2004; Ledlie, 2007; Rapley & Davidson, 2010).  However, there is a of lack a well-established and validated transition readiness measures that can be used as diagnostic instruments to assesses and determine preparedness of adolescents with SHCN to successfully transfer to adult care (Betz, 2004; Rapely & Davidson, 2010; Sawicki et al., 2011).  Consequently, the need for a valid and reliable measure of transition readiness has been recognized as an essential element of transition preparation and an area for further research investigation  (Betz, 2004; Betz & Telfair, 2007; Pai & Schwartz, 2011). Health care transition (HCT) models are largely based on adolescent development theory. To-date there have not been any investigations that confirm HCT recommendations and guidelines are in fact meeting the developmental needs of adolescents with SHCN.  The two research questions for this study were: 1) Do the normative stages of adolescent development identified in the ON TRAC (Taking Responsibility for Adolescent/Adult Care) transition model (stage and achievements at each age range) fit as the same stage-for-age among adolescents with special health care needs (SHCN)?  2) Are the behaviours and skills believed to be essential for successful transition to adult care developmentally appropriate expectations for adolescents with SHCN?  In order to answer these questions two transition readiness assessment measures, the Transition Readiness Assessment Questionnaire (TRAQ) and ON TRAC’s Am I ON TRAC for Adult Care questionnaire, were administered to youth ages 12-19 with a variety of chronic   6 conditions attending ambulatory specialist clinics at BC Children’s Hospital.  The psychometric properties and usefulness of these two tools were evaluated.  In addition, the Psychosocial Maturity Instrument (PMI) was used to examine the relationship between participants’ age, psychosocial and the two transition readiness assessment measures.  The results from this study have implications for improving the transitional care services as HCP will have access to a valid and reliable transition assessment tool which will facilitate the use of specific strategies and interventions that target adolescents knowledge and/or skill deficits to better prepare patients for HCT.   The primary research objectives were: a) Evaluate if the Transition Readiness Assessment Questionnaire (TRAQ) is a reasonable measure of adolescents with SHCN readiness to transfer from paediatric to adult health care services for a younger Canadian population b) Validate the ON TRAC’s (Taking Responsibility for Adolescent/Adult Care) Am I ON TRAC for Adult Care c) Assess if adolescent development theory is an appropriate framework for preparing youth with SHCN for transition to adult medical care Thesis Overview This chapter has identified the need for the study by presenting the research questions and objectives and has highlighted the dearth of empirically based research and lack of conclusive evidence to support whether adolescent development theory is an appropriate framework for HCT.  In chapter two, I provide a detailed review of the health care transition literature.  The review groups and discusses studies based on their research objectives: 1) exploring transition experiences, expectations, attitudes, needs and concerns 2) investigating the transition process and program designs 3) examining transition readiness 4) assessing transition outcomes.  Chapter   7 three provides the theoretical frameworks for the study and describes the methods used to collect and analyze the data.  The findings are presented in chapter four.  First, participant demographic information is provided and factor analyses of both the PMI and TRAQ are conducted.  Next, the psychometric properties of the ON TRAC knowledge scale are examined and the behaviour index raw scores and cut-off scores are explained.  Finally, the relationships between age, psychosocial maturity, and ON TRAC scores are explored through correlation, t-tests and linear regression.  In chapter five the key results of the study are summarized and discussed in relation to the current HCT literature.  Finally, recommendations for transition programs and future research are made.    8 CHAPTER TWO: REVIEW OF THE LTERATURE Introduction The health care transition (HCT) literature recognizes the need to improve transitional services from paediatric to adult medical services for adolescents with special health care needs (SHCN) generally and in several health care sub-specialties.  The literature has largely concentrated on developing transitional models, guidelines, recommendations, and practice standards that seek to facilitate an efficient transition process with the intent to optimize the health of young people with chronic medical conditions (Rosen et al., 2003). Relative to the amount of information available there is a dearth of research studies that examine HCT for youth with SHCN (Reiss, Gibson, & Walker, 2005).  Developing compelling HCT evidence through longitudinal studies or randomized controlled trials is especially problematic for two reasons. First, the population of interest - adolescents and young adults - is very transient which makes it extremely difficult to conduct longitudinal research.  Second, serious ethical dilemmas arise when contemplating any type of experimental design that would compare and evaluate different transition programs.  Consequently, a substantial proportion of published articles regarding HCT are editorials, program descriptions, expert opinions and recommendations. Furthermore, it is clear that transition research remains in the initial stages of development as the current body of HCT literature is founded on exploratory and descriptive research (Betz, 2004). Studies that assess and evaluate specific transition interventions and strategies are needed to provide more conclusive evidence to better meet the transitional care needs of youth with SHCN. A systematic review of the HCT literature was conducted.  Five electronic databases were accessed during the search process:  CINAHL, PubMed, Web of Science, Cochrane Database of Systematic Reviews, and PsycINFO.  A list of keywords pertinent to the research topic were generated and used to start, as well as refine the search process.  Keywords identified included:   9 adolescents, adolescence, teenagers, teens, youth, young adults, paediatric, children’s health services, adult health care, transition, transitional services, transition programs, chronic illness, chronic disease, and chronic disease management.  The search was further limited to articles written in English, the language of the researcher.  Titles and article abstracts were then evaluated to identify publications that met the following criteria: primary research studies or secondary analysis of HCT related data, and focused on transition from paediatric to adult health care services for adolescents with SHCN.  Articles were excluded if they were not data-based.  In addition, references of key studies were reviewed to identify any significant publications that were not found during the systematic search and as a means to ensure that noteworthy articles had not been over looked.  A total of 43 articles met the above inclusion criteria and an additional eight reviews of the HCT literature were located.  The literature reviews (Betz, 2004; Fleming et al., 2002; Jalkut & Jackson Allen, 2009; McDonagh & Kelly, 2010; Machado et al., 2010; Nakhla et al., 2008; Pai & Ostendorf, 2011; Rapley & Davidson, 2010;) provide an overview of the current body of knowledge and highlight clinical implications of key findings, identify pertinent research needs, and discuss the overall limitations of the HCT literature. The matrix method approach was used as described by Gerrard (2003) to review the literature.  A review matrix was created that systematically organized the information gathered and assisted with summarizing, analyzing and efficiently using the evidence.  Separate review matrixes were created for each type of research design: quantitative, qualitative and mixed methods.  Each matrix consisted of the following nine column topics: author(s), title, journal, year, and location, purpose and research question(s), study design, methodology, and methods, sample characteristics, data collection methods, findings, study limitations, study strengths, and additional comments. After all relevant articles were classified by design, the   10 matrixes were analyzed to identify each study’s purpose and research question(s) and then articles were grouped together based on common research objectives. The topic of HCT has been examined through a variety of research designs, methodologies, and data collection methods. Twenty quantitative studies (Annunziato et al., 2011; Benchimol et al., 2011; Betz, Redcay, & Tan, 2003; Clarizia et al., 2009; Hazel, Zhang, Duffy, & Campillo, 2010; Johnston, Bell, Tennet, & Carson, 2006; Kipps et al., 2002; Lotstein, McPherson, Strickland, & Newacheck, 2005; Lostein et al., 2009; McManus, Fox, O’Connor, Chapman, & MacKinnon, 2008; McPherson, Thaniel, & Minniti, 2009; Sawicki et al., 2011; Scal, Evans, Blozis, Okinow, & Blum, 1999; Scal & Ireland, 2005; Schwartz, Tuchman, Hobbie, & Ginsberg, 2011; van Staa, van der Stege, Jedeloo, Moll, & Hilberink, 2011b; Van Walleghem, MacDonald, & Dean, 2006; Whittaker, 2004; Williams et al., 2011; Wong et al., 2010;), 13 qualitative investigation (Fair, Sullivan, & Gatto, 2010; Hess & Straub, 2011; Hauser & Dorn, 1999; McCurdy et al., 2006; Moons et al., 2009; Patterson & Lanier, 1999; Reiss et al., 2005; Soanes & Timmons, 2004; Stabile et al., 2005; Tuchman et al., 2008; Valenzuela et al., 2011; van Staa, Jedeloo, van Meeteren, & Latour, 2011a; Visentin, Koch, & Kralik, 2006) and 10 mixed methods designs (Cappelli, MacDonald, & McGrath, 1989; de Beaufort, Jarosz-Chobot, Frank, de Bart, & Deja, 2010; Pacaud, McConnell, Huot, Aebi, & Yale, 1996; Pacaud, Yale, Stephure, Trussell, & Davies, 2005; Por et al., 2004; Reid et al., 2004; Scott, Vallis, Charette, Murray, & Latta, 2005; Telfair, Alexander, Loosier, Alleman-Velez, & Simmons, 2004; Van Walleghem, MacDonald, & Dean, 2008; Wiener, Kohrt, Battles, & Pao, 2011) were reviewed. HCT research has varied depending on how investigators have chosen to manage two significant design elements: the population of interest and timing of the study.  Some investigations concentrated on the transition process for youth with specific chronic conditions such as   11 congenital heart defects (CHD), transplant recipients, HIV and T1D, while others broadly address common transition issues that apply to all youth with SHCN by having a sample with a variety of conditions.  The research explores the concerns of various stakeholders involved in transition: adolescents, HCP who work with youth with SHCN, parents, and other family members.  A few studies have examined and compared multiple perspectives but most research has focused on a single stakeholder’s point of view.  With the exceptions of a longitudinal study (Tuchman et al., 2008), a participatory action research (PAR) design conducted over a 3 year period (Scott et al., 2005), evaluation of a systems navigator service that collected data every 6 months for 1 year (Van Walleghem et al., 2008), and three secondary analyses of HCT related data (Lostein et al., 2005; Lostein et al., 2009; Scal & Ireland, 2005) all the identified investigations used either cross-sectional or retrospective data collection methods.  Study designs differed in regards to the timing of data collection being either before or after transition had occurred.  A few studies explored HCT using more than one cohort of youth and/or young adults to assess the concerns, attitudes and experiences both before and after transition had occurred (Annunziato et al., 2011; Patterson & Lanier, 1999; Van Walleghem et al., 2006; Van Walleghem et al., 2008;).  Despite these fundamental design differences all 43 studies reviewed were descriptive and exploratory in nature. This review will discuss the HCT literature by examining the studies according to their research objectives.  The purposes can be grouped into four main categories with four investigations being multi-purposed (Lostein et al., 2009; McCurdy et al., 2006; Reiss et al., 2005; Visentin et al., 2006).  Figure 1 provides a conceptual map of the literature review process. The first grouping of studies explore or examine the expectations, attitudes, needs, concerns, and experiences of youth/young adults transferring (or who have transferred) from paediatric to adult   12 care and/or the needs, experiences and concerns of other major stakeholders including the adolescents’ family members, and the various HCP working with this population (Clarizia et al., 2009; Hauser & Dorn, 1999; McCurdy et al., 2006; McManus et al., 2008; Moons et al., 2009; Patterson & Lanier, 1999; Por et al., 2004; Reiss et al., 2005; Scott et al., 2005; Soanes & Timmons, 2004; Stabile et al., 2005; Telfair et al., 2004; Tuchman et al., 2008; Valenzuela et al., 2011; van Staa et al., 2011a; Visentin et al., 2006; Whittaker, 2004; Wiener et al., 2011; Wong et al., 2010).  A second cluster of investigations focus on the transition process and program designs (de Beaufort et al., 2010; Fair et al., 2010; Johnston et al., 2006; Kipps et al., 2002; Lotstein et al., 2005; Lostein et al., 2009; McCurdy et al., 2006; Pacaud et al., 1996; Pacaud et al., 2005; Reiss et al., 2005; Scal et al., 1999; Scal & Ireland, 2005; Van Walleghem et al., 2006; Van Walleghem et al., 2008; Visentin et al., 2006).  These studies explore the major challenges, concerns, and issues of transition, examine the existing transition structures and services, and identify best practices for HCT.  The literature also surveys adolescents with SHCN readiness to transition from paediatric to adult care and describes a new theoretical model of readiness for transition (Annuziato et al., 2011; Benchimol et al., 2011; Betz et al., 2003; Cappelli et al., 1989; Hess & Straub, 2011; McPherson et al., 2009; Sawicki et al., 2011; Schwartz et al., 2011; van Staa et al., 2011b; Williams et al., 2011).  This set of articles describe the self-care needs and level of health care self-care sufficiency required for transitioning, assess youths’ readiness and knowledge about the transition process, explore factors associated with readiness, determine if age, gender and disease   13  Figure 1    Conceptual Map of Literature Review Process                             * Indicates the group includes studies that were multi-purposed Mixed Methods Designs  10 Literature Reviews  8 Research Studies:  35 Primary Studies 3 Secondary Analyses Systematic Review of the HCT Literature Four main themes from analyzing the objectives and/or purposes of the research studies Transition Experiences, Expectations, Attitudes, Needs, & Concerns *  19  Transition Process & Program Designs *  15  Transition Readiness   10  Transition Outcomes *   3 Quantitative Designs  20 Qualitative Designs  13   14 severity are predictors of readiness, and include the development of readiness assessment instruments.  Lastly, three studies concentrate on transition outcomes (Hazel et al., 2010; Lostein, et al., 2009; Reid et al., 2004).  These investigations compare characteristics of adolescents who successfully and unsuccessfully transitioned from paediatric to adult care, determine the percentage of young adults who successfully transferred, examine the correlates of successful transition, and report United States (U.S.) national results on a transition core outcome based on the Maternal and Child Health Bureau’s (MCHB) core performance outcomes used to define a high-performing system of care of children with SHCN (Lostein et al., 2009). Transition experiences, expectations, attitudes, needs, and concerns  Nineteen studies explored key stakeholders experiences, expectations, attitudes, needs and concerns regarding HCT.  Eleven used qualitative research approaches (Hauser & Dorn, 1999; McCurdy et al., 2006; Moons et al., 2009; Patterson & Lanier, 1999; Reiss et al., 2005; Soanes & Timmons, 2004; Stabile et al., 2005; Tuchman et al., 2008; Vanlenzuela et al., 2011; van Staa et al., 2011a; Visentin et al., 2006), four employed mixed methods designs (Por et al., 2004; Scott et al., 2005; Telfair et al., 2004; Wierner et al., 2011) and four were quantitative investigations (Clarizia et al., 2009; McManus et al., 2008; Whittaker, 2004; Wong et al., 2010). The population of interest varied across the studies with eight having a broad focus on youth with SHCN (McManus et al., 2008; Patterson & Lanier, 1999; Por et al., 2004; Reiss et al., 2005; Soanes & Timmons, 2004; Tuchman et al., 2008; van Staa et al., 2011a; Wong et al., 2010) and the remaining concentrating on specific chronic conditions (Clarizia et al., 2009; Hauser & Dorn, 1999; McCurdy et al., 2006; Moons et al., 2009; Scott et al., 2005; Stabile et al., 2005; Telfair et al., 2004; Valenzuela et al., 2011; Visentin et al., 2006; Whittaker, 2004; Wiener et al., 2011). Three investigations sought only the opinions and perspectives of HCP regarding HCT   15 (McManus et al., 2008; Por et al., 2004; Telfair et al., 2004) and six studies examined transitional concerns, needs and experiences from multiple perspectives including youth, young adults, family members and HCP (Clarizia et al., 2009; Hauser & Dorn, 1999; Reiss et al., 2005; van Staa et al., 2011a; Visentin et al., 2006; Wong et al., 2010).  The remaining research samples were comprised only of adolescent and young adult participants with five exploring the experiences of those who had transferred to adult care (McCurdy et al., 2006; Stabile et al., 2005; Valenzuela et al., 2011; Whittaker, 2004; Wiener et al., 2011), three involving adolescents currently in the transition process (Scott et al., 2005; Soanes & Timmons, 2004; Tuchman et al., 2008), and two using separate cohorts to study the perspectives of youth prior to transition and young adults who had successfully transitioned (Moons et al., 2009; Patterson & Lanier, 1999). Tuchman et al. (2008) conducted the only longitudinal study that was reviewed.  Their investigation was part of a larger research project that collected information regarding adolescents’ health care preferences over an 18-month period (Tuchman et al., 2008).  This study examined how participants’ opinions regarding transition changed over time and provides unique insights into the experience of transition as 6 of the 22 participants transferred to adult care during the study interval. The qualitative investigations collected data through focus groups (Hauser & Dorn, 1999; McCurdy et al., 2006; Patterson & Lanier, 1999; Reiss et al., 2005) and a range of interview methods (McCurdy et al., 2006; Moons et al., 2009; Reiss et al., 2005; Soanes & Timmons, 2004; Stabile et al., 2005; Tuchman et al., 2008; Valenzuela et al., 2011; van Staa et al., 2011a; Visentin, et al., 2006).  The findings of these studies highlight prevalent transition barriers, identify common factors that impact transition, and suggest strategies for improving the transition process.  Differences between paediatric and adult medical care services were   16 recognized as a significant impediment to successful transition (McCurdy et al., 2006; Patterson & Lanier, 1999; Reiss et al., 2005; Soanes & Timmons, 2004; Valenzuela et al., 2011; van Staa et al., 2011a; Visentin et al., 2006).  Participants discussed anxieties related to leaving the comfort and familiarity of the paediatric setting and having to adjust to the new and unknown environment of adult care (Hauser & Dorn, 1999; Moons et al., 2009; Soanes & Timmons, 2004; Stabile et al., 2005).  Concerns accentuated in this literature included being forced to end close relationships with HCP (Reiss et al., 2005; Tuchman et al., 2008; Valenzuela et al., 2011), difficulties in establishing trusting relationships with adult HCP who young adults believe are inexperienced and unqualified to manage their conditions (Hauser & Dorn, 1999; Reiss et al., 2005; Tuchman et al., 2008), and struggling with the changes in relationships with both parents and HCP that occur as adolescents assume more responsibility in all aspects of their lives including health management (Hauser & Dorn, 1999; McCurdy et al., 2006; Moons et al., 2009; Stabile et al., 2005; Tuchman et al., 2008; van Staa et al., 2011a). A second theme that commonly emerged from this data was factors that impact transition. Findings emphasize a need for good communication between paediatric and adult HCP (Patterson & Lanier, 1999; Reiss et al., 2005; van Staa et al., 2011a), providing support for adolescents and family members throughout the transition process (Hauser & Dorn, 1999; Soanes & Timmons, 2004), the need to introduce transition early and view it holistically (Visentin et al., 2006) and adequately preparing adolescents for the adult health care environment and assuming more responsibility for their own health care (Hauser & Dorn, 1999; McCurdy et al., 2006; Moons et al., 2009; Patterson & Lanier, 1999; Soanes & Timmons, 2004; Tuchman et al., 2008; Valenzuela et al., 2011; van Staa et al., 2011a).  These pertinent factors were addressed when recommendations and strategies for improving the transition process were   17 discussed.  Research participants suggested that adolescents need to become experts regarding their own health condition (Patterson & Lanier, 1999), youth require more education and information regarding their condition, medications, high-risk behaviours, self-care skills, and vocational planning (Hauser & Dorn, 1999; McCrudy et al., 2006; Moons et al., 2009; Soanes & Timmons, 2004; van Staa et al., 2011a), formal transition programs are necessary (Moons et al., 2009; Visentin et al., 2006), and an orientation to adult care that enables adolescents to meet the adult HCP prior to transitioning and offers mentoring from young adults who have successfully transitioned would be beneficial (Hauser & Dorn, 1999; Moons et al., 2009; Patterson & Lanier, 1999; Visentin et al., 2006).  Additionally, Reiss et al. (2005) acknowledged that developmental delay and the progressive nature of the health condition significantly impact transition.  The transition process may be especially complex for persons with developmental delay as they may be unable to meet the independent care expectations of adult care (Reiss et al., 2005).  Similarly, the progressive nature of a condition may inhibit all HCT preparation, as some HCP believe that transition to adult care is unwarranted (Reiss et al., 2005).  Finally, financing and health insurance were raised as critical issues during the transition period.  Health care benefits that dependents and minors qualify for under their parents’ insurance are often terminated when adolescents turn 18 and consequently they no longer have access to key health care resources that were available to them as children (Reiss et al., 2005; Valenzuela et al., 2011).  Not all findings were negative and one study identified ‘leaving paediatric care is a logical step’ as a core theme related to moving on to adult medical care (van Staa et al., 2011a).  Young adults and parent participants expressed that transitioning to adult care was inevitable and appropriate for adults; comparably, HCP acknowledged transition to be a natural process that is age-appropriate but had different opinions regarding the optimal age for transfer (van Staa et al., 2011a).   18  The mixed methods designs employed questionnaires and telephone interviews to collect data from HCP involved with transition (Por et al., 2004; Telfair et al., 2004), adolescents in the transition process (Scott et al., 2005) and young adults who had transitioned to adult care (Wiener et al., 2011).  These investigations had similar research results to the qualitative studies. The HCP identify that youth and their families require support throughout all stages of transition (beginning, middle and end), detailed information about adult services, and to establish trusting relationships with adult HCP (Por et al., 2004).  Additionally, Por et al. (2004) examined the needs of HCP preparing adolescents with SHCN for transition and found them to include: on going communication and documentation between providers, good handovers and formal transitional care pathways, and education and professional development opportunities.  Of particular interest to this review was HCP were specifically questioned about criteria that should be used to assess readiness to transition to adult care. Mental maturity was the most frequently identified criterion followed by age (Por et al., 2004).  Unfortunately, there was no follow up regarding how mental maturity should be measured but the researchers concluded that this finding supports individually assessing adolescents’ self-care skills and condition-specific knowledge (Por et al., 2004). The HCP participants from this study suggest that transitional needs for adolescents, family and health providers could be best met by maintaining clearly written transition plans, improving communication between staff and patients, having adolescents meet with adult providers prior to transfer, and creating a specialized adolescent unit (Por et al., 2004). Telfair et al. (2004) surveyed both paediatric and adult HCP from multiple disciplines regarding transition to adult care for adolescents with sickle cell disease (SCD).  Participants were asked about what activities they participate in to signify transition and the results found   19 67% of respondents reported demonstrating transition practices including: seeing youth without their parents, encouraging more health care responsibility, and providing patients with literature regarding transition (Telfair et al., 2004).  Furthermore, 89% of these HCP acknowledged a need for a systematic transition process with paediatric and adolescent providers being significantly more likely than adult HCP to state a formal transition program would better meet their patients’ needs (Telfair et al., 2004).  A Canadian study used participatory action research (PAR) to gain a better understanding of why young adults drop out of diabetes care during the transition period and obtained suggestions from young adults on how to improve transitional care (Scott et al., 2005).  Scott and colleagues (2005) reported a 48% dropout rate from care at the time of transition for adolescents with T1D.  The PAR participants discussed feeling a sense of abandonment when relationships with paediatric providers ended, recommended longer initial appointments with adult care providers to assist with establishing a trusting relationship, visits to the adult clinic before transferring, and wanted more opportunities to interact with peers and form peer support groups (Scott et al., 2005).  Autonomy and self-direction issues emerged as critical elements for adolescents and young adults in regards to managing their diabetes (Scott et al., 2005) and reiterated the concept of youth becoming experts in their own condition, a prominent theme in some of the previously reviewed studies.  Wiener et al (2011) interviewed 59 young adults with HIV regarding their experiences with transitioning of which 71% reported they had transitioned from a specialized paediatric clinic to adult care. The results highlight that there continues to be both systematic and relational barriers that impede HCT as 45% of the 42 participants who had transitioned reported the transition from paediatric to adult care was more difficult than they had anticipated (Wiener et al., 2011).  Again respondent recommendations emphasized enhancing communication between all parties involved in the transition process,   20 increasing both the continuity and quality of care, providing assistance with logistical elements of transition, assessing the developmental level of adolescents, and individualizing transition plans based on patients’ unique capabilities (Wiener et al., 2011).  Finally, four studies used quantitative approaches (Clarizia et al., 2009; McManus et al., 2008; Whittaker, 2004; Wong et al., 2010).  Data collection consisted of questionnaires, semi- structured interviews and health record audits.  Whittaker (2004) retrospectively examined the transition experience of a cohort of young adults who had transitioned from paediatric to adult diabetes care.  Similar to the findings of Wiener et al. (2011), Whittaker (2004) found that 47% of the study participants stated they had difficulties with transitioning.  When questioned about a preferred scenario for young adult diabetes care 73.3% of respondents endorsed having a specialized health care team at a young adult clinic comprised of both paediatric and adult HCP (Whittaker, 2004).  A Canadian study explored patients, parents, and HCP perspectives of HCT preparation for adolescents with CHD found an evident gap in both patients’ and parents’ knowledge regarding the implications and changes associated with adolescents future transition to adult care (Clarizia et al., 2009).  Furthermore, 69% of the HCP respondents felt that youth are not being sufficiently prepared for transition and believe that not having a formal transition program is a significant barrier to successful transition (Clarizia et al., 2009).  The results of a national survey of American paediatricians by McManus et al. (2008) found that the majority of paediatric practices do not routinely offer HCT services that have been established as vital to ensuring successful transition to adult medical care in fundamental policy and position statements.  Approximately 62% of the respondents expressed that transition planning for youth with SHCN should begin at 18 to 20 of age (McManus et al., 2008), which contradicts the ON TRAC model, the basis for many transition programs around the world, which recommends the   21 formal transition process should begin at 10 years of age (Canadian Paediatric Society, 2011). This particular study reveals cardinal gaps in the implementation of essential procedures that are known to facilitate efficient and effective HCT (McManus et al., 2008).  A Hong Kong study examined adolescents’ and parents’ attitudes and perceptions towards transition care prior to transferring (Wong et al., 2010).  The investigators found the overall attitude towards transition care to be positive with 85.3% of youth and 82.5% of parents willing to transfer to adult care when the youth reached adult age (Wong et al., 2010).  Similar to other investigations, this study drew attention to the lack of preparation for transition reporting that less than 10% of patients had received transition information from their doctors (Wong et al., 2010).  Wong and colleagues (2010) identified three principal factors affecting willingness to transition: adolescents’ perceptions of their own responsibility towards the SHCN had a significant positive association, detailed explanations of transition by doctors encouraged adolescents to accept transition, and simply being opposed to change was the only significant barrier.  Participants regarded 18 as the most appropriate age for transfer to adult care with 47% of adolescents and 41.9% of parents reporting this to be the optimal age for transition care (Wong et al., 2010).  In this sample there was no significant difference in the transition decision based on age or gender and no evidence was found to support that age, gender, chronic disease type, or condition severity were predictive of willingness to transition (Wong et al., 2010). Transition process and program designs  A group of investigations examined, compared, and assessed existing transition programs and services in order to describe the proportion of youth with SHCN who receive transitional care, determine the adequacy of HCT services, and identify best practices for HCT.  The investigators used a range of research designs and methods as seven studies employed   22 quantitative designs (Johnston et al., 2006; Kipps et al., 2002; Lotstein et al., 2005; Lostein et al., 2009; Scal et al., 1999; Scal & Ireland, 2005; Van Walleghem et al., 2006;), four were qualitative investigations (Fair et al., 2010; McCurdy et al., 2006; Reiss et al., 2005; Visentin et al., 2006), and four used mixed methods approaches (de Beaufort et al., 2010; Pacaud et al., 1996; Pacaud et al., 2005; Van Walleghem et al., 2008). In 1999 Scal et al. conducted a quantitative study to identify and characterize the available transition health services for adolescents and young adults in the United States. Scal et al. (1999) identified 227 transition programs and clinics of which 126 completed a questionnaire about the services they provided. The researchers classified the transitional programs into two categories: condition focused and adolescent focused and observed that transition services mostly concentrate on disease outcomes as opposed to functional outcomes (Scal et al., 1999). The most salient transition impediments were related to health care system problems with 41% of the participating programs indicating that funding was a major barrier (Scal et al., 1999). Unexpectedly, this investigation found that patient and family resistance to transition were rarely reported as a barrier to effective transitional planning which contradicted one of the study’s a priori hypotheses (Scal et al., 1999) and the later findings of Wong et al.’s (2010) Hong Kong study.  Overall, the results indicate that programs are not providing collaborative, coordinated, and integrative care (Scal et al., 1999).  Three secondary analyses used previously collected data from the 2000-2001 National Survey of Children with Special Health Care Needs (NS CSHCN) in the U.S. (Lostein et al., 2005; Scal & Ireland, 2005;) and the results of the second NS CHSCN conducted in 2005-2006 (Lostein et al., 2009) to determine the adequacy of HCT services for youth with SHCN, describe the proportion of youth with SHCN who receive transition services, and discern factors   23 associated with addressing transition from paediatric to adult oriented health care.  As part of the NS CSHCN data collection parents/legal guardians of adolescents between 14-17 years of age with SHCN completed a telephone interview regarding their use of HCT services.  Both Scal and Ireland (2005) and Lostein et al. (2005) found that approximately 50% of the respondents reported discussing how their child’s health care needs might change in adulthood with their HCP and 30% had developed a plan to meet these changing needs with their HCP.  There was a slight discrepancy between the two studies regarding the percentage of participants who had also discussed transferring care to an adult provider with findings differing from 29.9% (Scal & Ireland, 2005) to 21% (Lostein et al., 2005).  Scal and Ireland (2005) used a HCT scale to assess the extent to which HCT had been addressed with families of youth with SHCN.  According to their findings older age, white race, a greater number of health care needs, and higher quality of parent-provider relationship are associated with significantly higher scores on the HCT scale (Scal & Ireland, 2005).  Comparably, using the same survey data Lostein et al. (2005) found that having received guidance and support in transition to adulthood was significantly higher for older adolescents and those who met the MCHB criteria for a medical home.  To meet the MCHB definition of a medical home the adolescent must satisfy the following five criterion: has a usual source of care, has a personal physician or nurse, has no problem obtaining referrals when needed, has received effective care coordination when needed, and has received family- centred care (Lostein et al., 2005).  Finally, both studies reported that HCT had been addressed in a comprehensive manner with only 16.35% (Scal & Ireland, 2005) and 15.3% (Lostein et al., 2005) of the respondents.  Lostein et al. (2009) examined the results of the second NS CSHCN in relation to meeting a core transition outcome comprised of four component measures. The core outcome was met when respondents reported having had discussions with their child’s HCP   24 regarding the following three key topics: transferring to adult providers, adult health care needs, and health insurance (Lostein et al., 2009).  The final component was fulfilled if parents reported that the HCP usually or always encourages the child to take responsibility for their care (Lostein et al., 2009).  Unfortunately, due to changes in measurement strategies the researchers were not able to directly compare the results of the two NS CSHCN surveys.  Nonetheless, the findings of this later study did not indicate that HCT services improved over the 5 year interval between the data collection periods.  The most recent investigation suggests that only 41% of respondents meet the transition core outcome (Lostein et al., 2009).  Significant disparities in meeting the core outcome exist based on participants’ race/ethnicity and social economic status, those most impacted by their health condition are less likely to have transition discussions with HCP, and overall the rates of meeting the transition core outcome do not significantly differ between adolescents with SHCN and a referent sample without SHCN (Lostein et al, 2009.).  A study conducted in the United Kingdom compared four different methods of transfer from paediatric to adult medical care for youth with T1D (Kipps et al., 2002).  The researchers found that attendance rates at the first clinic appointment after transfer ranged from 64% when transfer was directly from paediatric to adult services to 86% when patients first transferred to a young adult clinic in the same centre run jointly by paediatric and adult HCP before transferring to the adult clinic (Kipps et al., 2002).  More young adults reported being dissatisfied with their transition when they were directly transferred to adult care (Kipps et al., 2002).  Researchers discovered transfer methods that progressed to a young adult clinic in the same hospital and then to adult care had higher regular attendance rates 2 years post-transition compared to both direct transfer to adult care and transfer to a young adult clinic in a different hospital (Kipps et al., 2002).  The results demonstrate that the mode of transfer seems to affect subsequent clinic   25 attendance rates following HCT (Kipps et al., 2002).  Transitioning to a young adult clinic may be preferable and has potential to improve subsequent attendance outcomes compared to directly transferring to an adult clinic (Kipps et al., 2002).  However, there was a gradual but noticeable decline in the number of individuals who regularly attended clinic appointments for the entire study cohort as attendance dropped from 98% at 2 years pre-transition to 61% 2 years post- transition regardless of the transition method (Kipps et al., 2002).  Johnston et al. (2006) performed an audit of clinic attendance records 2 years before and 15-18 months following HCT for young adults with diabetes after the establishment of a Young Persons’ clinic.  Upon transfer to adult diabetes services patients chose whether they would attend the traditional weekday adult clinic or the Young Persons’ clinic that was held on Saturdays.  Of the 33 charts reviewed 18 individuals transitioned to the adult clinic and 15 chose to attend the Young Persons’ clinic (Johnston et al., 2006). Good clinic attendance was delineated as one or no missed appointments over the study period (Johnston et al., 2006). The researchers found that 18% of the 33 patients never attended any clinic appointments after leaving paediatric care (Johnston et al., 2006).  When attendance was compared between the two types of clinics the results provide some support for the use of young adult clinics because the Young Persons’ clinic was more than twice as likely to be associated with good attendance (Johnston et al., 2006).  In spite of these findings, the investigators were disappointed that the Young Persons’ clinic did not produce more positive results (Johnston et al., 2006).  The feasibility and acceptability of an administratively based transitional support and systems navigation service (the Maestro Project) for young adults between 18-30 years of age with T1D was examined (Van Walleghem et al., 2006).  The Maestro Project is a Canadian navigation service that was developed to increase the rate of medical follow-up and education   26 and decrease the morbidity and mortality from complications of diabetes for young adults between 18-30 years of age (Van Walleghem et al., 2006).  The study separated the 323 participants into two groups.  The first group had transitioned to adult care prior to the implementation of the Maestro Project and the second group consisted of adolescents who were directly referred to the Maestro Project during their transition to adult care.  In total there were 323 study participants of which 127 used the transitional support systems navigation service to request 230 community contacts for assistance to access care, education or optometry services and an additional 111 participants who contacted the Maestro Project 203 times for other related information (Van Walleghem et al., 2006).  Unfortunately, the researchers did not differentiate the results for the two groups and information regarding the usage of the intervention by participants who had access to the Maestro Project during their transition and those who were introduced to the navigation system after transition and used it to reconnect with diabetes care was not provided.  These initial findings demonstrate that an administrative systems navigation service is feasible and provides evidence that such a service has the potential to improve young adults access to care, health related resources and information during the vulnerable transitional period (Van Walleghem et al., 2006). Van Walleghem et al. (2008) conducted a second study, to further evaluate the Maestro Project, which concentrated on clinical outcomes and identifying barriers after transition.  Again participants were divided into two cohorts: younger participants who had the assistance of the Maestro Project as they transitioned and older participants who were transitioned without the initial support of the Maestro Project but were enrolled with the service 1 to 7 years after transition.  Van Walleghem et al. (2008) found that the systems navigator model assisted the older cohort to reconnect with adult medical services and reduced the first year fall-out from care   27 rate for the younger cohort.  Consistent with other research findings, difficulties or frustrations with establishing regular follow-up with the adult health care team after transitioning from paediatric care were reported by 42% of the older and 33% of the younger study participants (Van Walleghem et al., 2008).  This evaluation of the Maestro Project found improved medical surveillance during the first 2 years of its implementation, however, there was no evidence of improved short-term medical outcomes (Van Walleghem et al., 2008).  Pacaud et al. (1996) wanted to establish the importance of the HCT issue and did so with a mixed methods design that surveyed young adults who had attended paediatric diabetes clinics at two hospitals and were transferred to adult-oriented care within a specified 5 year period.  The two hospitals were independent from each other, had different approaches to the transition process and also transferred patients to different adult clinics.  The survey results found 50.4% of respondents reported a problem with transition that was classified as self-perceived, a delay in receiving care, or a loss in follow-up (Pacaud et al., 1996).  Two variables were associated with the presence of a problem: the difference between actual age at transition and the ideal transition age suggested by the participant, and the number of HCP seen by an individual before establishing regular follow-up care (Pacaud et al., 1996).  Despite only a small difference between mean age at referral to adult care (18.5) and the ideal age for transition suggested by respondents (18.8) 65% of the respondents felt they should have been transitioned to adult care at a later age (Pacaud et al., 1996).  A noteworthy study strength is the findings of the two participant subgroups (from the two paediatric diabetes clinics) were compared which enabled the researchers to provide evidence to suggest that the difficulties of HCT are a widespread problem and not just a process specific issue (Pacaud et al., 1996).  Furthermore, additional support for the findings of this first investigation was produced when Pacaud et al. (2005)   28 performed a replication study.  The second investigation was conducted in another Canadian province that employed a different care delivery model.  The same research methods and tools were used with only two minor alterations (Pacaud et al., 2005).  There was no significant difference in the main findings between the original and follow-up study with 52% of the participants in the second investigation expressing they had experienced a problem with transition, a delay of more than twelve months between care, or had no current follow-up (Pacaud et al., 2005). The respondents’ suggested mean ideal age for transition was 17.6 years which is slightly younger than in the original study, but the actual mean age of referral was also younger at 17.2 years and ranged from 14 to 19 years (Pacaud et al., 2005).  The percentage of individuals who would have preferred to transfer later decreased to 38% (Pacaud et al., 2005) from 65% (Pacaud et al., 1996). This second study increases the validity of the findings (Pacaud et al., 2005) and accentuates the significance of the HCT problem.  Furthermore, the results of these studies led the researchers to question if the problems of HCT may be related to psychosocial and maturational issues that were not considered during the transition process (Pacaud et al., 2005). One investigation assessed the transitional practices of paediatric diabetes HCP from 36 different countries (de Beaufort et al., 2010).  All members of the International Society for Pediatric and Adolescent diabetes were e-mailed a questionnaire that addressed the transition process and collected basic program information.  The key findings provide the following information: half of the centres surveyed had a structured transition program, 44% of the centres suggest transition should occur between the ages of 14-25, nine centres see more than 40 patients over the age of 25, evaluation of successful transfer to adult care is completed at only 35% of the   29 centres, and 90% of respondents recommend that transition planning begin at least 1 year before transfer to adult care occurs (de Beaufort et al., 2010).  Four qualitative studies explored the transition process (Fair et al., 2010; McCurdy et al., 2006; Reiss et al., 2005; Visentin et al., 2006), three of which were multi-purposed and reviewed in detail in the previous section on transition experiences (McCurdy et al., 2006; Reiss et al., 2005; Visentin et al., 2006).  Fair et al. (2010) interviewed medical professionals and social workers that provide care to youth with HIV in North Carolina to determine best HCT practices. Three levels of best practices were identified: patient, provider, and system (Fair et al., 2010) which are similar to the broad content domains - stages of transition, health care systems, and transition narratives - that emerged in the study by Reiss and colleagues (2005).  Many of the main research findings from the investigation by Fair et al. (2010) have been examined in other investigations.  One such prominent theme is the promotion of self-care skills.  Participants stressed that adolescents must possess the following five essential skills in order to function successfully in an adult clinic: take ownership of their medical care, be able to schedule their own medical appointments, communicate effectively with HCP, fill prescriptions, and practice medical adherence (Fair et al., 2010).  Both Reiss et al. (2005) and Fair et al. (2010) identified financing as an important issue of HCT.  Assisting youth to navigate the complicated health insurance and social services systems was identified as a best practice at the systems level of both American studies (Fair et al., 2010; Reiss et al., 2005).  Also, findings from Fair et al. (2010) provide further endorsement for creating a young adult clinic as the HCP felt that the unique psychosocial and medical needs of youth and young adults (ages 16-24) justify a specialized clinic.   30 An especially relevant finding by Fair et al. (2010) identified assessing patient readiness for HCT as an essential best practice of HCT.  HCP explicitly acknowledged the importance of evaluating adolescents' readiness for HCT by assessing their level of knowledge, skills, responsibility, and independence and further recognized a need for specific interventions to enhance youths’ preparedness if the readiness assessment reveals deficiencies (Fair et al., 2010). All four qualitative studies in this category emphasize the need to adequately prepare adolescents for the significant differences between paediatric and adult medical care services (Fair et al., 2010; McCurdy et al., 2006; Reiss et al., 2005; Visentin et al., 2006).  Each investigation also discussed key strategies for enhancing the transition preparation process and promoting independence in paediatric settings so that adolescents will have the necessary knowledge and skills to thrive in adult-oriented care. Transition Readiness The literature also explored the concept of ‘readiness to transition’ with eight articles (Annuziato et al., 2011; Benchimol et al., 2011; Betz et al., 2003; Cappelli et al., 1989; Hess & Straub, 2011; McPherson et al., 2009; Sawicki et al., 2011; Williams et al., 2011) concentrating on the self-care needs, health care knowledge, skills, and behaviours necessary for successfully transitioning to adult care.  Four of these studies developed and/or validated readiness assessment instruments (Betz et al., 2003; Cappelli et al., 1989 et al.; Sawicki et al., 2011; Williams et al., 2011) and one described the development and elements of a pilot school-based health care transition education program implemented in a large urban county in the U.S. (Hess & Straub, 2011).  A single investigation examined the relationship between youths’ perception of their readiness to transition with other factors (van Staa et al., 2011b).  Four investigations had condition-specific populations of interest (Annunziato et al., 2011; Benchimol et al., 2011;   31 Capelli et al., 1989; McPherson et al., 2009) and the remaining involved adolescents with a range of SHCN (Betz et al., 2003; Hess & Straub, 2011; Sawicki et al., 2011; van Staa et al., 2011b; Williams et al., 2011). An additional article described a newly developed social-ecological model of adolescents and young adults (AYA) readiness for transition that concentrated on a broader conceptualization of transition and included a pilot study which identified correlates of HCP decision making associated with transition (Schwartz et al., 2011). Two quantitative studies evaluated participants’ health care related knowledge (Benchimol et al., 2011; McPherson et al., 2009).  McPherson et al. (2009) assessed adolescents with SCD readiness and knowledge regarding the transition process and statistically analyzed if age, gender and disease severity can predict a patient’s level of transition readiness. As part of the development of a novel tool for HCT Benchimol et al. (2011) evaluated and compared the knowledge of adolescents with inflammatory bowel disease (IBD) and their parents in four major domains: disease characteristics, treatments, medical history, and health services and resources. In order to assess patient readiness and knowledge about transition McPherson et al. (2009) identified five components of transition readiness: amount of prior thought about transition, knowledge about the steps in the transition process, interest in learning more about the process, anticipated difficulty of transition, and perceived importance of continued care from a adult HCP who specialized in SCD.  The research findings suggest the greatest barrier to transition is inadequate knowledge and anticipatory guidance about the transition process, which was evidenced by knowledge of the basic steps in transition being ranked the lowest of the five readiness assessment components (McPherson et al., 2009).  McPherson et al. (2009) determined that age is a significant predictor of readiness, as participants 17-20 years of age demonstrated significantly greater knowledge and interest in transition than 14-16 years old.  Additionally,   32 severity of disease was found to negatively impact both participants interest in learning about transition and their anticipated difficulty with transition, and females reported higher levels of anticipated difficulty (McPherson et al., 2009). Benchimol et al. (2011) found that adolescents and their parents were equally likely to correctly answer questions relating to disease characteristics and treatment.   However, parents were significantly more likely to have greater knowledge about health services and resources (Benchimol et al., 2011).  Both of these studies accentuate that the amount of HCT preparation youth receive concerning the transition process, access to health resources, and the demands of adult care are insufficient.  Mastery of health care management behaviours of adolescents (14-17 year olds) were compared to those of young adults (18 years and older) who were expected to transition to adult care by Annunziato et al. (2011).  The self-report survey revealed that the majority of participants stated they “always” engage in behaviours related to medical management (Annuziato et al., 2011).  Yet less than half the respondents reported consistently demonstrating health care management behaviours related to: knowing chronic illness history, discussing diagnosis and health care management with parents and HCP, and keeping updated health records (Annuziato et al., 2011). Young adult respondents had significantly higher mean total scores for performing health care management behaviours compared to the adolescents (Annuziato et al., 2011) suggesting that partaking in health care management behaviours is associated and improves with age.  Despite these findings, less than half of young adult participants reported consistently managing their health care independently, making their own appointments, and understanding insurance issues (Annunziato et al., 2011).  This again highlights a lack of preparation for HCT, as young adults on the verge of transferring are not yet comprehensively managing their own care.  Furthermore, the researchers identified a need for   33 validated measures for assessing the health management behaviours of adolescents prior to transition to adult care (Annuziato et al., 2011).  Four investigations described the level of health care self-sufficiency required for successfully transitioning to adult care and developed and/or validated transition readiness assessment instruments (Betz et al., 2003; Capelli et al., 1989; Sawicki et al., 2011; Williams et al., 2011).  Capelli et al. (1989) identified that no clear guidelines or methods for assessing whether youth with cystic fibrosis (CF) were adequately prepared for transition existed and stressed that age alone is an insufficient criterion for transfer. More than 20 years later Sawicki et al. (2011) acknowledged that many transition programs concentrate on transition-related skill development yet programs do not assess adolescents’ transition readiness.  Williams and colleagues (2011) recognized HCP currently use a variety of transition checklists to evaluate adolescents’ level of medical self-management however, these tools are largely unvalidated.  As a result, Capelli et al. (1989), Sawicki et al. (2011), and Williams et al. (2011) independently developed and validated transition readiness assessment instruments. Capelli et al. (1989) interviewed paediatric and adult care givers as well as three adult CF patients to determine what behaviours and knowledge were perceived to be essential for successful transition to adult services.  Based on the interview responses a Readiness Questionnaire for CF patients was designed.  This 24-item questionnaire, consisting of 15 knowledge-items and nine behaviour items, was validated with a sample of 36 adolescents and young adults with CF. Comparably, Sawicki et al. (2011) developed and validated the Transition Readiness Assessment Questionnaire (TRAQ) a 29-item instrument which evaluates the self-care management and self-advocacy skills of adolescents with various SHCN preparing for HCT.  In their validation of the Readiness Questionnaire Capelli et al. (1989) found the behaviour subscale,   34 knowledge subscale, and full-scale were all better discriminators of readiness than age.  Sawicki and colleagues (2011) used the stages of change model as the framework for developing the TRAQ and therefore proposed this instrument can clearly identify the adolescent’s developmental stage of readiness for transition.  Validation of the TRAQ found older age to be associated with higher self-management scores but had no association with the self-advocacy domain (Sawicki et al., 2011).  Youth with mental health or cognitive problems had lower scores in all aspects of the TRAQ compared to individuals who had physical health problems (Sawicki et al., 2011) and provides additional evidence that developmental delay significantly impacts transition.  Both Capelli et al. (1989) and Sawicki et al. (2011) propose that the Readiness Questionnaire and TRAQ have important practice implications as they can be used to determine if patients have knowledge, skills and/or behaviour deficiencies that must be addressed prior to transition.  These instruments can assist HCP to develop targeted interventions to ensure adolescents are adequately prepared to meet the expectations of the adult health care system and to maximize lifelong functioning (Capelli et al., 1989; Sawicki et al., 2011). The Self-Management Skills Assessment Guide was developed for the Alberta Children’s Hospital and is a youth and parent-rated questionnaire that assesses medical self-management and readiness to transition in adolescents (Williams et al., 2011).  The Self-Management Skills Assess Guide is composed of items from transition readiness tools developed at other Canadian hospitals (Williams et al., 2011).  In the preliminary validation of this instrument it was found that adolescents and parents both report patients are obtaining the skills that HCP and experts consider necessary for medical self-management (Williams et al., 2011).  The medical self- management scores were strongly associated with scores of functional independence, but had small and non-significant correlations with age, which demonstrates the potential value of   35 readiness measures, opposed to relying on age, for predicting successful transition to adult care (Williams et al., 2011). Betz et al. (2003) conducted a validation study of the California Healthy & Ready to Work Transition Health Care Assessment (CA HRTW THCA) tool.  The CA HRTW THCA is comprised of 72 questions and 14 domains that assess the extent to which youth manage their health care self-care needs and if they possess the necessary knowledge and skills to function independently (Betz et al., 2003).  This validation study found adolescents most often report possessing the knowledge and skills required to: monitor their health condition, track health records, obtain information and reproductive counseling, and engage in responsible sexual behaviours (Betz et al., 2003).  The domains with the highest percentage of no responses, signify areas where respondents lack knowledge and/or skills to function independently included: knowledge of legal rights and protections, engaging in preventative health behaviours, accessing community resources, knowledge of health condition management, and using transportation safely.  The general findings demonstrate that adolescents with SHCN lack essential health care self-care knowledge and skills imperative for self-sufficiency and living independently (Betz et al., 2003).  Betz et al. (2003) state that more empirical data is required to gain a better understanding of adolescents’ self-care sufficiency educational needs during the HCT preparation period.  The researchers suggest that the CA HRTW THCA requires further refinement and interventions and programs that promote the acquisition of essential knowledge and skills for HCT need to be developed (Betz et al., 2003). A Dutch study examined the relationship between youth’s perception of their readiness to transition and five salient HCT factors: socio-demographic characteristics, variables associated with disease and health care, effect of the condition, self-management ability, and attitude   36 towards transition (van Staa et al., 2011b). To measure ability for self-management the researchers created a self-efficacy instrument that used items from available self-efficacy measures and from the California Healthy and Ready to Work Transition Assessment Tool (van Staa et al., 2011b).  Fifty-six percent of participants reported being “probably” or “definitely” ready to transition to adult medical services (van Staa et al., 2011b).  Readiness to transition was positively correlated with age, had weak associations with disease-related factors, and strong relationships with the self-management measures (van Staa et al., 2011b).  A majority of respondents viewed transition as a normal part of growing up but many still reported being concerned about the process (van Staa et al., 2011b) which aligns with qualitative findings from the literature about adolescents transition experiences, attitudes and concerns (Hauser & Dorn, 1999; Moons et al., 2009; Soanes & Timmons, 2004; Stabile et al., 2005; van Staa et al., 2011a). Transition discussions with HCP did not occur often with only 14.3% of respondents reporting these discussions happened “regularly” or “very often” and was related to older age (van Staa et al., 2011b).  Inadequate preparation and lack of information about transition has been frequently recognized as a major barrier (Clarizia et al., 2009; Lostein et al., 2005; Scal & Ireland, 2005; Wong et al., 2010) and Lostein et al. (2005) also found older age to be associated with more HCT discussions and preparation.  Lastly, the majority of the variance in transition readiness was explained by socio-demographic characteristics, self-management measures and the attitude towards transition (van Staa et al., 2011b).  The researchers reported age was a key contributor to adolescents’ perception of readiness; higher transition readiness was also associated with more self-efficacy skills related to independent hospital behaviour, increased independence during consultations, a more positive attitude towards transition, and more reported transition discussions (van Staa et al., 2011b).   37 A school-based health care transition education intervention designed to equip adolescents and young adults with crucial health literacy, self-advocacy, and self-determination skills needed to become successful adult health care consumers was piloted in a large urban county in the U.S. (Hess & Straub, 2011).  The 40 hour curriculum was implemented in 13 high school special education classes and involved 137 students in grades 9 through 12 who all had SHCN with some degree of cognitive limitation (Hess & Straub, 2011).  Following completion of the intervention both educators and students participated in focus groups and questionnaires that assessed the perceived relevance, interest, ease of implementation, and completeness of the program (Hess & Straub, 2011).  Preliminary results found educators and students both felt the curriculum was highly relevant and valuable, adolescents and young adults reported the intervention provided information they did not receive elsewhere, and students were least interested in the insurance unit and most interested in the unit on sexual health (Hess & Straub, 2011).  Transition policies and guidelines such as the American Academy of Pediatrics, American Academy of Family Physicians, and American College of Physicians-American Society of Internal Medicine’s consensus statement (2002) and the Canadian Paediatric Society’s reaffirmed position statement (2011) on transition to adult care for adolescents and young adults with SHCN emphasize that programs should provide multi-disciplinary and collaborative care. HCP identified a lack of reimbursement and time for transition services among the factors that impede their ability to sufficiently prepare youth with SHCN for transition (McManus et al., 2008).  Interventions that extend the health education components of transitional care to classrooms would create additional opportunities for addressing the complex health care management skills that youth with SHCN require to successfully navigate the health care system and are necessary for achieving healthy, independent living.   38 Schwartz and colleagues (2011) developed a new theoretical model of adolescents and young adults (AYA) readiness for transition, which aims to address the complex, dynamic and social-ecological elements of transition.  The SMART model “incorporates socio- demographics/culture and healthcare access, patient characteristics and the inter-related components of knowledge, skills/self-efficacy, beliefs/expectations, goals, relationships and psychosocial functioning of patients, parents and providers” (Schwartz et al., 2011, p.884), and is much more comprehensive than current transition frameworks that predominantly rely on patient age, knowledge and skills.  SMART uses the social-ecological framework developed by Bronfenbrenner to examine the interactions between immediate (microsystem, mesosystem) and broader (exosystem, macrosystem) systems and considers how these systems influence AYA readiness for transition (Schwartz et al., 2011).  A pilot study was conducted with HCP to determine if the components of SMART were pertinent to transition care.  HCP completed a questionnaire, constructed to identify correlates of provider decision making associated with transition, on patients 18 years and older at a paediatric cancer survivorship clinic (Schwartz et al., 2011).  Support for the model was found as several SMART components: patient disease knowledge and skills, patient autonomy and achievement of developmental milestones, present psychosocial crisis, parent involvement, HCP/patient communication regarding transition, and HCP ability to identify an adult provider, were significantly associated with the HCP’s plans for transitional care (Schwartz et al., 2011).  Furthermore, the results demonstrate that transitioning from paediatric to adult medical care is a complex process, which is associated with many factors and not just age, condition knowledge and skills (Schwartz et al., 2011).     39 Transition outcomes The HCT literature has established that facilitating efficient and effective transition between health care services promotes healthy management of SHCN and is critical to decreasing long-term complications, morbidity and mortality due to chronic conditions (Viner, 2008; Visentin et al., 2006).  However, there is a significant paucity of research that has assessed or evaluated HCT outcomes.  Only three such investigations were located during this systematic review. Reid et al. (2004) obtained outcome data for an entire birth cohort, from the largest paediatric cardiac center in Canada, with complex CHD who should be seen annually at a specialized adult CHD clinic (Reid et al., 2004).  Lostein et al. (2009) did a secondary analysis of the NS CSHN and reported national results for the U.S. regarding a transition core outcome. The core outcome was met when parents of youth with SHCN reported: having had at least one of three transition specified discussions, the transition related discussions would not have been helpful, and the HCP usually or always encouraged the adolescent to take responsibility for their care (Lostein et al., 2009).  Lastly, a Canadian study described the proportion of juvenile idiopathic arthritis (JIA) patients who had unsuccessfully transferred from paediatric to adult rheumatology care and compared the characteristics of individuals who had successfully transitioned to those who had not (Hazel et al., 2010). Results of the national U.S. data found that only 41% of adolescents with SHCN met the transition core outcome (Lostein et al., 2009).  The proportion of respondents who met each core component measure varied greatly and ranged from 34% to 78% with discussions of health insurance occurring least frequently and accounts of children usually or always being encouraged to take responsibility were reported most often (Lostein et al., 2009).  Reid et al. (2004) used a mixed methods design to determine the percentage of patients who successfully transitioned to   40 adult care.  Successful transfer to adult care was defined as having attended at least one appointment of any type at a Canadian Adult Congenital Heart (CACH) centre (Reid et al., 2004). Only 47% of the cohort successfully transferred and 27% of the young adults reported not having a cardiac appointment since leaving paediatric care (Reid et al., 2004). In their investigation Hazel et al. (2010) labeled unsuccessful transition as a patient never making contact with the appointed adult rheumatologist or if being lost to follow-up at 2 years post transfer.  They found 52% of patients met the criteria for unsuccessful transition with 17 making no initial contact with the identified adult rheumatologist and 35 patients being lost to follow-up 2 years after transfer (Hazel et al., 2010). These findings are similar to the results of both Lostein et al. (2009) and Reid et al. (2004).  These statistics are not encouraging and clearly demonstrate that HCT programs and interventions are not meeting the transition needs of adolescents with SHCN.  In addition to reporting statistical data on transition outcomes Reid et al. (2004) used qualitative techniques to examine the correlates of successful transfer and Hazel et al. (2010) quantitatively compared the characteristics of patients who had successfully transitioned to those who had not.  Hazel et al. (2010) found a significant association between unsuccessful transition and identifying a low active joint count (AJC), a measure of disease severity, at the last paediatric visit.  The researchers also noted that males tended toward a higher risk for unsuccessful transfer (Hazel et al., 2010).  Male gender was previously recognized as factor associated with inadequate HCT services (Scal & Ireland, 2005) and not meeting the transition core outcome (Lostein et al., 2009).  Results from a Dutch study contradict these findings as it found boys exhibited more transition readiness than girls (van Staa, van de Stege, Jedeloo, Moll, & Hilberink, 2011) and Reid et al. (2004) report no correlation exists between successful transition and gender, education attainment, average income, primary diagnosis, and residence in   41 parents’ home.  However, successful transfers were found to be associated with: more paediatric cardiovascular surgeries, older age at last visit to paediatric care, documented recommendations in the medical chart for follow-up at a CACH centre, living closest to a CACH centre in Ontario and elsewhere in Canada, and patient beliefs that adult CHD care should be at a CACH centre (Reid et al., 2004).  These three studies provide much needed outcome data and empirical evidence that demonstrate transition to adult health care continues to be problematic for many adolescents with SHCN. Summary  This body of evidence reveals a critical need to enhance transitional care services for adolescents with SHCN and establishes common factors that facilitate and impede the health transition process from paediatric to adult medical services.  It clearly identifies that all stakeholders in the transition process (adolescents, young adults, HCP, and family members) continue to have major concerns and unmet needs during this vulnerable period.  Significant factors that influence and impact transition from paediatric to adult medical care for adolescents with SHCN have been explored and described extensively.  Key findings of this review include: transition to adult services continues to be a problematic experience for many adolescents and young adults, transition should be viewed as a process that needs to begin early (preferably before the teenage years) and adequately prepares youth with SHCN for the responsibility shift and independent management skills required by adult medical services, practice differences between paediatric and adult services are major transition barriers, adolescent involvement in health care decision-making and independence in clinic attendance is believed to be essential, adolescents/young adults, family members, and HCP all need to be supported throughout the process, specialized adolescent/young adult clinics may potentially improve both transition   42 experiences and outcomes, youths’ health care knowledge, skills, and behaviours should be assessed prior to transition, it is necessary for paediatric providers to recommend and connect youth with adult providers, and health care system issues especially health insurance, financing, and funding are significant impediments to transition. Despite having ample information regarding transition experiences and process challenges there continues to be many unanswered questions concerning how to best prepare adolescents with SCHN to comprehensively manage their health care independently, maximize lifelong functioning, and successfully navigate the adult health care system. The research interest in transition services began over 30 years ago but has not evolved beyond exploratory and descriptive investigations.  The most prevalent obstacle related to advancing the current body of HCT knowledge is that the majority of the transition literature is in the form of program descriptions, discussion pieces, and suggestions for optimal transition practices (Reiss et al., 2005).  Additionally, study limitations and design flaws plague HCT research.  The investigations appraised in this literature review commonly cite the following limitations: small samples, low participation rates, potential for response bias due to the use of self-report instruments, and reporting findings that cannot be generalized to the larger population of interest. Design flaws diminished the validity of quantitative results and jeopardized the trustworthiness of the qualitative findings.  Absence of conceptual frameworks, methodological or theoretical approaches, poor descriptions of research methods, no evidence of reflexive practices, using unreliable and non-validated measurement tools, relying on convenience sampling, inadequate descriptions of coding processes and/or theme development, and conducting mostly retrospective and cross-sectional investigations frequently threatened the quality of the reviewed studies.   43 Many of the above identified limitations and design flaws have also been acknowledged in other reviews of the transition literature (Betz, 2004; McDonagh & Kelly, 2010; Nakhla et al., 2008). This review found significant gaps in the HCT literature that hinder the ability for HCPs to improve current transition services.  There is a well-recognized deficit of outcome based research and little conclusive evidence to support a definitive transition model (Rosen et al., 2003).  A need for empirical data with respect to transition models, preparation, timing of transfer (Soanes & Timmons, 2004) and to enhance understanding of transition difficulties (Betz et al., 2003; Reiss et al., 2005) has also been identified.  Studies that empirically evaluate the transition process and demonstrate the success and failure rates of different transitional programs are an essential next step. This research would provide vital evidence to support the use of one model over another and/or determine components of transition programs that are most beneficial for facilitating successful transition. Longitudinal investigations that follow a group of adolescents through the transition process would greatly add to the current body of transition literature for youth with SHCN, as it could identify important information about why and/or how these individuals drop out of care (Betz, 2004). There is also a scarcity of empirically supported interventions that facilitate self-management behaviours (Annunziatio et al., 2011) and a lack of valid and reliable tools that measure adolescents’ readiness for transition (Annunziato et al., 2011; Betz, 2004; Capelli et al., 1989; Sawicki et al., 2011; Williams et al., 2011). Developmentally appropriate care and services are accepted as a fundamental principle of transition planning (Blum et al., 1993; Canadian Paediatric Society, 2011; Rosen et al., 2003). Transition programs worldwide use the ON TRAC model (Canadian Paediatric Society, 2011), which is largely based on adolescent development theory.  However, no research could be found that evaluated or assessed if these programs are in fact meeting the developmental needs of   44 adolescents with SHCN.  Betz (2004) found that chronological age is the most frequently reported primary criteria for transition despite being recognized as an insufficient marker of transition readiness 15 years earlier (Capelli et al., 1989).  HCP suggest that maturity level should be the main criteria for assessing youths’ preparedness for adult care (Por et al., 2004) and that unaddressed psychosocial and maturation issues could be potential components of the transition problem (Pacaud et al., 2005; Williams et al., 2011).  In addition, researchers and HCT experts declare that the ideal time to transfer to adult medical care is when adolescents with SHCN are physically, medically, and socially prepared for transition (Lostein et al., 2005), but there is a lack of validated and standardized measures with which to evaluate youths’ level of preparedness.  A number of the studies reviewed acknowledged the need to establish transition readiness indicators and identified readiness assessment as a critical element of transitional care. Yet only three assessment instruments specifically developed and validated for assessing transition readiness in youth with SHCN were located.  These instruments assess skills related to health care knowledge, advocacy, and behaviours that are considered necessary for adult care but have only been validated by the developers.  Further validation of these and similar assessment tools are critically needed.  In addition, examining if transition programs such as ON TRAC are developmentally appropriate and if the expectations of the adult health care system are reasonable for adolescents with SHCN will add to the understanding of the HCT process.   45 CHAPTER THREE: RESEARCH DESIGN AND METHODS Introduction  The research questions for this study were: 1) Do the normative stages of adolescent development identified in the ON TRAC (Taking Responsibility for Adolescent/Adult Care) transition model (stage and achievements at each age range) fit as the same stage-for-age among adolescents with special health care needs (SHCN)? and 2) Are the behaviours and skills believed to be essential for successful transition to adult care developmentally appropriate expectations for adolescents with SHCN?  Furthermore the objectives were to: a) Evaluate if the Transition Readiness Assessment Questionnaire (TRAQ) is a reasonable measure of adolescents with SHCN readiness to transfer from paediatric to adult health care services for a younger Canadian population b) Validate the ON TRAC’s (Taking Responsibility for Adolescent/Adult Care) Am I ON TRAC for Adult Care c) Assess if adolescent development theory is an appropriate framework for preparing youth with SHCN for transition to adult medical care Four complementary and interrelated theoretical frameworks: developmental theory, Orem’s self-care theory, transtheoretical/stages of change model, and Meleis’ transition theory were used to address the research questions and study objectives.  First, developmental theory is the foundation for many transition programs, including ON TRAC, and a basic understanding of this theory was required to assess if it is the most suitable framework for preparing adolescents with SHCN for successful transition to adult health care services.  A key component of successful transition is adolescents’ ability to independently manage their health condition and long-term disability needs.  Pertinent health care transition (HCT) policies, guidelines and best practices stress the importance of promoting self-reliance and personal responsibility through knowledge   46 and skill development.  Although not explicitly acknowledged in the HCT literature Orem’s self- care theory is inherent to the HCT process. Next, evaluating the TRAQ included examining the aptness of the use of the transtheoretical (stages of change) model as a framework for assessing youths’ stages of readiness for adopting specific self-care behaviours.  Finally, the middle-range transition theory of Meleis and colleagues (2000) provided a nursing perspective that complements the previously mentioned theories, recognizes the complexity and multidimensionality of transition, and accentuates the philosophy of holistic care, which is also a fundamental element of HCT for adolescents with SHCN. Theoretical Frameworks Development Theory Adolescent development theory is a prevalent theoretical framework in transition literature, research, and program recommendations.  Developmentally appropriate care is a fundamental principle of transition services in policies by the Society for Adolescent Medicine (Blum et al., 1993; Rosen et al., 2003), American Pediatrics, American Academy of Family Physicians and American College of Physicians – American Society of Internal Medicine (2002), and the Canadian Paediatric Society (2011).  The ability to make appropriate health care related decisions is a vital component of successful transition.  Rew (2005) recognizes decision making as a critical aspect of development that occurs during adolescence.  Viner (2008) identified fourteen elements of successful transition that included developmentally appropriate education programs that focus on medical, psychosocial and vocational components of care.  Transition models, such as ON TRAC, differentiate adolescence into distinct stages that acknowledge the developing capabilities of youth.  Information processing, which includes short and long term- memory, information retrieval, and metacognition, is increased during adolescent development   47 (Rew, 2005).  Developing abstract thinking is a pertinent cognitive change that occurs during adolescence (Kaufman, 2006).  Information processing abilities and abstract thinking greatly enhance one’s ability to problem solve and make decisions which are crucial to transitioning to the adult care system (Kaufman, 2006).  Applying developmental theory to transition models is a method for concentrating on how to actively engage youth in the health care process and promote self-care skills that are appropriate for each stage of adolescence. The ON TRAC transition model emphasizes the need for developmentally appropriate care throughout the transition process (Paone et al., 2006).  It is based on the stages of normal adolescent developmental and is divided into three age-specific transitional stages: early (10-12 years), middle (13-15 years), and late (16-18 years) (Paone et al., 2006).  Each transitional period identifies developmental tasks related to transition that youth should achieve during that stage. For example during middle transition youth 13-15 years of age should be gaining an understanding of the transition process, practicing skills, gathering information, and setting goals to participate in their own care (Paone et al., 2006).  This model has established specific indicators and strategies for each developmental stage to assist adolescents in acquiring the necessary knowledge and skills to smoothly progress towards independently managing their health care (Paone et al., 2006). Orem’s Self-Care Theory Recommendations and guidelines for successful transition consistently stress that adolescents must have strong self-management skills, be actively engaged with and take responsibility of their health care (Paone et al., 2006; Sawicki et al., 2009; Tuchman et al., 2008; Viner, 2008; Visentin et al., 2006).  Despite rarely being acknowledged as a theoretical underpinning, Orem’s theory of self-care is central to the transition process.  As cited by Rew   48 (2005) Orem defined self-care as: “the personal care that individuals require each day to regulate their own functioning and development” (p. 129).  Adult care services expect patients to be autonomous, make independent health related decisions, and be responsible for comprehensively managing their health care (Hudsmith & Thorne, 2007; Visentin et al., 2006).  Consistently practicing the self-care skills required by adult health care services is a major challenge for youth transitioning from paediatric services (Visentin et al., 2006).  Betz and colleagues (2003) and Betz (2004) emphasize the need for adolescents to be self sufficient in regards to managing their health condition at the time of transition.  This entails having both the knowledge and skills to competently manage daily health care self-care needs as well as long-term disability needs (Betz et al., 2003). Orem defines self-care as “the performance or practice of activities that individuals initiate and perform on their own behalf to maintain life, health, and well being” (Wilson & Gramling, 2009 p.853). The HCT literature and research has established that transition preparation must promote independence and equip adolescents with the necessary knowledge and skills to successfully navigate the adult health care system.  The required knowledge and skills adolescents must possess to successfully transition to adult care are important components of the self care youth with SHCN need to practice daily.  Self-care agency and self-care requisites, two of the five concepts of Orem’s self-care theory, are especially pertinent to the HCT process.  As cited by Rew (2005) self-care agency pertains to the power and motivation a person has to manage healthy functioning and participate in self-care behaviours, and self-care requisites are the behaviours required to maintain life.  Both of these concepts are embedded in the objectives of facilitating an effective and efficient HCT.  Transition readiness assessment checklists, questionnaires, and tools evaluate adolescents’ self-care agency and self-care requisite   49 capabilities.  For example both domains of the TRAQ, chronic condition self-management and self-advocacy and health care utilization, involve self-care agency and requisites.  The self- advocacy domain includes skills related to maintaining health insurance, connecting with community services and communicating with the health care team (Sawicki et al., 2011).  The self-management domain is comparable to self-care requisites as it encompasses the skills needed for chronic disease self-management that are essential to maintaining positive health outcomes (Sawicki et al., 2011). Before transferring to adult care adolescents need to develop and be able to perform important self-care skills on a day-to-day basis. In addition, self-care theory takes into account that stage of development and maturity influence self-care behaviour and includes developmental as one of the three categories of self-care requisites (Rew, 2005). According to Orem’s theory, as cited by Rew (2005), for self-care practices to be therapeutic they must promote well-being, sustain life and closely correspond with the ultimate goal of transition, which is for youth with SHCN to be independent, self-sufficient, and maximize their life long functioning. Transtheoretical/Stages of Change Model Health care transition is a significant change for adolescents with chronic medical conditions, and is often a forced change (Visentin et al., 2006).  As discussed previously, it is essential that youth develop the necessary skills to successfully adapt and thrive in adult services (Viner, 2008).  Readiness to change and adopt self-care behaviours have been associated with an individual’s compliance to a medical regime and viewed as part of the learning process (Visentin et al., 2006).  Sawicki et al. (2011) used the transtheoretical model, also known as the stages of change model, as the framework to develop the TRAQ tool.  Sawicki et al. (2011) determined that the transtheoretical model was appropriate because it focuses on the individual’s decision-   50 making behaviour, which is key to acquiring new behaviours such as self-management skills. The transtheoretical model, as cited in Rew (2005), was developed by a group of behavioural scientists that discovered individuals who changed addictive behaviours moved through a sequence of stages.  It concentrates on decision-making and motivation as a state of readiness to progress through predictable and intentional stages of change and identifies six stages of change: precontemplation, contemplation, preparation, action, maintenance, and termination (Rew, 2005). It has been used to change both problem behaviour and for adopting new health behaviours (Rew, 2005). The TRAQ uses the transtheoretical model to assess youths’ readiness for transition. Adolescents are asked to describe their self-care skills and behaviours according to a five-point scale, which signifies which stage of change they are currently in for that particular skill/behaviour.   Identifying the adolescents’ current stage of change for these essential behaviours and skills enables clinicians to identify deficits and provide stage appropriate interventions and support for promoting necessary skill development (Sawicki et al., 2011). Transition Theory The emerging middle-range transition theory by Meleis et al. (2000) expands on the previous work of Meleis and her nursing colleagues (Chick & Meleis, 1986; Meleis & Trangenstein, 1994; Schumacher & Meleis, 1994) to further develop nursing therapeutics and interventions to improve health outcomes during critical periods.  This middle-range theory acknowledges that transitions are complex and multidimensional, identifies properties of the transition experience, and distinguishes between process and outcome response indicators (Meleis et al., 2000).  In their earlier concept analysis of transition, Chick and Meleis (1986) recognize that the concept of transition is prevalent throughout developmental, stress, and adaptation theories but built their conceptualization of the term from the first meaning in   51 Webster’s International Dictionary and define transition as: “…a passage from one life phase, condition, or status to another, is a multiple concept embracing the elements of process, time span, and perception” (p.239).  Furthermore, they propose that transition is inextricably linked to nursing theory, which addresses change due to health-illness events (Chick and Meleis, 1986). The emerging middle-range theory aligns well with the Society for Adolescent Medicine’s holistic definition of transition as it recognizes that events, experiences, and responses of clients are processes that require a longitudinal and multidimensional approach (Meleis & Trangenstein, 1994).  This theory identifies that transitions are not discrete or mutually exclusive events, but that multiple transitions can take place simultaneously, overlap, and can be related to one another (Meleis et al., 2000).  The concept of individuals experiencing multiple transitions at one time is particularly pertinent to adolescents with SHCN as transitioning to adult care is just one of the transitions that they are faced with during the adolescent period which itself is a developmental transition.  Meleis (2010) distinguishes between a healthy and unhealthy or ineffective transition.  A healthy transition, determined by both process and outcomes indicators, occurs when an individual is able to demonstrate mastery of the behaviours, sentiments, cues and symbols associated with their new role and/or identity (Meleis, 2010).  Whereas, an unhealthy transition results in role insufficiency and difficulties in understanding and/or performing a role or of the beliefs and goals associated with the role behaviour (Meleis, 2010). This theory has three major components of transition: nature of transitions, transition conditions, and patterns of response (Meleis et al., 2000).  Within each major component are factors that influence and affect the transition experience.  Nature of transition is composed of the following three aspects of transition: types, patterns, and properties.  Five essential and   52 interrelated properties of transition are described: awareness, engagement, change and difference, transition time span, and critical points and events (Meleis et al., 2000).  These key properties are frequently discussed throughout the HCT literature and acknowledged as significant transition issues and concerns.  Personal preparation and knowledge, one of the transition conditions that facilitate or inhibit transition, is extremely relevant to HCT and is addressed in this theory. Finally, the emerging middle-range theory accentuates that both process (feeling connected, interacting, location and being situated, and developing confidence and coping) and outcome indicators (mastery and fluid integrative identifies) need to be assessed and reassessed during the various stages of transition (Meleis et al., 2000). Meleis and colleagues (2000) identify transitions as a vital time for HCP to facilitate positive health outcomes by providing appropriate therapeutics to best meet patients’ needs. Focusing on both the process and outcome enables HCP, especially nurses, to assess what stage of transition a person is in and use interventions and strategies that address the individual’s current needs to facilitate healthy transition outcomes (Meleis & Trangenstein, 1994; Meleis et al., 2000).  This theory can be used to enhance current transitional care services and help develop strategies that can be used when the health care transition becomes disrupted. Furthermore, it demonstrates how nursing therapeutics can be used throughout all stages of transition to prevent negative health consequences and improve patients’ well-being (Schumacher & Meleis, 1994). Study Design This descriptive correlational, cross-sectional study evaluated the usefulness of two HCT readiness assessment questionnaires and assessed if the developmental framework of the ON TRAC model meets the developmental needs of adolescents with SHCN.  Adolescents with SHCN between 12-19 years of age who were currently outpatients of four clinics at BC   53 Children’s Hospital (BCCH) and their parents/guardians were recruited to complete the study package.  Evaluation of the TRAQ and validation of the Am I ON TRAC for Adult Care questionnaire were conducted using statistical procedures to determine if both instruments are reasonable measures of readiness for transition from paediatric to adult health care services. Statistical analyses were also used to correlate participants’ scores on the ON TRAC, with age, psychosocial maturity, ON TRAC stage, and the number of years participants had the condition and to compare mean scores between genders and ON TRAC stages. Population and Sample  The population of interest was adolescents with SHCN that will continue to require medical care in adulthood.  This Canadian study recruited participants from BCCH, which is the largest full-service child health centre in the country (BCCH, 2010).  The sample was drawn from outpatient clinics at BCCH that provide care for adolescents with SHCN who will be required to transition to adult health care services in the future.  Four clinics were identified as having appropriate patient populations from which to recruit: diabetes, cardiology, gastroenterology (GI), and neurology.  Inclusion criteria for youth participants were: adolescents between 12-19 years old, currently outpatients at one of the previously mentioned BCCH clinics, and having a SHCN that will require transitioning to an adult health care clinic.  Inclusion criteria for parent participants were: must be a parent/guardian of an adolescent who meets the study inclusion criteria, and must be present with their child at the clinic.  Youth participants were excluded from the study if they had a developmental delay because this is identified in the literature as a major impediment to the HCT process (Reiss et al., 2005; Sawicki et al., 2011) and would interfere with determining if transition programs are meeting the developmental needs of adolescents with SHCN.  Participants were also excluded if they did not speak English, as the   54 investigation instruments were only available in English.  Sawicki et al. (2011) had a sample of 192 for their initial validation of the TRAQ, which categorized participants by their primary diagnosis.  Of their 192 participants 80% had an activity limiting physical condition and the remaining 20% had either cognitive impairment or a mental health condition (Sawicki et al., 2011).  The present study concentrated on a variety of activity limiting physical conditions to further evaluate the TRAQ and validate the revised Am I ON TRAC for Adult Care.  It was estimated that approximately 20 to 30 participants in each age category and a total sample of 200 participants, which is equivalent to the sample used to validate the TRAQ, was required to meet the study objectives.  A convenience sample of 200 eligible youth who met the study criteria was obtained. Investigation Techniques  The study was guided by the Society for Adolescent Medicine’s holistic definition of transition, “…the purposeful, planned process that addresses the medical, psychosocial, and educational needs of adolescents and young adults with chronic physical and medical conditions as they move from child-centered to adult oriented healthcare systems” (Blum et al., 1993).  In addition, the purpose for assessing adolescents’ readiness for transition is to facilitate efficient and successful transitions to adult oriented care.  For the purpose of this research a successful transition was congruent with Meleis’ (2010) characterization of a healthy transition.  A healthy or successful transition occurs when the adolescent is able to demonstrate mastery of the behaviours, knowledge, and skills required to navigate the adult health care system and is determined by both process and outcome indicators.  The youth study package consisted of four self-report questionnaires: the TRAQ, Am I ON TRAC for Adult Care questionnaire, the work orientation and self-reliance sub-scales of the   55 Psychosocial Maturity Inventory (PMI), and the Child’s Report of Parental Behaviour Inventory (CRPBI) and four demographic questions (name of the clinic they attend, age, gender, and age at diagnosis).  Parents/guardians completed a parent version of the Am I ON TRAC for Adult Care, the Parent-Child Relationship Inventory (PCRI), and also provided demographic information (name of the clinic their child attends, their child’s age and gender, child’s age at diagnosis, their relationship with the child, and their own gender).  All the measures use Likert-type response scales and have reading levels appropriate for adolescents 12-19 years of age.  Both the TRAQ and Am I ON TRAC for Adult Care assess youths’ preparedness for transition by evaluating if they possess key behaviours, knowledge and skills that have been identified as essential for successfully transitioning to adult care.  The PMI is based on the concept of psychosocial maturity and was used to examine how well participants’ PMI scores correlate with their age and transition readiness scores.  The CRPBI and PCRI are measures of family relationship and interconnectedness.  The parent ON TRAC, CRPBI and PCRI will be analyzed in a separate project and will be used to explore the concordance between adolescents’ and their parents’ ON TRAC scores and potentially to help understand whether youth’s self-assessment of readiness for transition is influenced by how much parents support or try to control their teens with SHCN. Descriptive statistics were used to summarize the demographic information.  Factor analysis was performed to assess the component structure and usefulness of the TRAQ for this sample.  To validate the Am I ON TRAC for Adult Care questionnaire factor analysis was conducted and ON TRAC scores were correlated with age, PMI scores, ON TRAC stage, and years with condition.  These statistical procedures provided evidence that was used to determine if the ON TRAC model is a developmentally appropriate framework for youth with SHCN and   56 whether the two HCT questionnaires are reasonable and useful measures of transition readiness for Canadian adolescents with SHCN between the ages of 12-19. Instrumentation Transition Readiness Assessment Questionnaire. Participants completed the Transition Readiness Assessment Questionnaire (TRAQ), a self-administered instrument for adolescents with special health care needs (SHCN), which assesses preparedness for transition by measuring skills needed to effectively transfer from paediatric to adult health care settings (Sawicki et al., 2011). In their 2011 study Sawicki et al. deemed the TRAQ to be both a practical and reliable tool for assessing adolescents with chronic medical conditions readiness for transition.  The TRAQ is a 29-item questionnaire with two domains, skills for chronic condition self-management and skills for self-advocacy and communication with health providers, that uses the transtheoretical model as its’ framework to identify a youth’s stage of readiness for transition (Sawicki et al., 2011).  Each item requires participants to choose from a six-point ordinal scale of predetermined answers that represent each stage of change: pre-contemplation, contemplation, preparation, action and maintenance plus a not applicable option.  Each stage of change is given a distinct score where one represents the pre-contemplation stage and five the maintenance stage, with zero indicating that the skill or behaviour is not applicable to that individual.  Sawicki et al. divided their sample into three age groups: 16-18, 19-20, 21-26 years, which they propose represent the early, middle, and late transition age groups.  The TRAQ ages do not correspond with the ON TRAC stages of transition that define early as 10-12 years, middle as 13-15 years, and late as 16-18 years. Additionally, transfer from BCCH outpatient clinics to adult medical clinic is supposed to occur when patients are 18 years of age.  Due to the discrepancy between the ages and stages of   57 transition it was deemed important to evaluate the psychometric properties of the TRAQ with a younger population whose ages represent the ON TRAC stages of transition. Am I ON TRAC for Adult Care Questionnaire.  The ON TRAC project was initiated to aid in the development of a transition planning program for the large number of adolescents who transfer to adult care from BCCH each year (Paone et al., 2006).  ON TRAC was built using four central concepts: health promotion, empowerment, self-efficacy, and partnerships (Children’s & Women’s Health Centre of British Columbia, 2000).  The objective was to create a 2-tiered service that provided 1) education and information to assist HCP with transition issues and develop skills and strategies to integrate HCT planning into their practice and 2) to support HCP as they educate and empower youth and their families about transition (Paone et al., 2006).  The Am I ON TRAC for Adult Care questionnaire was created as one of the many tools, strategies, and initiatives of this transition planning framework.  The ON TRAC framework accentuates a holistic approach to transition planning that addresses six components of transition: 1) self-advocacy and self-esteem, 2) independent health care behaviours, 3) sexual health, 4) social supports, 5) educational, vocational, and financial planning, and 6) health and lifestyle (Paone et al., 2006).  The original Am I ON TRAC for Adult Care questionnaire consists of 23 Likert-response items and assessed youths’ capabilities of performing life skills required to actively participate in their health care and function independently in society and included both youth and parent versions.  This original measure was never validated and its psychometric properties have not been evaluated. The Am I ON TRAC for Adult Care has been adapted for use because the HCT literature has identified a need for a transition readiness measure.  The recently validated TRAQ is predominantly medically focused and does not include measures about normative adolescent risk   58 behaviours that may complicate the health of youth with SHCN.  In contrast, the ON TRAC questionnaire was developed from a holistic perspective and assesses the many facets of adolescent health and well-being including risk taking behaviours related to alcohol, drugs and sexual practices.  These are pertinent health related topics to address as the literature provides evidence to demonstrate that adolescents with SHCN are just as likely or more likely to engage in risk behaviours (Smith, Stewart, Poon, Saewyc, & the McCreary Centre Society, 2010; Smith et al., 2011).  An extensive review of current hospital transition programs was completed and a list of transition indicators was compiled.  Clinicians at specialized medical clinics at BCCH were asked to review the indicators, provide feedback and rank the indicators based on their importance.  The modified version of the measure was developed based on a review of the HCT literature, the indicators, and clinicians’ feedback.  During the revision of the Am I ON TRAC for Adult Care the goal was to ensure questions were not double-barreled, were clearly worded, and included the knowledge and skills that have been identified as essential for successful HCT. The modified youth version of the Am I ON TRAC for Adult Care is comprised of 26 questions that are phrased as either knowledge or behaviour based statements.  The knowledge statements, which form the knowledge subscale, use a four-point Likert-type scale ranging from strongly disagree to strongly agree and purposefully do not include a neutral response option. The behaviour statements, which are grouped as an index, employ a five-point Likert-type response scale that ask youth to report how frequently they engage in particular health related behaviours.  Responses for the behaviour statements range from never to always.  The corresponding parent version of the Am I ON TRAC for Adult Care has rephrased 25 statements from the youth measure to ask parents to report on their child’s health care related knowledge and behaviours.  One statement, I feel my family cares about me, was not included in the parent   59 version because it was thought that rephrasing the question to the parent perspective made it irrelevant.  Moreover, this question is not needed for the purposes of this research, as the PCBI was included in the study to collect data about family relationships. Psychosocial Maturity Inventory.  The Psychosocial Maturity Inventory (PMI) was developed by Greenberger and Sørensen (1974) and is based on their concept of psychosocial maturity.  The PMI uses a general model of maturity, combining goals of socialization with goals of development, and is concerned with the survival of both the individual and the society (Greenberger & Sørensen, 1974; Greenberger, Jossellson, Knerr, & Knerr, 1975).  Greenberger and Sørensen (1974) proposed that psychosocial maturity is an integrative concept that examines the requirements societies universally impose on mature individuals.  According to Greenberger et al. (1975): …in all societies ‘socialized’ and ‘developed’ individuals should be self-sufficient in some degree and take responsibility for their own survival, should be able to relate to others in stable and predictable ways, and should be able to meet threats to the integrity of the social group with efforts to restore social solidarity (p.128). The model identifies that psychosocial maturity is comprised of three general capacities that societies expect mature individuals to possess: the capacity to function on their own, the capacity to interact adequately with others, and the capacity to contribute to social cohesion (Greenberger & Sørensen, 1974; Greenberger et al., 1975).  The PMI consists of nine attributes that form the instrument’s subscales (self-reliance, identity, work orientation, communication skills, enlightened trust, knowledge of major roles, social commitment, openness to sociopolitical change, and tolerance of individual and cultural differences) that form the instrument’s three   60 subscales (individual adequacy, interpersonal adequacy, and social adequacy) and are indicators of the general capacities of mature individuals (Greenberger et al., 1975). The purpose of Form B of the PMI was to produce a single inventory of a practical length (Greenberger et al., 1975).  It contains 188 items dispersed among the inventory’s three subscales (Greenberger et al., 1975).  Each item is scored on a 4-point scale labeled “strongly agree”, “agree slightly”, “disagree slightly”, and “strongly disagree” with the high score depicting the most mature response (Greenberger et al., 1975).  The individual adequacy subscale of the PMI captures the individual’s capacity to function adequately on one’s own.  The self-reliance and work orientation, two of the three individual adequacy attributes, are especially relevant to assessing adolescents’ readiness for HCT and determining if the ON TRAC framework is developmentally appropriate for this population.  The PMI is a validated tool that is able to document the course of psychosocial maturity over the school years, to examine youth of different ages using the same measure, and is appropriate for comparing groups of individuals (Greenberger et al., 1975). Child’s Report of Parental Behaviour Inventory & Parent-Child Relationship Inventory. The data collected from these measure were not used within this thesis work and are included as part of a larger study being conducted by the principal investigator and two of the co- investigators.  This larger study will explore whether readiness among older adolescents differs by parental support and control.  The Child’s Report of Parental Behaviour Inventory-Short (CRPBI-S) is a brief 30-item instrument that is concrete enough to use with younger adolescent samples, has been used across various ethno-cultural backgrounds, and has a comparable parent measure – Parent-Child Relationship Inventory (PCRI) (Saewyc, 2010).  Internal consistency   61 reliability and factor analyses have been conducted for both the CRPBI-S and PCRI.  Saewyc (2010) validated the CRPBI-S subscales: the Mother Support and Mother Control subscales, the Father Support and Father Control subscales and the Parent Support and Parent Control subscales of the PCRI. All of these subscales were found to have acceptable internal consistency reliability, with all alphas above .75, and unidimensionality based on principal components factor analysis (Saewyc, 2010). Data Collection Data collection took place at BCCH’s diabetes, cardiology, gastroenterology and neurology outpatient clinic waiting rooms.  Gastroenterology patients were also recruited from the medical day unit during routine visits.  Study investigators contacted the medical specialty clinics (diabetes, cardiology, gastroenterology, and neurology) via e-mail that provided a brief description of the study and an information page which outlined: the research goals and purpose, participant recruitment process, type of data collection and collection procedures, potential risks and benefits of participation, length of the study, participant time commitment, involvement and time commitment of clinic staff and how the data would be used.  After this initial contact with the clinics the graduate student researcher followed-up with each clinic to discuss in detail how recruitment would take place to ensure minimum disruption of clinic operations.  A study flyer was posted in the waiting rooms of each participating clinic.  During the recruitment period the diabetes clinic included a short description of the study and informed patients that they might be asked to participate in the study at their next appointment in their mailed appointment reminders. The graduate researcher was given permission to be present in the clinic waiting rooms during arranged clinic days and worked closely with the nursing and clerical staff at each clinic. Youth and their parents who were deemed eligible for the study were approached by the graduate   62 researcher in the clinic waiting rooms and provided with the study information letters.  The informational letter clearly stated: participant confidentiality would be maintained throughout the study, choosing not to participate would in no way effect the care received, participants had the right to withdraw from the study, and that completing the questionnaire indicated that they had consented to participate.  In the waiting rooms potential participants had the opportunity to ask the researcher questions regarding the study and the researcher was able to address their concerns. After reviewing the study information letters adolescents and their parents who agreed to participate were provided with the appropriate study package to complete.  It was made clear to participants that the questionnaires had to be completed and returned to the investigator by the end of their clinic visit. Completion of the youth study packages took between 15 - 20 minutes depending on the participants’ age and reading abilities.  The parent questionnaire was shorter and on average took less than ten minutes for parents to fill-out.  Youth and parents completed the questionnaires independently but were provided clarification and assistance from the researcher as needed. Youth were given a ten dollar iTunes gift card upon return of their study package to the researcher.  The data was collected over a three-month period (April to June 2012). Data Analysis Plan  As the data was collected it was coded and entered into an excel spreadsheet by the graduate student researcher.  Upon completion of data collection a research assistant checked for entry errors and the cleaned excel file was then imported into SPSS 20 (Statistical Package for the Social Sciences) for statistical analyses.  The data had different levels of measurement including: categorical (gender, outpatient clinic), ordinal (ON TRAC stage of transition), and interval (age, years with condition, PMI score, ON TRAC scores).  The independent variables of   63 interest were: age, gender, clinic, years with condition, ON TRAC stage of transition, and PMI score.  The original plan included three dependent variables: TRAQ scores, the ON TRAC knowledge score and the ON TRAC behaviour index score. Based on the review of the literature it was hypothesized that participants’ transition readiness assessment scores from both the TRAQ and the ON TRAC questionnaires would differ based on age, PMI score, and number of years adolescents had the condition.  Before beginning the statistical analyses the data was examined to determine which statistical approaches would be appropriate.  This involved examining frequency distribution, measures of central tendency and variability, plotting histograms, assessing kurtosis and skewness, and using the Kolmogorov-Smirnov test (Field, 2009).  The data was also checked for missing responses and outliers. This information was used to describe and summarize the demographic and clinical characteristics of the study sample.  Factor analyses of the PMI and TRAQ were conducted to examine the psychometric properties of both instruments and evaluate the appropriateness of the TRAQ with a younger Canadian sample.  An exploratory factor analysis (EFA), specifically principal component analysis (PCA), of the ON TRAC knowledge scale was performed.  An exploratory approach was chosen because the ON TRAC knowledge scale has had no prior validation and this approach is most appropriate during the initial stages of measurement development when the research goals include explanation and data reduction (Floyd & Widaman, 1995).  The PCA method was chosen because it is a reliable approach for the exploratory development of measures and is not susceptible to improper solutions and factor indeterminacy issues like common factor analysis (Velicer & Jackson, 1990).  Velicer and Jackson (1990) emphasize that the mathematical procedures used in common factor analysis and PCA are similar and when the   64 same number of factors are extracted the solutions are usually comparable.  Moreover, the theory of self-care management and explicitly transition readiness lacks a soundly theorized latent construct; thus the study objective to identify observable or manifest variables aligns best with the purposes of PCA (Velicer & Jackson, 1990). The literature advises against using the Kaiser criteria (eigenvalue greater than one) for determining the number of components to retain after factor analysis because it is criticized for being inaccurate and overestimating the number of components (Patil, Singh, Mishra, & Donavan, 2008; Velicer & Jackson, 1990; Zwick & Velicer, 1986).  Zwick and Velicer (1986) evaluated five methods (Bartlett’s Test, Kaiser’s criterion, minimum average partial, scree test, and parallel analysis) for determining the appropriate number of components to extract and found parallel analysis to be the most frequently accurate method, the scree test to have moderate overall reliability and did not recommend Kaiser’s criterion due to its consistent over estimation. As a result, the literature advocates that researchers use these alternative methods for determining the number of components.  This study used both the parallel analysis and scree test methods to ensure component extraction was as accurate as possible.  A parallel analysis engine available online by Patil, Singh, Mishra & Donavan (2007) was used to compare eigenvalues obtained in the raw datasets with eigenvalues extracted from a randomly generated correlation matrix that had the same number of variables and sample size (Patil et al., 2008).  The engine generated a hundred random correlation matrices and provided the 95th percentile eigenvalues, which were then compared to the corresponding eigenvalue from our dataset.  Components were retained only when the eigenvalues from the present study’s raw data were larger than the corresponding eigenvalues produced by the randomly generated correlation matrix (Patil et al., 2008).   65  A theorized cut-off score was established for the ON TRAC behaviour index to identify the required skill sets that adolescents must achieve at a minimum level in order to be considered ready to transition to adult medical care.   Next, the distributions of these scores were examined and it was determined that they met the assumptions of normality (normally distributed, homogeneity of variance, interval data, and independence).  Since the data was found to be normal Pearson’s correlation coefficient was used (Field, 2009) to explore the relationships between the dependent variables and each of the independent variables.   To evaluate if the normal stages of adolescent development, as outlined by the ON TRAC model, are appropriate for youth with SHCN participants’ PMI scores were compared to their age and ON TRAC stage.  The relationships between these variables were then examined to assess whether participants’ scores fit the same stage-for-age that are outlined in the ON TRAC model’s normative stages of adolescent development trajectory.  Finally, linear regressions of both the ON TRAC knowledge scores and behaviour index raw scores were performed to investigate how the independent variables contribute to these scores. Rigor Efforts to enhance and ensure rigor were made throughout each phase of the study. Strategies that were employed included: clear participant inclusion and exclusion criteria, using a reliable and validated instrument to assess psychosocial maturity, and recruiting participants from multiple clinics.  A research assistant double-checked and cleaned the data.  Additional strategies such as seeking statistical consultation during the data analysis phase were also employed.     66 Ethical Considerations  This study did not involve any invasive procedures and was non-clinical.  Approval from the University of British Columbia’s (UBC) Behavioural Research Ethics Board (REB) and the Child Family Research Institute (CFRI) at BCCH was obtained because the research involved participation of human subjects.  All potential participants were provided with a study information letter that outlined the following: research goals and purpose, participant selection, type of data collection and collection procedures, potential risks and benefits of participation, length of the study, participant time commitment, how the data will be used, and additional study details.  The study information letter informed potential participants that they would be providing their consent by completing the study package.  The researcher’s contact information was included on the study information letter and participants could contact her if they had any questions or concerns regarding the study. Participant confidentiality was maintained throughout the study.  Data collection was anonymous, as participants did not provide any identifying information such as their names on the study forms.  All completed study packages were assigned an identification number for study purposes.  All questionnaires are securely stored in a locked filing cabinet in a locked office in the School of Nursing at the University of British Columbia.  Data entry was performed on password protected computer programs and only password protected data devices were used for transferring study information between computers. The data collected during this study was used only for the purposes specified in the study information letters.  Study data will be kept for five years after the final data analysis and study reports have been completed.  After this five-year period all study surveys and questionnaires will be shredded and computer files erased.   67 CHAPTER FOUR: FINDINGS Introduction  The following chapter presents the evaluation of the Transition Readiness Assessment Questionnaire (TRAQ), the results of the validity and reliability of the ON TRAC readiness measure, and the correlations between age, psychosocial maturity, and readiness to transition scores.  The chapter describes: a) the sample characteristics, b) factor analysis of the Psychosocial Maturity Inventory (PMI) and support for the use of a modified 18-item measure, c) factor analysis of the TRAQ, d) exploratory factor analysis of the ON TRAC knowledge scale, e) establishment of the ON TRAC behaviour index scores, f) the relationships between age, PMI scores, years with condition and ON TRAC knowledge and behaviour scores, g) comparison of PMI and ON TRAC scores by ON TRAC stage and gender and h) linear regression of the ON TRAC scores to understand the contribution and interrelationship of age, psychosocial maturity, gender, and years with condition for predicting readiness to transition. Sample Characteristics  A convenience sample of 201 youth and 192 parents were recruited from the diabetes, cardiology, gastroenterology and neurology outpatient clinics at BCCH. Two hundred and fifteen youth and 209 parents were approached to participate.  Each of the four clinics had individuals who declined participation.  Parents/guardians of six adolescents were not present during recruitment and an additional three parents declined.  Only one youth-parent pair had incomplete study packages and was considered a refusal.  Participation rate was 93.5% and 91.9% for youth and parents respectively.  Data from one youth-parent pair was not included in the data analysis because the youth did not meet the inclusion criteria of being diagnosed for at least six months. As a result, data from 200 youth has been analyzed.   68 Participants ranged in age from 12-19 with a mean age of 15.33 (SD=1.88).  Fifty-seven percent were male, and participants were distributed across the clinics, with 36% from diabetes, 27% cardiology, 27% gastroenterology, and 10% neurology.  Table 1 provides the demographic information of the sample. Table 1 Descriptive Statistics of the Youth Demographic Characteristics  Variable n % M (SD) Range Gender Male 115 57.5 Female 85 42.5 Clinic Diabetes 72 36 Cardiology 54 27 Gastroenterology 54 27 Neurology 20 10 ON TRAC Stage Early (12 years) 19 9.5 Middle (13-15 years) 82 41 Late (16-19 years) 99 49.5 Age   15.33 (1.88) 12-19 Years with condition   8.03 (5.46) 0.5-19 N = 200 Psychosocial Maturity Inventory The PMI was developed and originally validated in 1974 and has since been used as a measure of youth psychosocial maturity.  A principal component analysis (PCA) of the PMI was performed to verify its factor structure and appropriateness for adolescents in the present study because the PMI has primarily been validated with other populations and investigators were also concerned that items may be outdated. In the development of the PMI it was theorized that each of its nine subscales would combine and form three factors representing the main categories of psychosocial maturity (Greenberger et al., 1975).  The predicted structure of the Individual Adequacy category was   69 initially confirmed with factor analysis; furthermore, the three Individual Adequacy subscales plus a communication subscale formed the majority of the second factor (Greenberger et al., 1975).  Since only items from two subscales of the Individual Adequacy component were used in this study an essentially unidimensional factor structure was anticipated. A principal component analysis (PCA) without rotation was conducted. The inter-item correlation matrix was examined to identify items that correlate < .30 or > .90 with the other variables.  Six items (1, 3, 4, 8, 13 and18) had correlations of less than .30 with all other variables and were recognized as potential problems.  The remaining 14 items had correlations ranging from r = .31 to r = .55 with one to six items.  There were no correlations > .90 in the matrix.  The Kaiser-Meyer-Olkin measure (KMO) of sampling adequacy was .79 and Bartlett’s test of sphericity, χ2 (190) = 800.96, p < .001, which verified that the sample size was large enough and a factor analysis was useful (Tabachnick & Fidell, 2007).  A parallel analysis (see Table 2) and the scree plot both supported a two component solution.  The first and second components had eigenvalues of 4.65 and 1.79 respectively, and accounted for 23.3% and 8.9% of the explained variance.  There was a substantial amount of cross loading over the seven components yet 16 items had loadings > .4 with 12 items loading highest on first component. Only two items (8, 13) had loadings < .3 on component one.  Based on these results, an 18-item single factor scale was interpreted as the optimal solution. Table 2 Parallel Analysis for 20-items and n = 183 Component 95 th Percentile Eigenvalue Obtained Eigenvalue 1 1.77 4.65 2 1.61 1.79 3 1.50 1.30 Note: 95th percentile eigenvalue obtained from parallel analysis engine (Patil et al., 2007)   70 A second PCA on a modified 18-item PMI (excluding items 8 and 13) without rotation was performed to test the above conclusion.  The KMO was .80 and Bartlett’s test of sphericity, χ2 (153) = 757.39, p < .001, was significant indicating the sample size was large enough and a factor analysis was appropriate (Tabachnick & Fidell, 2007).  A parallel analysis (see Table 3) and the scree plot continued to support a two component solution. Component one had an eigenvalue of 4.58 and accounted for 25.5% of the explained variance.  Component two had an eigenvalue of 1.74 and explained 9.7% of the total variance.  Twelve items loaded highest on the first component with 16 items loading > .4, and no items loading < .3.  The number of items that cross loaded also decreased. This provides further support for a single factor 18-item measure. Table 3 Parallel Analysis for 18-items and n = 183 Component 95 th Percentile Eigenvalue Obtained Eigenvalue 1 1.69 4.58 2 1.56 1.74 3 1.45 1.30 Note: 95th percentile eigenvalue obtained from parallel analysis engine (Patil et al., 2007) A final PCA with only the 16 items that loaded > .4 on the first component was conducted (excluding items 3, 8, 13, and 19).  Sample size was adequate, KMO = .80, and Bartlett’s test of sphericity, χ2 (120) = 674.99, p < .001, was significant.  Both the parallel analysis (n=185, 16 items) and the scree plot indicated a two component solution.  The first two components had eigenvalues of 4.35 and 1.64 and accounted for 27.2% and 10.3% of the explained variance.  Items still cross loaded but all 16 items had loadings greater > .4 on the first component and 14 items loaded highest on this component. These results suggest that a 16-item modified PMI may also be justified.   71  Internal consistency and reliability statistics were run for the 20, 18 and 16-item scales and Table 4 displays the results.  It was determined that the essentially unidimensional 18-item PMI would be used in this study because Cronbach’s α was highest for the 18-item measure and the investigators were concerned that too much information would be lost with the 16-item measure. Table 4  Reliability of Psychosocial Maturity Inventory  Reliability 20-Item 18-Item 16-Item Cronbach’s α .81 .82 .81 Correlation between forms .62 .65 .65 Spearman-Brown Coefficient .77 .79 .79 Guttman Split-Half Coefficient .77 .79 .79  Two items, eight and 13, were removed from the measure.  Item eight - I get upset if I am not immediately successful in learning something new- is awkwardly worded, suggests impulsivity, and even a sense of self-competence.  It is also negatively worded which further confounds the item.  Research has demonstrated that negatively worded items contribute to problems with internal consistency, factor structures, and response distributions (Barnetter, 2000) and that respondents have more difficulty understanding these items (Hubley & Zumbo, in press).  Item 13 - I seldom get behind in my work - is one of the scale’s two reversed scored items.  An association between reversed items and psychometric problems has also been found and consequently reversed items have been identified as potential sources of misresponse (Swain, Weather, & Niedrich, 2008).  Additionally, during data collection the investigator observed that youth had difficulty with item 13 as participants commonly asked for the meaning of seldom. This indicates that ‘seldom’ may not be used as often as it was when the PMI was developed over thirty years ago. Both items had weak and low correlations (< .24) with all other items,   72 which provided further evidence that they were problematic (Field, 2009).  The component structure of the 18-item PMI is supported by the theory used to develop and originally validate this measure.  This analysis justifies the use of a modified 18-item PMI and demonstrates it is appropriate for the youth in this study.  Except for one item (13), which has been removed, the PMI does not appear to be outdated.  Tables of the component loadings and eigenvalues for the 20 and 18-item can be found in Appendix D. Psychosocial Maturity Inventory Scores  The 18-item PMI has a theoretical range of 18 to 72 and the total score is calculated by summing all responses.  Seventeen items are scored with Strongly Agree = 1, Agree = 2, Disagree = 3, Strongly Disagree = 4 and item 19 is reversed scored (Strongly Agree = 4 to Strongly Disagree =1).  Missing responses for items ranged from zero to five. Seventeen PMI scores could not be calculated due to missing responses resulting in a sample of 183.  Scores ranged from 27 to 72 with a mean of 54.97 (SD=7.25) and were slightly negatively skewed. However, because of the large sample size (n=183) the central limit theorem can be used to assume that the sampling distribution is normal (Field, 2009).  PMI scores were examined based on participant age and revealed only a small difference in mean scores.  Eighteen year olds had the highest mean (M=57.68, SD=6.02) and 14 year olds had the lowest (M=51.71, SD=6.65). There was a small trend for PMI mean score to increase with age, but these findings imply for this sample of adolescents with SHCN age may have a weak association with psychosocial maturity.  Figures 2 and 3 display the mean PMI scores based on age and ON TRAC stage.      73 Figure 2 Mean PMI Scores by Age  Figure 3 Mean PMI Scores by ON TRAC Stage    74 Transition Readiness Assessment Questionnaire The TRAQ was initially validated with a sample of youth aged 16-26 with a wide variety of complex health conditions and participants were categorized by three primary diagnoses: activity limiting physical conditions, mental health conditions, and cognitive impairment (Sawicki et al., 2011).  In comparison, the present sample was much younger (12-19 year olds), comprised of only activity limiting physical conditions and cognitive impairment was a criterion for exclusion.  Additionally, the current study used the most up-to-date version of the measure, TRAQ 4.1 received from the author, which has six response options instead of five. The TRAQ 3.0, used in the original validation study, did not include the ‘not needed for my care’ response. Due to the substantial differences between these two samples and the use of the revised measure it was considered reasonable to do an exploratory analysis, opposed to confirmatory, to evaluate the factor structure and appropriateness of the TRAQ in a younger Canadian sample. The TRAQ was initially validated with an exploratory factor analysis (EFA) with principal components (PCA) and oblique rotation (Sawicki et al., 2011).  The original criteria for retaining items were component loadings > .3 on one component and ≤ .3 on other components. Items were deleted if they had component loadings < .3 or > .3 and equally loaded on other components (Sawicki et al., 2011).  Based on eigenvalues ≥ 1, the scree plot, and percentage of variance explained the developers concluded that the TRAQ had two subdomains: Skills for Chronic Condition Self-Management (items 1-16) and Skills for Self-Advocacy and Communication with Health Providers (items 17-29) (Sawicki et al., 2011).  Due to the distinctions between the American and Canadian health care systems it was hypothesized that the items about health insurance (14 and 15) and payment for medications and medical supplies (3 and 9) may load differently or not at all as the current study had a Canadian sample and the original study was conducted in the United States.   75 In an effort to replicate the original component structure an unconstrained EFA with PCA and oblique rotation was performed on the 29-item TRAQ 4.1.  The r-matrix was examined and all items correlated ≥ .30 with at least one other variable and there were no correlations ≤ .90. However, seven items (20, 23, 25, 26, 27, 28, and 29) had correlations ≥ .30 with only one or two other items, indicating they could be problematic.  The KMO verified the sampling adequacy for the analysis, KMO = .80, and Barlett’s test of sphericity, χ2 (406) = 1827.55, p < .001, indicated that correlations between items were sufficiently large for PCA (Tabachnick & Fidell, 2007).  Since the initial study considered both Kaiser’s criteria and the scree plot to determine the appropriate number of components to extract, the Kaiser’s criteria from our analysis is reported for comparative purposes, even though this method is not recommended (Zwick & Velicer, 1986).  The present PCA produced eight components with eigenvalues larger than Kaiser’s criterion of one; the scree plot suggested a three component solution and a parallel analysis (see Table 5) supported four components. Table 5 Parallel Analysis for 29-items and n = 161 Component 95 th Percentile Eigenvalue Obtained Eigenvalue 1 1.99 7.27 2 1.82 2.62 3 1.73 1.89 4 1.62 1.74 5 1.56 1.46 Note: 95th percentile eigenvalue obtained from parallel analysis engine (Patil et al., 2007)      76 The first eight components of our analysis had eigenvalues ranging from 7.27 to 1.02 and accounted for 25.1% to 3.5% of the explained variance.  These findings are not consistent with the original study that reported four components with eigenvalues > 1 ranging from 9.73 to 1.26 and a scree plot with a clear elbow after the first two components (Sawicki et al., 2011).  Table 6 provides the component loadings, eigenvalues and percentage of variance for our unconstrained PCA of the TRAQ.                                     77 Table 6  Factor Loadings, Eigenvalues and Percentage of Variance Explained for the Unconstrained Principal Component Analysis of the TRAQ  n = 161 Note. Factor loadings > .30 are in boldface.  Only components supported by the parallel analysis are shown.   Component TRAQ Item 1 2 3 4 1. Do you fill a prescription if you need to? .34 -.01 -.63 .11 2. Do you know what to do if you are having a bad reaction to your medications? -.13 -.15 -.50 .13 3. Do you pay or arrange payments for your medications? .29 -.09 -.60 .08 4. Do you take medications correctly and on your own? -.26 -.21 -.73 -.13 5. Do you reorder medications before they run out? .13 -.19 -.67 -.08 6. Do you take care of your medical equipment and supplies? -.19 -.83 -.13 .02 7. Do you call the suppliers when there is a problem with the equipment? .15 -.81 .04 .05 8. Do you order medical equipment before they run out? .04 -.84 -.08 .01 9. Do you arrange payments for the medical equipment and supplies? .09 -.65 -.19 .03 10. Do you call the doctor’s office to make an appointment? .79 -.08 .01 -.03 11. Do you follow-up on any referral for tests or check-ups or labs? .61 .02 -.14 -.04 12. Do you arrange for your ride to medical appointments? .45 -.06 -.11 .05 13. Do you call the doctor about unusual changes in your health (For example: Allergic reactions)? .45 -.22 .11 -.22 14. Do you apply for health insurance if you lose your current coverage? .48 -.15 -.07 .18 15. Do you know what your health insurance covers?  .39 -.06 -.19 -.05 16. Do you manage your money & budget household expenses (For example: Use checking/debit card)? -.03 .01 .01 -.05 17. Do you fill out the medical history form, including a list of your allergies? .33 .03 .06 -.04 18. Do you keep a calendar or list of medical and other appointments? .20 -.12 .04 .25 19. Do you tell the doctor or nurse what you are feeling?  -.05 .13 -.21 -.09 20. Do you answer questions that are asked by the doctor, nurse or clinic staff? -.08 -.21 .05 .05 21. Do you ask questions of the doctor, nurse or clinic staff (For example: What medications or treatments are best for you)? .05 -.06 -.08 -.07 22. Do you make a list of questions before the doctor’s visit? -.00 .26 -.24 .03 23. Do you request the accommodations & support you need at school or work? -.18 -.04 .01 .22 24. Do you apply for a job or work or vocational services? .15 .09 .16 .27 25. Do you get financial help with school or work? .04 -.04 -.04 .03 26. Do you help plan or prepare meals/food? .03 -.10 .03 .81 27. Do you keep home/room clean or clean-up after meals? -.14 .03 -.03 .75 28. Do you use neighbourhood stores and services (For example: Grocery stores and pharmacy stores)? .15 -.07 .04 .22 29. Do you use community support services (For example: After school programs) or advocacy services (For example: Legal services) when you need them? .03 -.21 .26 -.22  Eigenvalue 7.27 2.62 1.89 1.74 Percentage of variance  25.07 9.05 6.53 5.98   78 The structure of the TRAQ 4.1 from our study was examined using the component loading criteria outlined by Sawicki et al. (2011) (factor loadings > .3 on one component and ≤ .3 on the other component).  Seven items in the current analysis (1, 2, 5, 12, 13, 17, and 23) cross loaded on at least two components and based on the component loading criteria need to be deleted.  Our findings indicate five (3, 4, 6, 7, and 8) of the eight components with eigenvalues greater than one are problematic because when the cross loaded items are removed these components would each contain only two items with substantial loading which is the minimum required for retaining a component with PCA (Zwick & Velicer, 1986).  Consequently, if we relied on Kaiser’s criteria the EFA from the present study would result in a 12-item measure comprised of three components (1, 2, and 5) each with four items.  In our analysis the strongest correlation between components was .28 and a correlation of at least .32 is needed to warrant an oblique rotation (Tabachnick & Fidell, 2007), therefore the oblique rotation was not justified. The present PCA was unable to replicate the original component structure reported by Sawicki et al. (2011). Using the data from the present study a constrained two-component PCA of the TRAQ with oblique rotation was performed to explore if forcing two components would improve our results and better replicate the TRAQ’s original component structure.  This analysis produced components that explained 25.1% and 9.1% of the total variance and cumulatively account for 34.2% of the explained variance.  Using the same component criteria stated above two items (1 and 26) need to be removed from the measure.  Item one cross loaded > .3 on both components and item 26 loaded < .3 on both components.  Of the remaining 27 items, 19 items load on component one (items 10-29) and seven load on component two (items 2-9).  The two components have a correlation of r = -.34, which indicates there is enough overlap in the   79 variance among factors to justify an oblique rotation (Tabachnick and Fidell, 2007).  However, even with this constrained two-component PCA we were unable to adequately reproduce the TRAQ’s original two components.  Table 7 summarizes the results of the constrained two- component PCA of the TRAQ. The findings of the present EFA suggest the TRAQ is a 27 item, not a 29 item, measure. Our first TRAQ component is composed of 19 items (10-25, 27-29) that address the many different facets of self-care currently thought to be necessary for successful self-management of a special health care need and includes communicating with HCP, managing activities of daily living (ADLs) and applying for work.  In comparison, the first component reported by Sawicki and colleagues (2011) was labeled Skills for Chronic Condition Self-Management and contains items 1-16.  Six items (10-16) loaded onto the first components in both analyses.  Component one from the present analysis is much broader than Sawicki et al.’s (2011) component one as it also contains 12 items that were originally part of the second Skills for Self-Advocacy and Communicating with Health Providers component.  The second component from the present PCA consists of seven items (2-9) that specifically focus on managing medications, medical equipment and/or supplies and all originally loaded on Sawicki and colleagues’ (2011) Skills for Chronic Condition Self-Management component.  Unlike our first component that shared items with both of the TRAQ components reported in the initial factor analysis, our second component is markedly different and concentrates only on medical management.       80  Table 7 Factor Loadings for Principal Component Analysis with Forced Two Components and Oblique Rotation of the TRAQ  n = 161 Note. Factor loadings > .30 are in boldface.  Components TRAQ Item 1 2 1. Do you fill a prescription if you need to? .32 -.49 2. Do you know what to do if you are having a bad reaction to your medications? .17 -.46 3. Do you pay or arrange payments for your medications? .29 -.51 4. Do you take medications correctly and on your own? .25 -.72 5. Do you reorder medications before they run out? .15 -.69 6. Do you take care of your medical equipment and supplies? -.13 -.80 7. Do you call the suppliers when there is a problem with the equipment? .12 -.66 8. Do you order medical equipment before they run out? .02 -.80 9. Do you arrange payments for the medical equipment and supplies? .15 -.68 10. Do you call the doctor’s office to make an appointment? .65 -.09 11. Do you follow-up on any referral for tests or check-ups or labs? .53 -.12 12. Do you arrange for your ride to medical appointments? .50 -.20 13. Do you call the doctor about unusual changes in your health (For example: Allergic reactions)? .54 -.10 14. Do you apply for health insurance if you lose your current coverage? .48 -.16 15. Do you know what your health insurance covers?  .45 -.19 16. Do you manage your money & budget household expenses (For example: Use checking/debit card)? .52 .07 17. Do you fill out the medical history form, including a list of your allergies? .74 .11 18. Do you keep a calendar or list of medical and other appointments? .54 -.01 19. Do you tell the doctor or nurse what you are feeling?  .42 -.05 20. Do you answer questions that are asked by the doctor, nurse or clinic staff? .37 -.13 21. Do you ask questions of the doctor, nurse or clinic staff (For example: What medications or treatments are best for you)? .58 -.10 22. Do you make a list of questions before the doctor’s visit? .57 .13 23. Do you request the accommodations & support you need at school or work? .46 .04 24. Do you apply for a job or work or vocational services? .61 .20 25. Do you get financial help with school or work? .39 -.04 26. Do you help plan or prepare meals/food? .22 -.05 27. Do you keep home/room clean or clean-up after meals? .33 .06 28. Do you use neighbourhood stores and services (For example: Grocery stores and pharmacy stores)? .45 -.06 29. Do you use community support services (For example: After school programs) or advocacy services (For example: Legal services) when you need them? .41 .04  Eigenvalue 7.27 2.62 Percentage of variance 25.07 9.05   81 Our hypothesis regarding the four TRAQ items relating to health insurance and payment for medications/supplies was not supported.  In the present PCA the health insurance items (14 and 15) both loaded onto the first component that encompasses self-care skills and management of a SHCN.  Items three and nine about payment for medications and medical supplies both loaded onto the second component which specifically focused on managing medications and supplies.  However, based on an examination of the item response distributions it is evident that these items may not be appropriate for a Canadian sample.  Fifty-seven percent of our participants reported they did not now how to pay or arrange payment for their medications, 71% did not know what their health insurance covers and 74% did not know how to apply for health insurance if their coverage was lost.  In a country that has universal health care assessing youths’ knowledge about health insurance and arranging medical payments could be irrelevant and unrelated to their readiness to transition.  Using the data from the present study the relationship between individual TRAQ items and age were examined.  Fifteen of the 29 items had significant relationships with age although these correlations were only weak to moderate, Spearman’s rho ranged from .14 to .49, p < .05. Item 24 – Do you apply for a job or work or vocational services? – had the strongest correlation (rs = .49, p > .01) and Item 6 regarding taking care of medical equipment and supplies had the poorest association with age (rs = .00, p > .05).   These results suggest for the current younger, Canadian sample just over half the TRAQ items were significantly related to age.   We were unable to replicate the component structure from the original TRAQ validation study with either an unconstrained or constrained two-component approach with this younger adolescent sample.  Given the inability to reproduce the component structure, the results of this psychometric evaluation raises questions about whether the TRAQ is a suitable measure to   82 evaluate readiness to transition among younger adolescents.  As a result, the TRAQ could not be used as a gold standard with which to compare the newly developed ON TRAC measure as originally intended. ON TRAC Questionnaire The Am I ON TRAC for Adult Care questionnaire was conceptualized as a combination of knowledge, skills and behaviours that was developed to have two independent scores: the knowledge scale and the behaviour index.  The knowledge scale has 17 items that asks adolescents to rate their level of knowledge or ability to perform tasks (strongly disagree to strongly agree) related to both their SHCN as well as more general medical self-management statements.  We anticipated that the knowledge items would be related to one another and together have the potential to provide HCP with salient information regarding adolescents’ readiness to transition from paediatric to adult medical care.  In contrast, the nine items on the behaviour index were not conceptualized as theoretically connected elements.  These items assess the frequency (never to always) that youth engage in behaviours that are conceptually related to the developmental model of transition such as communicating with their health care providers and participating in activities to stay healthy.  However, since many elements can motivate individuals to engage in different behaviours (friends, family, living situation, socioeconomic status etc.), we decided the behaviour items would be more useful as an index that measures consistent patterns of behaviours. ON TRAC Knowledge Scale Structure An exploratory factor analysis (EFA), specifically principal component analysis (PCA), of the knowledge items was conducted.  It was hypothesized that the knowledge scale could either be unidimensional or have two subscales.  If it were composed of subscales it was predicted that   83 the first subscale would be composed of the nine health condition specific items (1, 4, 8, 9, 10, 15, 16, and 17) and the second subscale would consist of the remaining eight more general health related self-management items (2, 3, 5, 6, 7, 11, 13, and 14).  Additionally, if there were subscales it was hypothesized they would be correlated and an oblique rotation would be appropriate. Initially a PCA without rotation was conducted on all 17 knowledge items.  The inter- item correlation matrix was examined; all items correlated > .3 with at least one other variable and there were no loadings ≥ .9.  However, four items (2, 3, 14, and 16) correlated > .3 with only one other item and two additional items (15 and 17) loaded > .3 with two other items suggesting these six items could be potential problems (Field, 2009).  The KMO verified the sampling adequacy for the analysis, KMO = .84 and Barlett’s test of sphericity, χ2 (136) = 822.54, p < .001, indicated that the correlations between items were sufficiently large for PCA (Field, 2009). Both a parallel analysis and the scree test were examined to ensure accurate extraction of components.  The scree plot showed a clear inflexion after the first component suggesting a one component solution and the parallel analysis (see Table 8) also supported a one component solution.  The first component accounted for 30.3% of the explained variance, had item loadings ranging from .37 to .67 and sixteen items with loadings > .4.  There was a modest amount of cross loading with five items (2, 3, 11, 15 and 16) loadings > .4 on more than one component and five items (2, 3, 11, 14, and 15) loading highest on component 2, 3 or 4. Table 9 provides the component loadings, eigenvalues and percentage of variance for the 17-item PCA without rotation.     84 Table 8 Parallel Analysis 17-items and n = 179 Component 95 th Percentile Eigenvalue Obtained Eigenvalue 1 1.71 5.14 2 1.55 1.47 Note: 95th percentile eigenvalue obtained from parallel analysis engine (Patil et al., 2007)  Table 9 Component Loadings, Eigenvalues, and Percentage of Variance for Principal Component Analysis with No Rotation of the 17-item ON TRAC Knowledge Scale                                  n = 179 Note. Component loadings > .40 are in boldface. Only components supported by the parallel analysis are shown.   Component ON TRAC Knowledge Item 1 1. I can describe my health condition to others .55 2. I have a family doctor .45 3. I visit my family doctor when I need to (For example: to have check-ups, get birth control, or if I have the flu) .44 4. I know what my long-term health problems might be .61 5. I know what patient confidentiality means .61 6. I understand the risks and benefits of health care treatments before consenting to those treatments .65 7. I know how to get my medical records .63 8. I know the names of my medications .67 9. I know what each of my medications are for .65 10. I know the side effects of the medications I take .59 11. I can get to my clinic appointments on my own .48 12. I know how my condition might affect my sexual health .57 13. I know how to prevent sexual health risks such as pregnancy and sexually transmitted infections (STIs) .56 14. My family supports me in managing my health .37 15. I know how my health condition might limit my career choices .48 16. I know how my health condition affects my physical activities .46 17. I know how alcohol, drugs, and tobacco can affect my medications .47  Eigenvalue 5.14 Percentage of variance  30.25   85 Since it was hypothesized that the knowledge scale could potentially be composed of two correlated components an oblique rotation was performed to explore this possibility.  The rotation did not provide clear and distinct components.  Five items (1, 8, 9, 10, and 14) loaded highest on the first component with loadings > .4 and two items (6 and 17) loaded > .3 on component one but also loaded > .4 on other components.  The second component had three items (5, 7, and 13) that loaded exclusively on it and an additional two items (12 and 14) that loaded > .3 but loaded > .4 on components one and four.  Component three was composed of only two items (2 and 3) that both inquired about family doctors.  Two items with substantial loadings is the minimal requirement for component extraction using PCA (Zwick & Velicer, 1986) and the literature warns researchers that retaining components with only two items is risky (Tabachnick & Fidell, 2007).  Three items (4, 15, and 16) loaded uniquely on component four and other two items (12 and 17) loaded > .3 but cross loaded on other components.  Components one and four had the strongest correlation with r = .34 in the component correlation matrix.  The remaining component correlations were less than .32 and did not support an oblique rotation (Tabachnick & Fidell, 2007).  Table 10 provides the component loadings, eigenvalues and percentage of variance for the 17-item PCA with oblique rotation.          86 Table 10 Component Loadings, Eigenvalues, and Percentage of Variance for Principal Component Analysis with Oblique Rotation of the 17-item ON TRAC Knowledge Scale   Component ON TRAC item 1 2 3 4 1. I can describe my health condition to others .64 -.00 -.07 .15 2. I have a family doctor .01 -.04 .85 .01 3. I visit my family doctor when I need to (For example: to have check-ups, get birth control, or if I have the flu) -.12 -.15 .85 .03 4. I know what my long-term health problems might be 1 -.24 .16 .43 5. I know what patient confidentiality means .19 -.67 .08 -.06 6. I understand the risks and benefits of health care treatments before consenting to those treatments .41 -.42 .05 .04 7. I know how to get my medical records .15 -.64 .11 .04 8. I know the names of my medications .77 -.17 -.01 -.04 9. I know what each of my medications are for .64 -.25 -.04 .01 10. I know the side effects of the medications I take .44 -.26 .01 .11 11. I can get to my clinic appointments on my own -.07 -.73 .10 -.04 12. I know how my condition might affect my sexual health -.12 -.49 -.05 .53 13. I know how to prevent sexual health risks such as pregnancy and sexually transmitted infections (STIs) .15 -.46 .03 .16 14. My family supports me in managing my health .64 .30 .26 -.04 15. I know how my health condition might limit my career choices -.10 -.08 -.06 .81 16. I know how my health condition affects my physical activities .14 .27 .21 .68 17. I know how alcohol, drugs, and tobacco can affect my medications .32 .08 -.01 .43  Eigenvalue 5.14 1.47 1.29 1.18 Percentage of variance  30.25 8.67 7.60 6.96 n = 179 Note. Component loadings > .40 are in boldface. All components with eigenvalues  > 1 are displayed. Based on the parallel analysis, scree plots and a comparison of the non-rotated and obliquely rotated component structure it was determined that the non-rotated single component solution was optimal.  Moreover, a unidimensional scale also fits better with the theoretical development of the ON TRAC.  Upon further review of the non-rotated solution it was decided that items 2, 3, and 14 would be removed from the scored knowledge measure.  All of these items correlated > .3 with only one other item in the inter-item correlation matrix and were identified as potential problems.  As discussed previously at least two items are required to   87 warrant retaining a component and even then retaining a component with only two items is risky. Items 2 and 3, addressing family doctors, solely form component three in both the non-rotated and oblique solutions.  Although these items load > .4 on the non-rotated PCA they do not assess either adolescents’ knowledge or capabilities.  In fact, item 3 – I visit my family doctor when I need to (For example: to have check-ups, get birth control, or if I have the flu) – is inquiring about a behaviour and belongs on the behaviour index.  Item 14 – My family supports me in managing my health – is the only item that loaded < .4 on the first component on the non-rotated solution and again does not evaluate youth’s knowledge or capabilities.  Even though items 2 and 14 do not conceptually fit with what the knowledge scale is measuring these items are still pertinent to transition readiness.  These items are important topics for HCP to discuss with youth because having a family doctor and a supportive family can both directly impact adolescents’ abilities to manage their SHCN.  As a result these two items (2 and 14) were not included in the scoring of the knowledge scale and item 3 was removed and will be included on the behaviour index. A second PCA was conducted on a 14-item knowledge scale without rotation.  The inter- item correlation matrix was examined; all items correlated > .3 with at least one other variable and there were no loadings ≥ .9.  Two items (15 and 16) correlated > .3 with only one other variable and item 17 correlated > .3 with only two other variables.  The KMO verified the sampling adequacy for the analysis, KMO = .85 and Barlett’s test of sphericity χ2 (91) = 676.78, p < .001, indicated that correlations between items were sufficiently large for PCA.  Both the scree test and a parallel analysis (see Table 11) (14 items, n=180) signified a one component solution was optimal.  Component one accounted for 33.66% of the explained variance.  The item loadings ranged from .44 to .66.  There was some cross loading with four items (5, 11, 13,   88 and 15) loading > .4 and five items (7, 8, 9, 12, and 17) loading > .3 on more than one component.  Table 12 summarizes the results of the non-rotated PCA of the 14 ON TRAC knowledge items. Table 11 Parallel Analysis for 14-items and n = 180 Component 95 th Percentile Eigenvalue Obtained Eigenvalue 1 1.61 4.71 2 1.46 1.34 Note: 95th percentile eigenvalue obtained from parallel analysis engine (Patil et al., 2007)  Table 12  Component Loadings, Eigenvalues, and Percentage of Variance for Principal Component Analysis with No Rotation of the 14-item ON TRAC Knowledge Scale                         Note. Component loadings > .40 are in boldface. Items 2, 3, and 14 removed. Only components supported by the parallel analysis are shown.  Component ON TRAC item 1 1. I can describe my health condition to others .56 4. I know what my long-term health problems might be .60 5. I know what patient confidentiality means .62 6. I understand the risks and benefits of health care treatments before consenting to those treatments .66 7. I know how to get my medical records .65 8. I know the names of my medications .66 9. I know what each of my medications are for .65 10. I know the side effects of the medications I take .59 11. I can get to my clinic appointments on my own .50 12. I know how my condition might affect my sexual health .60 13. I know how to prevent sexual health risks such as pregnancy and sexually transmitted infections (STIs) .57 15. I know how my health condition might limit my career choices .50 16. I know how my health condition affects my physical activities .44 17. I know how alcohol, drugs, and tobacco can affect my medications .46  Eigenvalue 4.71 Percentage of variance  33.66   89 The internal consistency reliabilities were run for the 17 and 14-item unidimensional scales.  Cronbach’s α for the 17 and 14-item scales were .85 and .84 respectively.  For both the 17 and 14-item versions the overall Cronbach’s α are larger than alpha’s obtained if any of the items are deleted which demonstrates that no items need to be removed.  Split-half reliabilities were examined and the 17-item scale had a slightly stronger correlation between forms r = .70 and Spearman-Brown Coefficient .82 than the 14-item scale (r = .69 and Spearman-Brown Coefficient = .81).  Overall these two solutions had very similar reliabilities.  However, based on the previous discussion regarding removing items 2 and 14 from the scoring of the knowledge scale, but still including them in the measure, and moving item 3 to the behaviour index it was decided that the 14-item scale provided the optimal solution and fits best with the theoretical development of the ON TRAC measure. ON TRAC Knowledge Scale Scores  The ON TRAC knowledge scale has a theoretical range of 14 to 56 and is scored by summing the 14 item responses (strongly disagree=1, disagree=2, agree=3, strongly agree=4). Study scores ranged from 23 to 56 with a mean of 41.65 (SD=6.44).  Three items were negatively skewed with a skewness > 1, eight were slightly negatively skewed and three were slightly positively skewed.  The distribution was not expected to be normal as it was anticipated that youth would be knowledgeable about their condition and a slightly negative distribution is reasonable.  Items regarding medications (8, 9, 10, and 17) had the most missing responses because ten participants were not taking medications.  In the future it would be beneficial to include a ‘not applicable’ response option for these items.  Otherwise, missing responses for items ranged from one to four. Table 13 provides the descriptive statistics for the ON TRAC knowledge scale.   90 Table 13  Descriptive Statistics for ON TRAC Knowledge Scale                     Note: Items 2 and 3 are not included in the Knowledge score and item 14 has been moved to the Behaviour Index         Item/ Subscale n M SD Min. Max. Theoretic al Range Skewness Std. Error of Skewness Kurtosis Std. Error of Kurtosis 1 199 3.31 .65 1 4 1 – 4  -.64 .17 .39 .34 4 197 3.05 .77 1 4 1 – 4  -.62 .17 .23 .35 5 198 3.16 .91 1 4 1 – 4  -.72 .17 -.55 .34 6 198 3.00 .63 1 4 1 – 4  -.25 .17 .36 .34 7 196 2.15 .74 1 4 1 – 4  .51 .17 .30 .35 8 190 3.43 .68 1 4 1 – 4  -.88 .18 .00 .35 9 190 3.42 .70 1 4 1 – 4  -1.16 .18 1.41 .35 10 188 2.89 .78 1 4 1 – 4  -.36 .18 -.20 .35 11 199 2.32 .99 1 4 1 – 4  .30 .17 -.91 .34 12 197 2.24 .85 1 4 1 – 4  .33 .17 -.44 .35 13 197 3.25 .94 1 4 1 – 4  -1.20 .17 .52 .35 15 198 2.82 .01 1 4 1 – 4  -.28 .17 -.77 .34 16 199 3.30 .78 1 4 1 – 4  -1.21 .17 1.50 .34 17 194 3.11 .85 1 4 1 – 4  -.67 .18 -.22 .35 Knowledge Score 180 41.65 6.44 23 56 14 – 56  -.06 .18 -.30 .36   91 As hypothesized the mean, minimum and maximum knowledge scores differed based on participant age with scores slightly increasing with age.  Twelve year olds had the lowest mean (M = 36.44, SD = 4.41) and maximum scores and 18 year olds had the highest mean score (M = 46.37, SD = 6.33) and second highest maximum score.  A deviation from this general trend was the 19 year olds who had the third lowest mean (M = 39.50, SD = 3.11) and second lowest maximum score.  However, it was anticipated that the 19 year olds would have lower scores since they were still patients at BCCH.  It is hospital policy to transfer patients to adult care at 18 and it is reasonable to assume that the 19 year old participants lacked the knowledge and/or skills required for adult medical care which is why they remain at Children’s.  Scores between the different clinics were comparable with means ranging from 40.7 for cardiology to 43.6 for diabetes.  The cardiology clinic had the most missing responses, which was due to ten patients not taking medications.  Table 14 provides the descriptive statistics for ON TRAC knowledge scores by clinic and Figures 4 and 5 display the mean ON TRAC knowledge scores based on age and ON TRAC stage. Table 14  Descriptive Statistics for ON TRAC Knowledge Scores by Clinic   Clinic n Missing M SD Min. Max. Diabetes 68 4 43.63 6.53 29 56 Cardiology 43 11 40.65 7.42 23 53 Gastroenterology 52 2 40.10 4.77 31 49 Neurology 17 3 41.00 6.35 29 51   92 Figure 4 Mean ON TRAC Knowledge Scores by Age  Figure 5 Mean ON TRAC Knowledge Scores by ON TRAC Stage    93 Each ON TRAC knowledge item was categorized according to what stage (early, middle, late) the ON TRAC model indicates youth should possess the knowledge or skill the statement addressed.  Knowledge scale responses were differentiated by ON TRAC stage and the item response frequencies were examined.  When 75% of respondents in a stage agreed or strongly agreed to a knowledge question it was determined that achievement was in the appropriate developmental stage for youth with SHCN.  Five knowledge items (1, 6, 8, 13, & 16) had 75% of participants possessing the knowledge and skill corresponding with its’ designated ON TRAC stage and additional two items (4 & 9) were achieved earlier than the model suggested.  For the issues addressed by these seven items the ON TRAC model appears to be developmentally appropriate for adolescents with SHCN.  Item 17 that enquires about knowledge related to the affect of drugs, alcohol, and tobacco on medications was achieved at by 75% of participants in the early and late ON TRAC stages but not in the middle stage where the model indicates adolescents should possess this knowledge.  This result indicates that the younger participants may not fully comprehend the statement.  Item 5 about understanding patient confidentiality was achieved by participants in the late stage, not the middle as anticipated by the model.  The other five items (7, 10, 11, 12, & 15) had less than 75% of youth in all three stages agreeing with them suggesting that these items maybe not be developmentally appropriate, HCP are not discussing these topics with patients, or that they may be unrealistic expectations for successful transition to adult care. Upon further review of the unattained times, two knowledge items (10 and 11) were recognized as possibly being problematic for the study sample.  Item 10 asked respondents if they knew the side effects of the medications they take and it was not clear if diabetes clinic participants interpreted this item the same way as participants from the other clinics.  Even   94 though insulin is a medication and there are dangers related to taking too much or too little insulin, knowledge of potential side effects for insulin are not comparable to side effects from cardiac, epileptic or inflammatory bowel disorder medications and may not be relevant for the diabetic sample.  Response frequencies for item 10 were examined by clinic and 82% of diabetes clinic participants at least agreed with knowing the side effects of their medications compared to 54%, 67%, and 60% of cardiology, neurology, and gastroenterology participants respectively. Unfortunately, this does not provide insight into why youth did not possess the required knowledge of medication side effects by the middle stage as outlined by ON TRAC’s model of normal adolescent development.  Instead, this suggests that HCP may not be adequately communicating to youth the potential side effects of their medications.  Item 11 asked participants if they were able to get to their clinic appointments on their own.  BCCH provides paediatric outpatient clinics for a large geographical area and when adolescents transition to adult care it may be possible for them to attend a clinic closer to home.  During the data collection process the researcher noted a few participants (3%) came from long distances including: Vancouver Island, Prince George, and the Kootaneys.  Participants were not asked to provide information about where they lived, therefore, it was not feasible to determine the number of participants who traveled long distances to attend appointments.  As a result, we were unable to investigate if responses to item 11 truly reflect participants’ abilities to get to clinic appointments independently or if responses were distorted due to a few patients travelling long distances to attend clinics at BCCH. ON TRAC Behaviour Index  The behaviour index is composed of nine items that were conceptually related to the developmental model of transition.  The items require youth to rate the frequency, from never to   95 always, that they engage in particular behaviours.  The ON TRAC questionnaire was created with a holistic approach to evaluate adolescents’ overall health and wellbeing as well as their specific knowledge, skills and behaviours related to SHCN.  As a result, four items (18, 23, 24, and 25) explicitly address behaviours identified as salient for successful self-management of SHCN in the adult medical system and the remaining items focus on general health related behaviours.  One item, I feel my family cares about me, was removed from the index because it did not assess a behaviour.  Due to the complex nature of behaviour theory it was determined that these eight items did not have a latent or manifest construct thus, they would not form a scale and factor analysis was inappropriate.  Instead, an index that measures consistent patterns of behaviour was established with the items hypothesized as skill sets that youth must achieve at a minimum level in order to be considered ready to transition to adult care and self-manage their condition.  Since there are no empirically derived cut-off scores for determining the minimal level of behaviours required for successful transition the investigator created theorized cut-offs.  Items 23 to 25 required a minimum frequency of ‘often’ because these items directly influence adolescents’ self-management of their SHCN.  Items 20, 21 and 22 are general health and wellness statements that are more indirectly related to self-care management and needed a minimum response of ‘sometimes’.  In the transition literature youth meeting with HCP on their own is presented as a crucial skill, however, during data collection it was evident that very few youth actually do this. As a result, the cut-off for item 18 (I meet with my health care providers on my own) was placed at ‘sometimes’.  Finally, item 26, which asks about thinking beyond high school, required a minimum response of ‘often’ since transitioning to adult health care is another step towards adulthood and it is important that adolescents are thinking and preparing for their future at this   96 time.  Responses below the designated cut-off responses received a zero and responses at or above the cut-off were counted as one.  The cut-offs represent what the investigator deemed the minimum frequency adolescents’ need to practice the behaviours to be considered ready to transition to adult care.  The highest achievable cut-off score is eight and scores of seven and eight are needed for adolescents to meet the minimal level and be considered ready to transition. ON TRAC Behaviour Index Scores   The behaviour index has both a raw score and a cut-off score.  The raw score has a theoretical range of 8 to 40 and is created by summing the eight item responses (never=1, rarely=2, sometimes=3, often=4, always=5).  The raw scores ranged from 16 to 40 and had a mean of 27.75 (SD=4.71). Item 18 was positively skewed, item 24 was negatively skewed and the remaining six were slightly negatively skewed.  A normal distribution was not expected, as many youth should be consistently practicing many of these behaviours.  Item 24 asked about medications and had 11 missing responses.  A ‘not applicable’ response option should be added for this item.  Missing responses ranged from zero to two for the other seven items. The Behaviour cut-off score has a theoretical range of zero to eight and is formed by converting the item responses to either a zero or one depending on whether the minimum cut-off for each item has been met.   The cut-off scores ranged from one to eight, had a mean score of 5.19 (SD=1.58) and were very slightly negatively skewed.  Table 15 provides the descriptive statistics for the ON TRAC behaviour index.     97 Table 15  Descriptive Statistics for ON TRAC Behaviour Index          Note: Item 19 was removed from the index  Item/ Score n M SD Min. Max. Theoretical Range Skewness Std. Error of Skewnes s Kurtosis Std. Error of Kurtosis 18 200 1.90 1.02 1 5 1 – 5  1.03 .17 .11 .34 20 200 3.92 1.07 1 5 1 – 5  -.63 .17 -.49 .34 21 200 3.19 1.21 1 5 1 – 5  -.19 .17 -.83 .34 22 200 3.99 .99 2 5 1 – 5  -.45 .17 -.1.02 .34 23 200 3.23 1.07 1 5 1 – 5  -.09 .17 -.44 .34 24 189 4.48 .82 1 5 1 – 5  -1.79 .18 3.46 .35 25 198 2.99 1.37 1 5 1 – 5  -.04 .17 -1.19 .34 26 200 4.10 .94 1 5 1 – 5  -.79 .17 -.17 .34 Raw Score 189 27.75 4.71 16 40 8 – 40 .03 .18 -.14 .35 Cut-off Score 189 5.19 1.58 1 8 1 – 8 -.24 .18 -.38 .35      98 The behaviour index raw score means increased slightly with age ranging from 25.8 to 31.1.  Again the 19 year old participants did not fit the general trend of the data.  This group had a lower mean score than both the 17 and 18 year olds, had the highest minimum and the lowest maximum scores.  There were only four participants who were 19, which affects the mean score and makes it difficult to draw meaningful conclusions from such a small sample.  As discussed in the knowledge scale section these adolescents continue to be patients at BCCH even though they should have transitioned to adult care at 18 suggesting they are not ready for the adult system.  Excluding the 19 year olds there was not a lot of difference between the age groups. The minimum scores range from 18 to 21 and maximum scores range from 34 to 40.  The largest increase in the maximum score occurs between 15 and 16 year olds.  Table 16 summarizes the behaviour index raw scores by age and ON TRAC stage. Table 16  Descriptive Statistics for ON TRAC Behaviour Index Raw Scores by Age & ON TRAC Stage              The behaviour index cut-off score means also increased slightly with age ranging from 4.41 to 6.35.  The percentage of adolescents who met the theorized cut-off score of seven and  n Missing M SD Min. Max. Age 12 17 2 26.18 5.48 18 35 13 23 0 25.78 4.38 16 34 14 20 1 26.25 4.66 17 34 15 38 0 27.34 4.08 20 34 16 35 5 28.00 4.43 19 40 17 32 1 28.94 4.68 21 40 18 20 1 31.10 4.64 21 39 19 4 1 28.50 3.00 27 33 Stage Early 17 2 26.18 5.48 17 35 Middle 81 1 26.63 4.31 16 34 Late 91 8 29.03 4.60 19 40      99 satisfied the minimal requirements for successful transition increased steadily after 14 years of age.  The 12 to 14 year olds had similar percentages of acceptable cut-off scores and then percentage nearly doubled at 15 and continued to increase until 18.  The greatest increase in the number youth deemed ready for transition occurred between 17 and 18 with 27% of 17 year olds and 62% of 18 year olds obtaining cut-off scores of seven or eight.  These results suggest that there is not a set age at which adolescents consistently perform the behaviours necessary for successful transition and some youth may potentially be ready for adult care at 15 and others may take longer to satisfy the minimum requirements.  Figure 6 displays the percentage of youth who meet the behaviour cut-off score at each age and Table 17 summarizes the descriptive statistics for the behaviour index scores by clinic. Figure 6 Percentage of Youth that Met the Behaviour Index Cut-off Score > 7      0 10 20 30 40 50 60 70 80 90 100 19 18 17 16 15 14 13 12 Percentage of Participants that Met Cut-off Score A ge       100 Table 17 Descriptive Statistics for ON TRAC Behaviour Index Scores by Clinic       Similar to the knowledge items each ON TRAC behaviour index item was categorized according to what ON TRAC stage (early, middle, late) the model expects adolescents to be consistently engaging in each particular health related behaviour.  When 75% of participants in stage met the theorized cut-off frequency for a behaviour statement that behaviour was deemed developmentally appropriate for adolescents with SHCN.  Half of the behaviour items (20, 22, 24, and 26) were achieved by 75% of respondents at the stage specified in the ON TRAC model indicating that they are developmentally appropriate.  The cut-off requirements were not met by 75% of youth in any stage for the remaining four behaviour items (18, 21, 23, and 25) and indicate that these behaviours may be unrealistic expectations for transition, they may not be developmentally appropriate for this sample, or they HCP may not be addressing these behaviours with adolescents. ON TRAC Model The nine ON TRAC questionnaire items (five knowledge and four behaviour) that were not achieved by 75% of participants at any stage as well as the two knowledge items attained later than indicated by the ON TRAC model were investigated further to assess if the model’s Clinic n Missing M SD Min. Max. Raw Score Diabetes 72 0 27.78 4.85 17 40 Cardiology 43 11 27.95 5.06 16 39 Gastroenterology 54 0 27.63 4.42 18 40 Neurology 20 0 27.50 4.50 19 35 Cut-off Score Diabetes 72 0 5.06 1.69 1 8 Cardiology 43 11 5.42 1.67 2 8 Gastroenterology 54 0 5.17 1.42 2 8 Neurology 20 0 5.25 1.58 3 8      101 normative stages of adolescent developmental are appropriate for youth with SHCN.  Responses were examined by age starting at 13 years.  The 12 year olds were excluded because none of the unattained items were designated as achievable in early transition.  The four 19 year old participants were also excluded because their responses are not reliable and were not representative of the larger population due to the small sample size.  None of these 11 items were achieved sufficiently by age 13 or 14 years.  Item 5 about understanding patient confidentiality and item 21 regarding talking to friends about problems or worries were attained by three- quarters of the 15 year olds.  At age 16 items 5 and 17 (knowing medication interactions with drugs, alcohol and tobacco) but not 21 were attained by more than 75% of respondents.   Five items were adequately attained by both the 17 and 18 year old participants and four were common to both ages: meaning of patient confidentiality, knowing the side effects of their mediations, knowing the potential effects of drugs, alcohol and tobacco on their medications, and talking to friends about problems or worries.  Interestingly, 84.8% of 17 year olds but only 66.7% of 18 year olds agreed/strongly agreed to item 15, I know how my health condition might limit my career choices.  Lastly, by age 18 the majority of participants agreed/strongly agreed that they can get to their clinic appointments independently while only 64% of 17 year olds reported being able to do so. While the percentage of participants achieving the other five items generally increased with age they were not sufficiently attained by age 18 indicating these items may not be realistic expectations of adolescents before they transition to adult care.  These items include: I know how to get my medical records, I know how my condition might affect my sexual health, I meet with my health care providers on my own, I ask health care providers questions about my health at my visits, and when my symptoms are getting worse I contact the clinic for help.      102 Statistical Analysis  The data was examined to determine if the assumptions of normality were met.  Sample distributions, histograms, p-plots, skewness and kurtosis were assessed.  Skewness and kurtosis for the ON TRAC and PMI scores were both less than one signifying that the data is approximately normal.  The Kolmogorov-Smirnov test for the ON TRAC knowledge scale, D(162) = .06, p = .20, and the PMI, D(162) = .06, p = .20 were both non-significant indicating these scores are approximately normally distributed but, the K-S test for both the ON TRAC behaviour index raw score, D(162) =.08, p = .02 and the cut-off score, D(162) = .16, p = .00 were significantly non-normal.  However, a limitation of the K-S test is that it’s more likely to produce significant findings with larger samples (Field, 2009).  Homogeneity of variance was assessed with the Levene’s test.  Variances were equal for male and female participants for all the measures: knowledge scale, F(1, 160) = .47, ns, behaviour index raw score, F(1, 160) = .09, ns, behaviour index cut-off score, F(1, 160) = .69, ns, and PMI, F(1, 160) = .99, ns.  Variances of the PMI scores were significantly different for the four clinics, F(3, 158) = 3.19, p < .05, but were equal for the other three instruments knowledge scale, F(3, 158) = 2.42, ns, behaviour index raw score, F(3, 158) = .528, ns, and behaviour index cut-off score, F(1, 158) = .72, ns.  The variance was not significantly different based on age for any of the instruments: knowledge scale, F(7, 154) = 1.32, ns, behaviour index raw score, F(7, 154) = .43, ns, behaviour index cut-off score, F(7, 154) = .58, ns, and PMI, F(7, 154) = 1.16, ns. The study sample is fairly large (n=161) and based on the evaluation of all the results and taking into consideration the central limit theorem the data appears to be approximately normal.         103 Relationships Between the Variables  Correlations.  The relationships between age, psychosocial maturity, ON TRAC stage, years with condition and the three ON TRAC scores were examined and table 18 provides the results.  The correlation between psychosocial maturity and age was significant but not very strong, r = .16, p < .05, which suggests that age is a very loose proxy for development and maturity, and for youth with SHCN age may have poorer relationship with maturity.  This weak relationship was also evident when the PMI scores were examined by age.  Age had a significant moderate relationship with the knowledge scale, r = .43, p < .01, and both behaviour indexes (raw score, r = .30, p < .01, and cut-off score, r = .34, p < .01).  The PMI also had significant moderate correlations with the ON TRAC scales: knowledge r = 30, p < .01, behaviour raw score and cut- off score, r = .39, p < .01.  Age had a stronger relationship with knowledge but psychosocial maturity had a stronger relationship with behaviour.  The strongest relationships were found between the ON TRAC scales.  There was a strong significant correlation between the knowledge scale and the behaviour index raw score, r = .53, p < .01, as well as with the cut-off score, r = .49, p < .01.  These results verify the relationship between health care knowledge and self-care behaviours.  All correlations between ON TRAC scores and years with condition were non-significant and did not support the hypothesis that participants’ readiness assessment scores would differ based on the number of years adolescents had their condition.              104 Table 18 Correlations between Age, PMI, ON TRAC Knowledge Score, ON TRAC Behaviour Index Raw Score, ON TRAC Behaviour Index Cut-off Score, ON TRAC Stage and Years with Condition   Age PMI  Knowledge Scale Behaviour Index Raw Score Behaviour Index Cut- off Score ON TRAC Stage Years with Condition Age 1 PMI  .16*   1 Knowledge Scale .43** .30** 1 Behaviour Index Raw Score .30** .39** .53** 1  Behaviour Index Cut-off Score .34** .39** .49** .91** 1  ON TRAC Stage .89** .12 .41** .25** .29** 1 Years with Condition .35** .14 .13 .05 .06 .26** 1 *p < .05, **p < .01   Comparing Means.  A one-way ANOVA was used to assess if the ON TRAC and PMI scores were significantly different based on participants’ ON TRAC stage.  A significant difference was found for the three ON TRAC scores.  Table 19 summarizes the results of the ANOVA. Table 19 ANOVA of ON TRAC and PMI Scores Based on Participant ON TRAC Stage Scale Degrees of Freedom F Statistic Significance Knowledge Score 2, 177 18.10 .000 Behaviour Index Raw Score 2, 186 7.04 .001 Behaviour Index Cut-off Score 2, 186 8.52 .000 PMI Score 2, 180 1.51 .224  In order to determine which ON TRAC stages had significantly different ON TRAC scores a series of independent t-Tests were performed.  The knowledge scores had significantly      105 different means between each of the stages: early/middle t (90) = -2.21, p = .03; middle/late t (160) = -4.44, p < .001; early/late t (104) = -5.35, p < .001.  The differences between participants classified as early and middle ON TRAC stages for both behaviour index scores were non- significant.  However, behaviour index scores were significantly different between both middle/late, raw score t (170) = -3.52, p = .001, cut-off score t (170) = -3.51, p = .001, and early/late stages, raw score t (106) = -2.79, p = .03, cut-off score t (106) = -2.93, p = .004.  These results indicate that participants’ knowledge scores significantly increased at each ON TRAC stage and behaviour patterns had the most significant change between the middle and late stages.     An independent-means t-test was conducted to compare ON TRAC and PMI scores between male and female participants.  ON TRAC knowledge scores for boys (M = 41.90, SE = .63) and girls (M = 41.32, SE = .75) were not significantly different t (178) = .60, p = .55, and represented a small-sized effect r = .04.  Similarly, PMI scores for boys (M = 55.11, SE = .72) and girls (M = 54.79, SE = .81) were not significantly different, t (181) = .295, p = .77 and also had a small-sized effect size r = .02.  The hypothesis that PMI scores would differ based on gender was not supported.  A difference between the two groups was found for both behaviour index scores, with girls having a significantly higher mean score than boys.  The behaviour index raw scores for boys (M = 27.02, SE = .45) and girls (M = 28.72, SE = 4.68) had significantly different means, t (187) = -2.49, p = .01 and a small-sized effect r = .17.  The behaviour index cut-off scores for boys (M = 4.99, SE = .16) and girls (M = 5.46, SE = .16) were also significantly different, t (187) = -2.02 with a small-sized effect r = .14.  These results indicate that girls reported consistently practicing more health related self-management behaviours than boys.  After finding no difference in mean PMI scores based on gender it was surprising to find      106 that girls had significantly higher behaviour scores than boys, especially since there was a moderate correlation between PMI and behaviour scores.  A one-way ANOVA compared the mean scores between the four clinics (diabetes, cardiology, gastroenterology, and neurology).  There were no significant differences between clinics for either of the behaviour index scores or the PMI scores.  However, there was a significant difference between the mean knowledge scores, F (3, 176) = 3.72, p = .01.  The diabetes clinic had the highest mean knowledge score (M = 43.63, SE = .79) and the gastroenterology had the lowest mean knowledge score (M = 40.10, SE = .66). Linear Regression  Simple linear regression with the entry method was performed to test the theory that age, psychosocial maturity, gender and years with condition would predict ON TRAC knowledge scores.  Assumptions of multicollinearity, homoscedasticity, and normally distributed residuals were met. The model was significant, F(4, 159) = 12.47, p < .001, and accounted for 23.9% of the variance in knowledge scores.  Age and maturity were significant predictors, however, gender and years with condition did not significantly predict knowledge scores.  Table 20 summarizes the results. Table 20  Linear Regression of the ON TRAC Knowledge Score  Predictor B SE B β Constant 7.82 4.94 Age 1.38 .25 .41* PMI  .23 .06 .26* Gender .38 .91 .03 Years with Condition  -.06 .09 -.05 n = 164. Note: R2 = .24 (p < .001). * p < .001      107  A second simple linear regression with the entry method was conducted to examine the extent to which age, psychosocial maturity, ON TRAC knowledge scores, gender and years with condition relate to adolescents’ level of self-care behaviour as measured by the ON TRAC behaviour index raw scores.  The data was examined to ensure assumptions of multicollinearity, homoscedasticity, and normally distributed residuals were met.  The model explained 42.9% of the variance in behaviour index raw scores and was significant, F(5, 156) = 23.41, p < .001. Years with condition was the only variable that did not significantly contribute to the model. Table 21 provides the results of the regression of the ON TRAC behaviour index raw scores. Table 21  Linear Regression of the ON TRAC Behaviour Index Raw Scores  Predictor B SE B β Constant -3.01 3.31 Age .41 .18 .16* PMI  .22 .04 .32*** ON TRAC Knowledge .29 .05 .39*** Gender 1.86 .60 .19** Years with Condition  -.04 .06 -.04 n=162.  Note: R2 = .43 (p < .001). * p < .05, **p < .01, *** p < .001  Summary The psychometric evaluation of the TRAQ revealed that it may not be an appropriate measure for younger adolescents.  Its’ original component structure could not be replicated with either an unconstrained or constrained two-component PCA.  Consequently, the TRAQ could not be used as the comparison measure for the revised ON TRAC questionnaire as originally planned. The Am I ON TRAC for Adult Care questionnaire is comprised of two separate measures: knowledge scale and behaviour index.  Based on an EFA of the knowledge items it was concluded that a 14-item unidimensional scale was the optimal solution.  Two items (2 and      108 14) were removed from the scoring of the scale but will remain on the measure, as they are pertinent issues that have potential to affect successful transition.  A third item (3) was removed from the knowledge score and will be added to the behaviour index in the future.  The behaviour index has eight items and has both a raw score and a theorized cut-off score.  The theorized cut- off score represents the minimally accepted frequency for engaging in behaviours considered necessary for successful transition to adult care.  Each individual item has a specific cut-off (scored as zero or one) and an overall cut-off score of seven or eight is required for youth to be deemed ready for transition.  Based on item response frequencies seven of the knowledge and four of the behaviour items were achieved by 75% of participants either at or before the developmental stage designated by the ON TRAC model. The relationships between the ON TRAC scores and age, ON TRAC stage, PMI score, years with condition and gender were examined.  Age and psychosocial maturity had a small but significant correlation suggesting that age may not be the best indicator for maturity for this sample.  Age had significant moderate relationship with all the ON TRAC scores but had the strongest correlation with the knowledge score.  In contrast, the PMI had significant moderate correlation with all the ON TRAC scores but had the strongest relationship with the behaviour scores.  Years that participants had the condition did not have a significant relationship with the ON TRAC score or the PMI.  Significant differences were found for ON TRAC scores but not psychosocial maturity scores based on participants’ ON TRAC stage.  Knowledge scores significantly increased with each ON TRAC stage and behaviour index scores had significant differences between early/late and middle/late stages.  No significant difference in mean psychosocial maturity and knowledge scores were found between boys and girls.  However, mean behaviour index scores were significantly different and indicated that girls reported      109 consistently practicing more health related self-management behaviours than boys. Finally, linear regressions revealed that age and PMI score significantly predicted the ON TRAC knowledge score and that age, PMI score, ON TRAC knowledge score and gender all significantly contribute to the ON TRAC behaviour index raw score.                         110 CHAPTER FIVE: DISSCUSSION AND IMPLICATIONS Introduction  This chapter discusses the key study findings and their relevance to the primary research questions.  The purpose of the study was to determine if the normative stages of adolescent development identified in the ON TRAC model (stage and achievements at each age range) fit as the same stage-for-age among adolescents with SHCN and examine if the behaviours and skills believed to be essential for successful transition to adult care are developmentally appropriate expectations for adolescents with SHCN.  After the primary research questions are addressed the study findings are situated within the health care transition (HCT) literature and recommendations for transitional care models and future research are made. Key Study Findings Based on participant responses to the ON TRAC questionnaire the normative stages of adolescent development identified in the ON TRAC model (stage and achievements at each stage) moderately fit the same stage-for-age among adolescents with SHCN.  Participants classified in the early ON TRAC stage had the lowest mean score and adolescents in the late stage had the highest mean score on both the ON TRAC knowledge scale and the behaviour index.  Significant differences between mean knowledge scores were found between each ON TRAC stage (early/middle, middle/late, early/late).  Mean behaviour index scores were significantly different between the middle/late and early/late stages.  This evidence provides reasonable support for the ON TRAC stages being developmentally appropriate for the sample as the results demonstrate that condition specific and general health knowledge increased with each ON TRAC stage.  Furthermore, participants in the late transition stage were consistently      111 engaging in necessary self-management behaviours more often than those in either the early or middle transition stages.  It is important to note that the ON TRAC responses provided by participants in the early stage are not representative of the ON TRAC’s early transition stage and these responses may be inflated.  The ON TRAC model identifies the early stage of transition as youth aged 10 to 12 years, however, 12 year olds were the youngest participants in the present sample.  Consequently, the achievements of respondents in the early transition stage may be over estimated.  There is also evidence to suggest that these younger participants may not have fully understood some questionnaire items.  According to Piaget’s theory of cognitive development, as cited by Wong, Perry and Hockenberry (2002), abstract thinking, adaptability and flexibility are acquired between 11 and 15 years of age during the formal operations period.  It is possible that some of the 12 year olds may not have been able to accurately report on their own level of knowledge because their abstract thinking abilities are not well established and they were unaware of their own knowledge gaps.  Examination of the ON TRAC item response frequencies grouped by participant ON TRAC stage produced more uncertain findings regarding the model’s stage-for-age fit among youth with SHCN.  Three-quarters of respondents achieved 36% of the knowledge and 50% of the behaviour items at the stage designated by the model.  This demonstrated the normative stages of adolescent development corresponded with the same stage-for-age among adolescents with SHCN for the areas addressed by five knowledge and four behaviour items that represent 41% of the questionnaire content.  The knowledge items achieved at the appropriate ON TRAC stage included: I can describe my health condition to others, I understand the risks and benefits of health care treatments before consenting, I know the names of my medications, I know how to      112 prevent sexual health risks, and I know how my health condition affects my physical activities. Participants achieved two knowledge items before the stage specified by the ON TRAC model. Seventy-five percent of youth in the middle stage agreed/strongly agreed that they knew what their long-term health problems might be while the model indicates this knowledge should be attained during late transition.  Similarly, the majority of adolescents in the early ON TRAC stage reported they agreed/strongly agreed to knowing what their medications are for and the ON TRAC model placed this knowledge in the middle stage.  One item, I know what patient confidentiality means, was achieved by participants in the late stage, one stage after the normative developmental model suggests is appropriate.  According to ON TRAC’s normative stages of adolescent developmental participants should possess knowledge related to interactions between their medications and recreational drugs by the middle transition stage.  Seventy-nine and 80% of youth in the early and late ON TRAC stages respectively reported knowing how alcohol, drugs and tobacco can effect their medications, however, only 70% of adolescents in the middle stage agreed/strongly agreed with this item.  This is an example of younger participants not comprehending the question and/or not being aware of their own knowledge gaps.  Finally, four knowledge items were not achieved by three-quarters of the participants at any stage: I know how to get my medical records, I know the side effects of the medications I take, I can get to my clinic appointments on my own, and I know how my condition might affect my sexual health. Half of the behaviours items were achieved by 75% of the respondents during the appropriate ON TRAC stage-for-age and included: I participate in clubs, groups, sports or activities, I take my medications on my own, I participate in activities/exercise to stay healthy, and I think about what I’d like to do after high school.  However, the theorized minimum cut-off      113 scores for the remaining four behaviour items were not attained by the majority of participants at any ON TRAC stage.  Items not sufficiently achieved were: I meet with my health care providers on my own, I talk to my friend(s) about my problems or worries, I ask health care providers questions about my health at my visits, and when my symptoms are getting worse I contact the clinic for help. Overall, of the 22 scored items, 41% were achieved during the designated ON TRAC stage, two items were attained earlier and two later than expected.  The remaining 41% of items were not sufficiently achieved at any stage.  Potential explanations for why these items were not attained include: for these specific issues the normative stages of adolescent development do not fit the same stage-for-age among youth with SHCN, HCP are not adequately discussing these topics with adolescents, or these items represent unrealistic expectations and are not developmentally appropriate for adolescents with and even without SHCN.  The guidelines and recommendations used to construct the ON TRAC questionnaire are based on the set of knowledge, skills, and behaviours HCP believe are essential for youth with SHCN to possess in order to successfully transition to adult medical care but, these assumptions have never been tested.  A salient issue that arose from our findings is: do HCP hold adolescents with SHCN to a higher standard when it comes to health care self-management compared to the general adult population?  As an example, the HCT literature states adolescents should be meeting with HCP independently before transitioning to adult care, however, do many adults attend health care appointments alone and is this practical?  There are benefits to having a second person present at medical appointments, which include providing support for the patient and clarifying misunderstood information.  However, HCP working with adolescents with SHCN and adult providers who take on these patients as young adults may need to re-evaluate and adjust the      114 transition readiness indicators to differentiate between the ideal level of knowledge, skills, and behaviours required for transition and what is realistic and actually needed for successful transfer to adult care. Generally, the ON TRAC model appears to moderately fit the same stage-for-age among adolescents with SHCN.  Nevertheless, based on the analysis of the ON TRAC responses by age and taking into consideration that the youngest participants in the sample were 12 year olds, which is the oldest age of the early ON TRAC transition stage, it may be appropriate to modify the ages for each stage of the model.  Using our findings, I would suggest that the ON TRAC transitions stages should have the following age ranges: early 13 to 15, middle 16 to 17, and late 18 to one year after transfer to adult medical services. Discussion  The HCT literature acknowledges there is a lack of validated measures that assess adolescents’ with SHCN readiness to transition to adult medical care (Fair et al., 2010; Sawicki et al., 2011; Williams et al., 2011).  To address this need Sawicki et al. (2011) developed and validated the Transition Readiness Assessment Questionnaire (TRAQ), a self-administered assessment tool, which measures transition readiness skills.  The TRAQ was originally validated with a diverse group of youth and young adults ranging in age from 16 to 26 years in the United States (Sawicki et al., 2011).  An objective of the present study was to evaluate if the TRAQ is a reasonable measure of transition readiness for younger Canadian adolescents (12 to 19 years) with SHCN. The original validation of the TRAQ reported two subdomains: Skills for Chronic Condition Self-Management and Skills for Self-Advocacy and Communication with Health Providers (Sawicki et al., 2011).  Unfortunately, our psychometric evaluation could not replicate the TRAQ’s original component structure with either an unconstrained or constrained two-      115 component principal component analysis.  Our unconstrained analysis supported a four component solution and using forced component loadings was the only method that enabled us to produce two components and even then the solution was different from the original. Our inability to replicate the TRAQ’s component structure raises questions regarding how the TRAQ was developed, its’ psychometric properties and whether it is an appropriate measure of transition readiness for younger youth and for populations with universal health care. If the TRAQ is to be used with younger, non-American populations further evaluation and additional work on the instrument is needed.  The TRAQ employs the transtheoretical model as its’ framework to identify adolescents’ stage of transition readiness (Sawicki et al., 2011).  The response options measure readiness but the items are mostly asking about behaviours (Do you fill a prescription if you need to? Do you tell the doctor or nurse what you are feeling?) and measuring behaviour is very challenging.  The TRAQ assumes that age and maturity are driving the youths’ behaviours, when in reality the adolescent’s condition and context may in fact be influencing the behaviours and their age, maturity or capability to perform a skill may be irrelevant.  Behaviour theory is exceptionally complex and there are a plethora of factors that potentially affect an individual’s motivation to perform a particular behaviour.  As a result, it may not be pragmatic to assume that all the behaviours thought necessary for successful self- management of a SHCN are driven by the same latent construct or even have a common latent construct.  When the TRAQ was administered to youth 12 to 19 years of age we found that participants had problems understanding the response options.  The six response options each reflect a particular stage of change (i.e. No, I don’t know how = precontemplation; No, but I am learning to do this = preparation) and are not concrete choices.  During data collection the researcher frequently had to clarify the response options to participants of all ages because they      116 had difficulty understanding what the response options meant and how they applied to them. Focus group work and cognitive processing interviewing would provide valuable insights into how younger adolescents interpret the response options and if revisions are required.  Finally, since Canada has a universal health care system a few of the TRAQ items, specifically the items about paying for medications, medical equipment and supplies, applying for health insurance and what health insurance covers, were not as relevant to our sample.  Further psychometric and model reduction testing of the TRAQ are also recommended if researchers and HCP are going to use this instrument with a younger adolescent population. Adolescent development theory is the prevalent theoretical foundation for the ON TRAC model and appropriate developmental care is a salient principle of transitions services (Blum et al., 1993; Rosen et al., 2003).  Hospitals have constructed transition programs with the goal of providing continuous, coordinated, developmentally suitable and psychologically sound care to youth throughout the process of transferring from paediatric to adult medical services (Canadian Paediatric Society, 2011). Despite this, a 2005 study of youth with type 1 diabetes found common transition difficulties across different health care delivery models which led the researchers to question if the problems may be partly due to transitional care not addressing psychosocial maturation issues (Pacaud et al.).  The HCT literature speculates that SHCN during childhood and adolescence may affect the normal maturational process, which potentially results in these youth being less independent than their peers when transition from paediatric to adult care occurs (Williams et al., 2011).  Furthermore, a UK survey of HCP perceptions of transitional care reported that mental maturity was the most commonly suggested criteria for assessing youths’ readiness to transfer to adult care (Por et al., 2003).  To the best of the researchers’ knowledge the present study is the only investigation to examine adolescents’ with      117 SHCN psychosocial maturity, assess the relationships between maturity, age and readiness to transition measures, and evaluate if a transitional care model (ON TRAC) is developmentally appropriate for youth with SHCN.  Our results from the Psychosocial Maturity Inventory (PMI) demonstrated that mean psychosocial maturity scores increased both by age and ON TRAC stage. This indicates that as youth with SHCN get older they are also becoming more psychosocially mature.  The difference between mean PMI scores was small and not significantly different between any of the ON TRAC stages with a mean psychosocial maturity score of 52.53 for 12 year olds and 57.68 for 18 year olds.  The mean score of 57.68 is 14.32 points below the maximum score of 72 and only one participant achieved the maximum PMI Score.  Normative ranges for PMI scores by age are not available and consequently we were unable to compare our sample’s scores to expected scores and assess if adolescents with SHCN mature at the normal rate.  When examining the psychosocial maturity scores it is important to take into consideration that the PMI was developed in the 1970s and since then a new period of development has been proposed.  Arnett (2000) acknowledged that the late teens and early twenties (18-25) have become a period of emerging adulthood.  Until recently, it was normal for young adults, between 18 to 25 years, to be participating in and establishing their long-term adult roles.  In contrast, the new period of emerging adulthood is characterized by: “relative independence from social roles and from normative expectations” (Arnett, 2000, p.469).  During this developmental period individuals often explore the various potential opportunities that are available in all aspects of their lives including love, work, and worldviews (Arnett, 2000).  Arnett (2000) reports that young adults do not perceive themselves as adolescents, but many also do not view themselves entirely as adults either.  This change in normative development may have affected the normative maturational process and HCP may need to incorporate this developmental theory into      118 their transitional care models.  A critical outcome from the current study is that it raises questions regarding whether the knowledge, skills and behaviours that HCP believe youth need to possess to effectively self-manage their SHCN as an adult are developmentally appropriate or even realistic expectations of adolescents.  Re-evaluating the expectations of youth at the time of transition and taking into consideration the emerging adulthood theory may help improve the fit between the ON TRAC stage-for-age model, as well as the overall HCT process. To validate the revised Am I ON TRAC for Adult Care questionnaire we examined the correlations between adolescents’ age, psychosocial maturity, ON TRAC knowledge score, and both ON TRAC behaviour index scores. We found that age and psychosocial maturity had a significant but weak correlation, which suggests age is a very loose proxy for maturity.  Age had the strongest relationship with the knowledge scores but had significant moderate correlations with both the knowledge and behaviour scores.  Other investigations that have assessed readiness have also identified a significant relationship between age and transition related knowledge (McPherson et al., 2009; Sawicki et al., 2011).  We found that psychosocial maturity had the strongest relationship with the ON TRAC behaviour scores but was also significantly and moderately correlated with the ON TRAC knowledge scores.  Additionally, a strong and significant correlation between the knowledge and behaviour scores was found.  These results are consistent with earlier research and provide further evidence to support that chronological age may not be the best indicator of transition readiness. Studies by Cappelli et al. (1989) and Williams et al. (2011) also determined that age was not the most important marker for evaluating readiness for transition.  Cappelli et al. (1989) found the results from their readiness questionnaire for adolescents with cystic fibrosis to be a better predictor of transition readiness than age and Williams et al. (2011) reported that adolescents’ medical self-management was      119 closely associated with their independence in other areas.  Together this research suggests hospitals may need to re-consider having a standard age for determining when transition occurs and look more closely at health and self-management behaviours. Our findings provide unique and valuable insight into the relationships and interactions between age, psychosocial maturity, health and self-care related knowledge and behaviours.  A linear regression found that age, psychosocial maturity, gender and years with condition accounted for only 24% of the variability in ON TRAC knowledge scores.  It is clear that other factors, which have yet to be identified, contribute to adolescents’ health and self-care related knowledge.  Not surprisingly, knowledge and behaviour scores had the strongest relationship among all the variables.  However, the regression model that included age, maturity, knowledge scores, gender and years with condition explained 43% of the variance in ON TRAC behaviour scores with the remaining 57% unaccounted for.  This finding emphasizes that knowledge alone does not influence behaviour, and other elements such as family support, HCP expectations and management style, access to resources and other stressors in adolescents’ lives may be predictors of the consistency with which youth are able to manage their own health care. In their development of the social-ecological model of adolescents and young adults readiness for transition (SMART), Schwartz et al. (2011) considered a variety of theoretical frameworks such as: the social-ecological framework, the family management style framework, the disability-stress-coping model, the chronic care model, the medical traumatic stress model and the paediatric psychosocial preventative health model, in order to broaden the conceptualization of transition.  The SMART integrates adolescent/young adult characteristics (disease, neurocognitive and developmental maturity) with the interconnected components of knowledge, skills/self-efficacy, beliefs/expectations, goals, relationships and psychosocial      120 functioning of adolescents/young adults, parents and HCP (Schwartz et al., 2011).  SMART also considers socio-demographic, cultural, and health care access factors (Schwartz et al., 2011). Although the results of this study verify that ON TRAC’s developmental model of stage-for-age moderately corresponds with the developmental needs of youths with SHCN, our findings highlight that a developmental approach to transition by itself may be insufficient.  The SMART model includes factors that are not addressed in the ON TRAC model such as socio- demographics, access to resources, and the effect of relationships that potentially contribute to adolescents’ with SHCN knowledge, skills and self-care management behaviours.  Incorporating and addressing at least some of these elements into health care transition programs may enable HCP to better prepare youth for adult medical care by meeting more of their transitional needs. The Am I ON TRAC for Adult Care questionnaire is embedded in theoretical practice and captures adolescents’ knowledge, skills and behaviour related to SHCN and general health care self-management.   The initial evaluation of the ON TRAC questionnaire’s psychometric properties established it is a psychometrically sound measure with strong internal consistency and reliability.  Adolescent participants between 12 to 19 years of age from a variety of outpatient clinics (cardiology, diabetes, gastroenterology, and neurology) had no difficulty completing the questionnaire, which indicated that the items are clear and understandable.  An issue that did arise during the data collection process was youth who were not taking medications did not know how to answer the questions that specifically asked about medications.  To resolve this problem the four medication related items have been modified and to include a “I do not take any medications” response option.  Based on the results of the present study the ON TRAC questionnaire has the potential to be used as a readiness assessment tool to measure change in      121 adolescent patients over time and ultimately to predict successful transition from paediatric  to adult medical care. Strengths and Limitations The most significant limitation of this study was the inability to infer causal relationships due to its correlational design (Polit & Beck, 2008).  In addition, because this was a cross-section study it was not possible to infer changes over time (Polit & Beck, 2008).  Selection bias was another concern as individuals who chose to participate may have been significantly different than those who did not participate; this was a minor issue due to the high participation rate.  The PMI does not provide any information related to expected normative scores based on age.  As a result, we were unable to examine if participants’ psychosocial maturity scores were within an acceptable normative range for their age and whether adolescents’ with SHCN mature at the normal rate.  Lastly, all study materials were self-report measures, which raises some uncertainties regarding the validity and accuracy of the data.  Strengths of this research include its’ correlational and cross-section design which are efficient, effective and economical methods for collecting large amounts of data (Polit & Beck, 2006) and suitable for examining transition readiness and psychosocial maturity levels of adolescents ranging in ages from 12 to 19.  Using questionnaires was inexpensive, required minimal time commitment from participants, and the study package was easy to administer (Polit & Beck, 2006).  Our sample of 200 participants was adequate and increases our confidence in the findings.  Finally, the PMI is established as a reliable and valid measure for assessing the psychosocial maturity level of youth ages 11-18 (Greenberger et al., 1975).         122 Future Research  The ON TRAC questionnaire shows promise for being able to predict successful transition but we still do no know if this measure of readiness will lead to better transition outcomes or successful navigation of the adult medical system.  A critical next step is to put the questionnaires to use in clinics to assess patients’ transition readiness before, during and after transferring to adult care.  By following adolescents longitudinally through the transition process it will be possible to examine if ON TRAC scores are able to differentiate between individuals who successfully transition and those who do not.  Following-up with young adults in adult medical care will enable researchers to observe who is doing well in the adult system, who is not, and examine the relationships between the variables that are believed to influence transition.  It is also important to determine what are realistic and reasonable expectations of adolescents so that HCP feel comfortable that their patients are capable of self-managing their SHCN in the adult medical system.  Administering the Am I ON TRAC for Adult Care questionnaire to adults, aged 24 to 45 years, at adult clinics is one way we could determine what behaviours, knowledge and skills are needed for successful self-management as an adult.  Research that explores the role of family support and connectedness and how these factors impact adolescents’ transition readiness is also needed.            123 REFERENCES American Academy of Family Physicians, American Academy of Family Physicians, American College of Physicians – American Society of Internal Medicine (2002).  A consensus statement on health care transitions for young adults with special health care needs. Pediatrics, 110(6), 1304-1306. Retrieved on March 20, 2010 from: http://pediatrics.aappublications.org/content/110/Supplement_3/1304.full.pdf Annunziato, R, A., Parkar, S., Dugan, C. A., Barsade, S., Arnon, R., Tamir, M., Iyer, K., Kerkar, N., & Shemesh, E.  (2011).  Brief report: Deficits in health care management skills among adolescent and young adult liver transplant recipients transitioning to adult care settings.  Journal of Pediatric Psychology, 36(2), 155-159.  doi: 10.1093/jpepsy/jsp110 Arnett, J. J. (2000).  Emerging adulthood: A theory of development from the late teens through the twenties.  American Psychologist, 55(5), 469-408.  doi: 10.1037//0003-66X.55.5469 Barnette, J. J. (2000).  Effects of stem and Likert response option reversals on survey internal consistency: If you feel the need, there is a better alternative to using those negatively worded stems.  Educational and Psychological Measurement, 60(3), 361-370. doi: 10.1177/00131640021970592 BC Children’s Hospital.  (2010). Facts. Retrieved on June 22, 2011 from: http://www.bcchildrens.ca/AboutUs/BCCH/Facts.htm Benchimol, E. I., Walters, T. D., Kaufman, M., Frost, K., Fiedler, K., Chinea, Z., & Zachos, M. (2011).  Assessment of knowledge in adolescents with inflammatory bowel disease using a novel transition tool.  Inflammatory Bowel Disease, 17(5), 1131-1137.  doi: 10.1002/ibd.21464      124 Betz, C. L., Redcay, G., & Tan, S.  (2003).  Self-reported health care self-care needs of transition-aged youth: A pilot study.  Issues in Comprehensive Pediatric Nursing, 26(3), 159-181.  doi: 10.1080/01460860390223880 Betz, C. L.  (2004).  Transition of adolescents with special health care needs: Review and analysis of the literature.  Issues in Comprehensive Pediatric Nursing, 27(3), 179-241. doi: 10.1080/01460860490497903 Betz, C. L., & Telfair, J.  (2007).  Health care transitions: An introduction.  In W. M. Nehring (Ed.), Promoting health care transitions for adolescents with special health care needs and disabilities.  (pp. 3-19).  Baltimore, MD US: Paul H Brookes Publishing. Blum, R. W. M., Garell, M. D., Hodgman, C. H., Jorissen, T. W., Okinow, N. A., Orr, D. P., & Slap, G. B.  (1993).  Transition from child-centered to adult-centered systems for adolescents with chronic conditions: a position paper of the society for adolescent medicine.  Journal of Adolescent Health, 14(7), 570-576.  doi.org/10.1016/1054- 139X(93)90143-D Canadian Paediatric Society (2011).  Transition to adult care for youth with special health care needs.  Retrieved on April 10, 2011 from: http://www.cps.ca/english/statements/AM/AH07-01.htm Cappelli, M., MacDonald, N. E., & McGrath, P. J.  (1989).  Assessment of readiness to transfer to adult care for adolescents with cystic fibrosis.  Children’s Health Care, 18(4), 218-224. doi: 10.1207/s15326888chc1804_4 Chick, N., & Meleis, A.  (1986).  Transitions: A nursing concern.  In P.L. Chinn (Ed), Nursing research methodology (pp.237-257).  Boulder, CO: Aspen Publication.      125 Children’s & Women’s Health Centre of British Columbia.  (2000).  Setting the TRAC: A resource for health care professionals supporting youth with chronic health conditions and their families. Clarizia, N. A., Chahal, N., Manlhoit, C., Kilburn, J., Redington, A. N., & McCrindle, B. W. (2009).  Transition to adult health care for adolescents and young adults with congenital heart disease: Perspectives of the patient, parent, and health care provider.  Canadian Journal of Cardiology, 25(9), e317-e322.  doi: 10.1016/S0828-282X(09)70145-X de Beaufort, C., Jarosz-Chobot, P., Frank, M., de Bart, J., & Deja, G. (2010).  Transition from pediatric to adult diabetes care: Smooth or slippery?  Pediatric Diabetes, 11(1), 24-27. doi: 10.1111/j.1399-5448.2009.00524.x Fair, C. D., Sullivan, K., & Gatto, A.  (2010).  Best practices in transitioning youth with HIV: Perspectives of pediatric and adult infectious disease care providers.  Psychology, Health & Medicine, 15(5), 515-527.  doi: 10.1080/13548506.2010.493944 Field, A.  (2009).  Discovering Statistics Using SPSS.  (3rd ed.).  Los Angeles: Sage. Fleming, E., Carter, B., & Gillibrand, W.  (2002).  The transition of adolescents with diabetes from the children’s health care service into the adult health care service: A review of the literature.  Journal of Clinical Nursing, 11(5), 560-567.  doi: 10.1046/j.1365- 2702.2002.00639.x Flume, P. A.  (2009).  Smoothing the transition from pediatric to adult care: Lessons learned. Current Opinion in Pulmonary Medicine, 15(6), 611-614.  doi: 10.1097/MCP.0b013e3283314dec Floyd, F.J., & Widaman, K.F. (1995).  Factor analysis in the development and refinement of clinical assessment instruments.  Psychological Assessment, 7(3), 286-299.  Retrieved on      126 September 24, 2012 from: http://mres.gmu.edu/pmwiki/uploads/Main/FloydWidaman95.pdf Garrard, J. (2007). Health Sciences Literature Review Made Easy:  The Matrix Method.  (2nd  ed.).  Toronto:  Jones and Bartlett Publishers. Greenberger, E., Josselson, R., Knerr, C., & Knerr, B.  (1975).  The measurement and structure of psychosocial maturity.  Journal of Youth and Adolescence, 4(2), 127-143. Greenberger, E., & Sørensen, A. B.  (1974).  Toward a concept of psychosocial maturity. Journal of Youth and Adolescence, 3(4), 329-358. Hauser, E., & Dorn, L.  (1999).  Transitioning adolescents with sickle cell disease to adult centered care.  Pediatric Nursing, 25(5), 479-488.  Retrieved on January 15, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/detail?sid=a8dd482e-5361-4f6b-935c 4a8e6da37cef%40sessionmgr14&vid=1&hid=7&bdata=JnNpdGU9ZWhvc3Q tbGl2ZQ%3d%3d Hazel, E., Zhang, H., Duffy, C., Ciaran, M., & Campillo, S.  (2010).  High rates of unsuccessful transfer to adult care among young adults with juvenile idiopathic arthritis.  Pediatric Rheumatology, 8(2).  Retrieved from: http://www.ped-rheum.com/content/8/1/2 Hess, J. S., & Straub, D. M.  (2011).  Brief report: Preliminary findings from a pilot healthcare transition education intervention for adolescents and young adults with special health care needs. Journal of Pediatric Psychology, 36(2), 172-178. doi: 10.1093/jpepsy/jsq091 Hubley, A. M. & Zumbo, B. D. (in press).  Psychometric characteristics of assessment procedures: An overview.  In K.F. Geisinger (Ed.), APA handbook of testing and assessment in psychology.  Washington, DC: American Psychological Association.      127 Hudsmith, L. E., & Thorne, S. A. (2007). Transition of care from paediatric to adult services in cardiology. Archives of Disease in Childhood, 92(10), 927-930.  doi: 10.1136/adc.2006.103812 Jalkut, M. K., & Jackson Allen, P. L. (2009).  Transition from pediatric to adult health care for adolescents with congenital heart disease: A review of the literature and clinical implications.  Pediatric Nursing, 35(6), 381-387.  Retrieved on April 20, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/pdfviewer/pdfviewer?vid=3&hid=7&sid=1e6855de-df5a-4573-a35f-58a997364cd6%40sessionmgr11 Johnston, P., Bell, P. M., Tennet, H., & Carson, D.  (2006).  Audit of young people with type 1  diabetes transferring from paediatric to adult diabetes services.  Practical Diabetes International, 23(3), 106-108.  Retrieved on February 12, 2010 from: http://onlinelibrary.wiley.com/store/10.1002/pdi.914/asset/914_ftp.pdf?v=1&t=h9bnnbsg&s=84d3478612547c00c0676fc470c10898c278989a Kipps, S., Bahu, T., Ong, K., Ackland, F.M., Brown, R. S., Fox, C. T., Griffin, N. K., Knight, A. H., Mann, N. P., Niel, H. A. W., Simpson, H., Edge, J. A., & Dunger, D. B.  (2002). Current methods of transfer of young people with Type1 diabetes to adult services. Diabetic Medicine, 19(8), 649-654.  doi: 10.1046/j.1464-5491.2002.00757.x Kaufman, M. (2006). Role of adolescent development in the transition process. Progress in Transplantation, 16(4), 286-290.  Retrieved on May 15, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/pdfviewer/pdfviewer?sid=76691f4c-9571-4a21-a81f-64f73df42461%40sessionmgr14&vid=4&hid=7      128 Ledlie, S. W. (2007).  Methods of assessing transition health care needs.  In W. M. Nehring (Ed.), Promoting health care transitions for adolescents with special health care needs and disabilities.  (pp. 119-135).  Baltimore, MD US: Paul H Brookes Publishing. Lostein, D. S., Ghandour, R., Cash, A., McGuire, E., Strickland, B., & Newacheck, P.  (2009). Planning for health care transitions: Results of the 2005 2006 national survey of children with special health care needs.  Pediatrics, 123(1), e145-e152.  doi: 10.1542/peds.2008- 1298 Lostein, D, S., McPherson, M., Strickland, B., & Newacheck, P. W.  (2005).  Transition planning for youth with special health care needs: Results from the national survey of children with special health care needs.  Pediatrics, 115(6), 1562-1568.  doi: 10.1542/peds2004-1262 Machado, D. M., Succi, R. C., & Turato, E. R.  (2010).  Transitioning adolescents living with HIV/AIDS to adult-oriented health care: An emerging challenge.  Jornal de Pediatria, 86(6), 465-472.  doi: 10.2223/JPED.2048 McCrudy, C., DiCenso, A., Boblin, S., Ludwin, D., Bryant-Lukosius, D., & Bosompra, K. (2006).  There to here: Young adult patients’ perceptions of the process of transition from pediatric to adult transplant care.   Progress in Transplantation, 16(4), 309-316. Retrieved on April 27, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/pdfviewer/pdfviewer?sid=f670bd99-63a6-4309-b7e1-f1e55c9e5e89%40sessionmgr10&vid=2&hid=7 McDonagh, J. E., Southwood, T. R., & Shaw, K. L.  (2006).  The impact of a coordinated transitional care programme on adolescents with juvenile idiopathic arthritis. Rheumatology, 46, 161-168.  doi: 10.1093/rheumatology/kel198      129 McDonagh, J. E., & Kelly, D. A. (2010).  The challenges and opportunities for transitional care research.  Pediatric Transplantation, 14, 688-700.  doi: 10.1111/j.1399- 3046.2010.01343.x McManus, M., Fox, H., O’Connor, K., Chapman, T., & MacKinnon, J.  (2008).  Pediatric perspectives and practices on transitioning adolescents with special needs to adult health care.  The National Alliance to Advance Adolescent Health.  Retrieved on March 30, 2010 from: http://test.medicalhomeinfo.org/downloads/pdfs/FactsheetTransitioningAdolescents-McManus.pdf McPherson, M., Thaniel, L., & Minniti, C. P. (2009).  Transition of patients with sickle cell disease from pediatric to adult care: Assessing patient readiness.  Pediatric Blood Cancer, 52, 838-841.  doi: 10.1002/pbc.21974 Meleis, A. I.  (2010).  Transitions from practice to evidence-based models of care.  In A. I. Meleis, Transitions theory: middle-range and situation-specific theories in nursing research and practice.  (pp. 1- 11)  New York:  Springer Publishing Company. Meleis, A.I., Sawyer, L.M., Im, E.O., Hilfinger Messias, D.K., & Schumacher, K. (2000). Experiencing transitions: An emerging middle-range theory.  Advances in Nursing Science, 23(1), 12-28.  Retrieved on September 30, 2010 from: http://journals.lww.com/advancesinnursingscience/Abstract/2000/09000/Experiencing_Transitions__An_Emerging_Middle_Range.6.aspx Meleis, A.I., & Trangenstein, P. A.  (1994).  Facilitating transitions: redefinition of the nursing mission.  Nursing Outlook, 42(6), 255-259.  doi: 10.1016/0029-6554(94)90045-0      130 Moons, P., Pinxten, S., Dedroog, D., Van Deyk, K., Gewillig, M., Hilderson, D., & Werner, B. (2009).  Expectations and experiences of adolescents with congenital heart disease on being transferred from pediatric cardiology to an adult cardiology heart disease program. Journal of Adolescent Health, 44, 316-322.  doi: 10.1016/j.jadohealth.2008.11.007 Nakhla, M., Daneman, D., Frank, M., & Guttman, A.  (2008).  Translating transition: A critical review of the diabetes literature.  Journal of Pediatric Endocrinology & Metabolism, 21(6), 507-516.  Retrieved on November 9, 2012 from: http://www.slideshare.net/roger961/translating-transition-a-critical-review-of-the- diabetes Pacaud, D., McConnell, B., Huot, C., Aebi, C., & Yale, JF.  (1996).  Transition from pediatric care to adult care for insulin-dependent diabetes patients.  Canadian Journal of Diabetes, 20(4), 14-20. Pacaud, D., Yale, JF., Stephure, D., Trussell, R., & Davies, D. H. (2005).  Problems in transition from pediatric care to adult care for individuals with diabetes.  Canadian Journal of Diabetes, 29(1),13-18.  Retrieved on March 30, 2010 from: http://www.diabetes.ca/Files/Pacaud.pdf Pai, A. L. H., & Ostendorf, H. M.  (2011).  Treatment adherence in adolescents and young adults affected by chronic illness during the health care transition from pediatric to adult health care: a literature review.  Children’s Health Care, 40(1), 16-33.  doi: 10.1080/02739615.2011.537934 Pai, A. L. H., & Schwartz, L. A.  (2011).  Introduction to the special section: Health care transitions of adolescents and young adults with pediatric chronic conditions. Journal of Pediatric Psychology, 36(2), 129-133.  doi: 10.1093/jpepsy/jsq.100      131 Paone, M. C., Wigle, M., & Saewyc, E.  (2006).  The ON TRAC model for transitional care of adolescents.  Progress In Transplantation, 16(4), 291-302.  Retrieved on April 27, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/pdfviewer/pdfviewer?vid=3&hid=7&sid=76691f4c-9571-4a21-a81f-64f73df42461%40sessionmgr14 Patil, V. H., Singh, S.N., Mishra, S., & Donavan, D. T. (2007), "Parallel Analysis Engine to Aid Determining Number of Factors to Retain [Computer software]. Available from http://ires.ku.edu/~smishra/parallelengine.htm; Utility developed as part of Patil, V. H., Singh, S.N., Mishra, S., & Donovan, D.T. (2008), “Efficient Theory Development and Factor Retention Criteria: A Case for Abandoning the ‘Eigenvalue Greater Than One’ Criterion,” Journal of Business Research, 61 (2), 162-170.  doi: 10.1016/j.jbusres.2007.05.008 Patil, V.H., Singh, S.N., Mishra, S., & Donavan, D.T. (2008).  Efficient theory development and factor retention criteria: Abandon the ‘eigenvalue greater than one’ criterion.  Journal of Business Research, 61, 162-170.  doi:10.1016/j.jbusres.2007.05.008 Patterson, D, L., & Lanier, C.  (1999).  Adolescent health transitions: Focus group study of teens and young adults with special health care needs.  Family Community Health, 22(2), 43-58. Retrieved on February 16, 2010 from: http://ovidsp.tx.ovid.com.ezproxy.library.ubc.ca/sp- 3.7.1b/ovidweb.cgi?&S=IDHHFPGKINDDCGMLNCPKJGJCHFHGAA00&Link+Set= S.sh.25.26.29.33.36|6|sl_10 Polit, D. F., & Beck, C. T.  (2006).  Essentials of Nursing Research:  Methods, Appraisal, and Utilization.  (6th ed.). Philadelphia: Lippincott Williams & Wilkins. Polit, D.F., & Beck, C.T. (2008). Nursing Research: Generating and Assessing Evidence For      132  Nursing Practice (8th ed.). Philadelphia: Lippincott Williams & Wilkins. Por, J., Golberg, B., Lennox, V., Burr, P., Barrow, J., & Dennard, L.  (2004).  Transition of care: Health care professionals’ view.  Journal of Nursing Management, 12(5), 354-361. doi:10.1111/j.1365-2834.2004.00428.x Rapley, P., & Davidson, P.M.  (2010).  Enough of the problem: A review of time for health care transition solutions for young adults with a chronic illness.  Journal of Clinical Nursing, 19(3-4), 313-323.  doi: 10.1111/j.1365-2702.2009.03027.x Reid, G. J., Irvine, M. J., McCrindle, B. W., Sananes, R., Ritiva, P. G., Siu, S. C., & Webb, G. D. (2004).  Prevalence and correlates of successful transfer from pediatric to adult health care among a cohort of young adults with complex congenital heart defects.   Pediatrics, 113 (3), e197-e205.  Retrieved on February 16, 2010 from: http://pediatrics.aappublications.org/content/113/3/e197.full.pdf+html Reiss, J. G., Gibson, R. W., Walker, L. R.  (2005).  Health care transition: Youth, family and provider perspectives.   Pediatrics, 115(1), 112-120.  doi: 10.1542/peds.2004-1321. Rew, L.  (2005).  Adolescent Health: A multidisciplinary approach to theory, research and intervention.  Thousand Oaks, California: Sage Publications. Rosen, D. S., Blum, R. W., Britto, M., Sawyer, S. M., & Seigel, D.  (2003).  Transition to adult health care for adolescents and young adults with chronic conditions: Position paper of the Society  for Adolescent Medicine.  Journal of Adolescent Health, 33, 309-311. Retrieved on February 16, 2010 from: http://www.adolescenthealth.org/AM/Template.cfm?Section=Position_Papers&Te mplate=/CM/ContentDisplay.cfm&ContentID=1488      133 Sawicki, G. S., Lukens-Bull, K., Yin, X., Demars, N., Huang, I., Livingood, W., Reiss, J., & Wood, D.  (2011).  Measuring the transition readiness of youth with special healthcare needs: validation of the TRAQ – transition readiness assessment questionnaire.  Journal of Pediatric Psychology, 36(2), 160-171.  doi: 10.1093/jpepsy/jsp.128 Scal, P., Evans, T., Blozis, S., Okinow, N., & Blum, R.  (1999).  Trends in transition from pediatric to adult health care for young adults with chronic conditions.  Journal of Adolescent Health, 24(4), 259-264.  doi: 10.1016/S1054-139X(98)00127-X Scal, P., & Ireland, M.  (2005).  Addressing transition to adult health care for adolescents with special health care needs.  Pediatrics, 115(6), 1607-1612.  doi: 10.1542/peds.2004-0458 Scal, P., Davern, M., Ireland, M., & Park, K.  (2008).  Transition to adulthood: Delays and unmet needs among adolescents and young adults with asthma.   The Journal of Pediatrics, 152, 471-475.  doi: 10.1016/j.jpeds.2007.10.004 Schumacher, L.K., & Meleis, A.I. (1994).  Transitions: A central concept in nursing.  Image Journal of Nursing Scholarship, 26(2), 119-127.  Retrieved on September 30, 2010 from: http://www.transitiesinzorg.nl/files/schumacher-k-l-a-i-meleis-transitions-a-central- concept-in-nursing.pdf Schwartz, L. A., Tuchman, L. K., Hobbie, W. L., & Ginsberg, J. P.  (2011).  A social-ecological model of readiness for transition to adult-oriented care for adolescents and young adults with chronic health conditions.  Child: care, health and development, 37(6), 883-895. doi: 10.1111/j.1365-2214.2011.01282.x Scott, L., Vallis, M., Charette, M., Murray, A., & Latta, R.  (2005).  Transition of care: Researching the needs of young adults with type 1 diabetes.  Canadian Journal of      134 Diabetes, 29(3), 203-210.  Retrieved on February 16, 2010 from: http://www.diabetes.ca/Files/Scott--Transition pages 203-210.pdf Singh, S. P.  (2009).  Transition of care from child to adult mental health services: The great divide.  Current Opinion in Psychiatry, 22(4), 386-390.  doi: 10.1097/YCO.0b013e32832c9221 Soanes, C., & Timmons, S.  (2004).  Improving transition: A qualitative study examining the attitudes of young people with chronic illness transferring to adult care.  Journal of Child Health Care, 8(2), 102-112.  doi: 10.1177/1367493504041868 Stabile, L., Rosser, L., Porterfield, K. M., McCauley, S., Levenson, C., Haglund, J., & Chistman, K.  (2005).  Transfer versus transition: Success in pediatric transplantation brings the welcome challenge of transition.  Progress in Transplantation, 15(4), 363-370.  Retrieved on February 16, 2010 from: http://web.ebscohost.com.ezproxy.library.ubc.ca/ehost/pdfviewer/pdfviewer?sid=f3 5724e1-a02d-4615-8b92-5370c36ea5b6%40sessionmgr15&vid=4&hid=7 Swain, S.D., Weathers, D., & Niedrich, R.W. (2008).  Assessing three sources of misresponse to reversed likert items.  Journal of Marketing Research, 45(1), 116-131. doi: 10.1509/jmkr.45.1.116 Tabachnick & Fidell (2007). Using Multivariate Statistics.  Boston: Pearson Education Inc.  Telfair, J., Alexander, L.R., Loosier, P. S., Alleman-Velez, P.L., & Simmons, J.  (2004). Provider’s perspectives and beliefs regarding transition to adult care for adolescents with sickle cell disease.  Journal of Health for the Poor and Underserved, 15(3), 443-461. doi: 10.1353/hpu.2004.0049      135 Tuchman, L. K., Slap, G. B., & Britto, M. T. (2008). Transition to adult care: Experiences and expectations of adolescents with a chronic illness. Child: Care, Health & Development, 34(5), 557-563. doi: 10.1111/j.1365.2214.2008.00844.x Valenzuela, J. M., Buchanan, C. L., Radcliffe, J., Ambrose, C., Hawkins, L. A., Tanney, M., & Rudy, B. J.  (2011).  Transition to adult services among behaviorally infected adolescents with HIV – a qualitative study.  Journal of Pediatric Psychology, 36(2), 134-140.  doi: 10.1093/jpepsy/jsp051 van Staa, AL., Jedeloo, S., van Meeteren, J., & Latour, J. M.  (2011a).  Crossing the transition chasm: Experiences and recommendations for improving transitional care of young adults, parents and providers.  Child: care, health and development, 37(6), 821-831. doi:10.1111/j.1365-2214.2011.01261.x van Staa, AL., van der Stege, H. A., Jedeloo, S., Mool, H. A., & Hilberink, S. R. (2011b). Readiness to transfer to adult care of adolescents with chronic conditions: Exploration of associated factors.  Journal of Adolescent Health, 48, 295-302.  doi: 10.1016/j.jadohealth.2010.07.009 Van Walleghem, N., MacDonald, C. A., & Dean, H. J.  (2006).  Building connections for young adults with type1 diabetes mellitus in Manitoba: Feasibility and acceptability of a transition initiative.  Chronic Diseases Canada, 27(3), 130-134.  Retrieved on October 12, 2010 from: http://www.phac-aspc.gc.ca/publicat/cdic-mcbc/27-3/pdf/cdic273-5e.pdf Van Walleghem, N., MacDonald, C. A., & Dean, H. J.  (2008).  Evaluation of a systems navigator model for transition from pediatric to adult care for young adults with type 1 diabetes.  Diabetes Care, 31(8), 1529-1530.  doi:10.2337/dc07-2247      136 Velicer, W.F. & Jackson, D. N. (1990).  Component analysis versus common factor analysis: Some issues in selecting an appropriate procedure.  Multivariate Behavioral Research, 25(1), 1-28.  Retrieved on September 25, 2012 from: http://www.uam.es/personal_pdi/psicologia/carmenx/Articulo_ACP_AFE.pdf Viner, R. M. (2008). Transition of care from paediatric to adult services: One part of improved health services for adolescents. Archives of Disease in Childhood, 93(2), 160-163.  doi: 10.1136/adc.2006.103721 Visentin, K., Koch, T., & Kralik, D.  (2006).  Adolescents with type 1 diabetes: Transition between diabetes services.  Journal of Clinical Nursing, 15(6), 761-769.  doi: 10.1111/j.1365-2702.2006.01347.x Whittaker, C. B.  (2004).  Transfer of young adults with Type1 diabetes from pediatric to adults diabetes care.  Diabetes Quarterly, 2, 10-14.  Retrieved on February 16, 2010 from: http://www.diabetes.ca/files/Professional Pub Archives/DiabetesQuarterly/Diabetes Quarterly--FINAL--Spring2004.pdf Wiener, L. S., Kohrt, BA., Battles, H. B., & Pao, M.  (2011).  The HIV experience: Youth identified barriers for transitioning from pediatric to adult care. Journal of Pediatric Psychology, 36(2), 141-154.  doi: 10.1093/jpepsy/jsp129 Williams, T. S., Sherman, E. M. S., Dunseith, C., Mah, J. K., Blackman, M., Latter, J., Mohamed, I., Slick, D. J., & Thornton, N.  (2011).  Measurement of medical self-management and transition readiness among Canadian adolescents with special health care needs. International Journal of Child and Adolescent Health, 3(4), 527-535.  Retrieved on June 15, 2012      137 from: https://www.novapublishers.com/catalog/product_info.php?products_id=22999 Wilson, J., & Gramling, L.  (2009).  The application of Orem’s self-care model to burn care. Journal of Burn Care & Research, 30 (5), 852-858.  doi: 10.1097/BCR.0b013e3181b48a2d Wong, D. L., Perry, S. E., & Hockenberry, M. J.  (2002).  Maternal Child Nursing Care 2nd Edition.  St. Louis, MO: Mosby Inc. Wong, L. H. L., Chan, F. W. K., Wong, F. Y. Y., Wong, E. L. Y., Huen, K. F., Yeo, E-K., & Fok, T-F.  (2010).  Transition care for adolescents and families with chronic illness.  Journal of Adolescent Health, 47, 540-546.  doi: 10.1016/j.jadohealth.2010.04.002 Zwick, W.R., and Velicer, W.F. (1986).  Comparison of five rules for determining the number of components to retain.  Psychological Bulletin, 99(3), 432-442.        February 28, 2012    138 Appendix A Youth Study Information Letter     Youth Study Information Letter  Research Project:  Assessing adolescents with special health care needs: Readiness to transition from pediatric to adult medical care  Principal Investigator: Dr. Elizabeth Saewyc, Professor, UBC School of Nursing  Division of Adolescent Health & Medicine        Phone  (xxx) xxx-xxxx xxxxxxxxxxxx@nursing.ubc.ca  Co Investigators: Melissa Moynihan Graduate Student, UBC School of Nursing Phone (xxxx) xxx-xxxx xxxxxxxx@interchange.ubc.ca  Dr. Gladys McPherson Assistant Professor, UBC School of Nursing  Mary C. Paone, RN, MSN Transition Coordinator B.C’s Children’s Hospital  Dr. Sandy Whitehouse Division of Adolescent Health & Medicine B.C’s Children’s Hospital  This research is part of Melissa Moynihan’s master in nursing science degree and will be part of her thesis.  You are being invited to take part this research study because we are trying to learn more about assessing adolescents’ readiness to transition from B.C. Children’s Hospital clinics to adult hospital clinics.  Study Procedures: You are being asked to participate in this study because you are 12-18 years old, an outpatient at a medical specialties clinic B.C. Children’s Hospital outpatient clinics, and have a special health care condition.  You will fill out a questionnaire that asks you about your readiness to transition to an adult hospital clinic, your general attitudes, and your relationships with your parent(s).  The questionnaire will take 15-20 minutes for you to fill out.  You are asked to fill out the questionnaire here in the clinic waiting room and return it to the investigator before you leave the clinic today.        (Please turn over) The University of British Columbia School of Nursing Vancouver Campus T201-2211 Wesbrook Mall Vancouver, BC Canada V6T 2B5  Phone  xxx xxx xxxx Fax  xxx xxx xxxx www.nursing.ubc.ca  February 28, 2012   139  Participating in this study is entirely voluntary and you do not need to fill out the questionnaire if you do not wish to.  Deciding not to participate will not change anything for you and will not affect your health care in the clinic at all. If you agree to participate you can choose not to answer any question, or stop at any time and it will not affect the care you receive at B.C.’s Children’s Hospital. Completing and returning this questionnaire indicates your consent to participate in the study.  Confidentiality Your confidentiality will be respected.  We will not tell other people that you are in this research and we won't share information about you to anyone who does not work on the research study. You will not put your name on the questionnaire and only the researchers will see your answers. Questionnaires will be kept in a locked filing cabinet.  Data will be stored on a password protected computer.   The data collected during this study will be used only for the purposes outlined in this written consent form.  Risks and Benefits There are no anticipated risks to participating in this study.  Benefits of participating in this study include contributing to the improvement of health care transition services for youth with special health care needs.  Reimbursements You will be given a $10 iTunes gift card for filling out the questionnaire.  The gift card will be given to you when you return your completed questionnaire to me.  Contact for information about the study: If you have any questions or if you would like more information about this study, contact Melissa Moynihan at (xxx) xxx – xxxx or at xxxxxxxxx@interchange.ubc.ca  Contact information about the rights of research subjects: If you have any concerns your treatment or rights as a research subject, you may contact the Research Subject Information Line in the UBC Office of Research Services at RSIL@ors.ubc.ca, toll free at (1-877-822-8598) or at the local number (604) 822-8598.                February 27, 2012   140 Appendix B Recruitment Flyer  Assessing Readiness to Transition  Youth 12 to 18 yrs & their parents are needed for research: Researchers at UBC and B.C’s Children’s Hospital want to learn more about youths’ readiness to transition from children’s clinics to adult medical clinics.  Are you: • 12 – 18 years old • A patient at one of the following outpatient clinics at B.C. Children’s Hospital: ♦ Cardiology, Diabetes, Gastroenterology, or Neurology •  Diagnosed with a health condition and will need to transition to an adult clinic when you are older   Fill out a survey (15 – 20 minutes) that asks about your readiness to transition to an adult clinic, your attitudes, and your relationship with your parent(s).  Your parents will also fill out a survey about your readiness to transition and their relationship with you.  Youth get an iTunes Gift Card For more information, please approach the nurse researcher in the clinic waiting room or e-mail moyniha3@interchange.ubc.ca  The principal researcher for this study is Dr. Elizabeth Saewyc, Professor, Division of Adolescent Health & Medicine & UBC School of Nursing, xxxxxxxxxxx@nursing.ubc.ca   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 141 Appendix C Participant Study Package         Assessing Readiness to Transition from Pediatric to Adult Medical Care Youth Version    Date: ____________________________________   Participant Number: ________________________        RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 142  Please completely fill in the circle of your answer choice for each survey question.      1. Are you:      male      female   2. What clinic are you attending?  Diabetes  Cardiology  Gastroenterology  Neurology   3. How old are you?  12  13  14  15  16  17  18   4. How old were you when you were diagnosed with your health condition?         _____________ (age)            Background   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 143   Direction:  We would like to know how you describe your skills in the areas that are important in your care.  Your answers will help us provide services and education that will be important in preparing you to transition to adult care.  There is no right or wrong answer and your answers will remain confidential and private.  Please check the box  that you feel best describes you.    Not needed for my care  No, I do not know how No, I do not know how but I want to learn  No, but I am learning to do this Yes, I have started doing this  Yes, I always do this when I need to Skills for Chronic Condition Self- Management 1. Do you fill a prescription if you need to? 2. Do you know what to do if you are having a bad reaction to your medications? 3. Do you pay or arrange payments for your medications? 4. Do you take medications correctly and on your own? 5. Do you reorder medications before they run out? 6. Do you take care of your medical equipment and supplies? 7. Do you call the suppliers when there is a problem with the equipment? 8. Do you order medical equipment before they run out? 9. Do you arrange payments for the medical equipment and supplies? 10. Do you call the doctor’s office to make an appointment? 11. Do you follow-up on any referral for tests or check-ups or labs? 12. Do you arrange for your ride to medical appointments? 13. Do you call the doctor about unusual changes in your health (For example: Allergic reactions)?  14. Do you apply for health insurance if you lose your current coverage? 15. Do you know what your health insurance covers? 16. Do you manage your money & budget household expenses (For example: Use checking/debit card)?  Transition Readiness Assessment Questionnaire 4.1   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 144              Not needed for my care  No, I do not know how No, I do not know how but I want to learn  No, but I am learning to do this Yes, I have started doing this Yes, I always do this when I need to Skills for Self-Advocacy and Health Care Utilization  17. Do you fill out the medical history form, including a list of your allergies?  18. Do you keep a calendar or list of medical and other appointments?  19. Do you tell the doctor or nurse what you are feeling?  20. Do you answer questions that are asked by the doctor, nurse or clinic staff?  21. Do you ask questions of the doctor, nurse or clinic staff (For example: What medications or treatments are best for you)?  22. Do you make a list of questions before the doctor’s visit?  23. Do you request the accommodations & support you need at school or work?  24. Do you apply for a job or work or vocational services?  25. Do you get financial help with school or work?  26. Do you help plan or prepare meals/food?   27. Do you keep home/room clean or clean- up after meals?  28. Do you use neighbourhood stores and services (For example: Grocery stores and pharmacy stores)?  29. Do you use community support services (For example: After school programs) or advocacy services (For example: Legal services) when you need them?    RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 145    For each of the following statements, select the response that best describes you: 1. I can describe my health condition to others  Strongly disagree  Disagree  Agree  Strongly Agree  2. I have a family doctor  Strongly disagree  Disagree  Agree  Strongly Agree  3. I visit my family doctor when I need to (For example: to have check-ups, get birth control, or if I have the flu)  Strongly disagree  Disagree  Agree  Strongly agree  4. I know what my long-term health problems might be  Strongly disagree  Disagree  Agree  Strongly agree  5. I know what patient confidentiality means  Strongly disagree  Disagree  Agree  Strongly agree  6. I understand the risks and benefits of health care treatments before consenting to those treatments  Strongly disagree  Disagree  Agree  Strongly agree  7. I know how to get my medical records  Strongly disagree  Disagree  Agree  Strongly agree 8. I know the names of my medications  Strongly disagree  Disagree  Agree  Strongly agree  9. I know what each of my medications are for  Strongly disagree  Disagree  Agree  Strongly agree  10. I know the side effects of the medications I take  Strongly disagree  Disagree  Agree  Strongly agree  11. I can get to my clinic appointments on my own  Strongly disagree  Disagree  Agree  Strongly agree  12. I know how my condition might affect my sexual health  Strongly disagree  Disagree  Agree  Strongly agree  13. I know how to prevent sexual health risks such as pregnancy and sexually transmitted infections (STIs)  Strongly disagree  Disagree  Agree  Strongly agree  14. My family supports me in managing my health  Strongly disagree  Disagree  Agree  Strongly agree  15. I know how my health condition might limit my career choices  Strongly disagree  Disagree  Agree  Strongly agree Am I ON TRAC for Adult Care?   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 146    16. I know how my health condition affects my physical activities  Strongly disagree  Disagree  Agree  Strongly agree  17. I know how alcohol, drugs and tobacco can affect my medications  Strongly disagree  Disagree  Agree  Strongly agree  18. I meet with my health care providers on my own  Never  Rarely  Sometimes  Often  Always  19. I feel my family cares about me  Never  Rarely  Sometimes  Often  Always  20. I participate in clubs, groups, sports or activities  Never  Rarely  Sometimes  Often  Always  21. I talk to my friend(s) about my problems or worries  Never  Rarely  Sometimes  Often  Always  22. I participate in activities/exercise to stay healthy  Never  Rarely  Sometimes  Often  Always  23. I ask health care providers questions about my health at my visits  Never  Rarely  Sometimes  Often  Always  24. I take my medications on my own  Never  Rarely  Sometimes  Often  Always  25. When my symptoms are getting worse I contact the clinic for help  Never  Rarely  Sometimes  Often  Always  26. I think about what I would like to do after high school  Never  Rarely  Sometimes  Often  Always     RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 147     How much do you agree or disagree with the following statements?  If you haven’t been chosen as the leader, you shouldn’t suggest how things should be done…………  In a group I prefer to let other people make the decisions…………………………………………………………………  You can’t be expected to make a success of yourself if you had a bad childhood………………………………………  When things have gone wrong for me, it is usually because of something I couldn’t do anything about..    How much do you agree or disagree with the following statements?  Hard work is never fun ……………………………………………  I find it hard to stick to anything that takes a long time to do………………………………………………….……….…..  No one should expect you to do work that you don’t like…………………………………………………………………  I get upset if I am not immediately successful in learning something new………………………………………….    How much do you agree or disagree with the following statements?  Luck decides most things that happen to me…………..  The main reason I’m not more successful is that I have bad luck………………………………………………………….  When things go well for me, it is usually not because of anything I myself actually did………………………………..  I feel very uncomfortable if I disagree with what my friends think…………………………………………………………… Strongly             Disagree                 Agree                     Strongly Disagree                                                                             Agree   ....................................................................................   ....................................................................................   ....................................................................................    ....................................................................................    Strongly             Disagree                 Agree                     Strongly Disagree                                                                             Agree  ....................................................................................   ....................................................................................    ....................................................................................   ....................................................................................    Strongly             Disagree                 Agree                     Strongly Disagree                                                                             Agree  ....................................................................................   ....................................................................................    ....................................................................................   ....................................................................................  My Attitudes   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 148                  How much do you agree or disagree with the following statements?  I seldom get behind in my work…………………………………  I tend to go from one thing to another before finishing any one of them…………………………………………  I often don’t finish work I start………………………………....  I often leave my homework unfinished if there are a lot of good TV shows on that evening………………………..    How much do you agree or disagree with the following statements?  It is best to agree with others, rather than say what you think, if it will keep peace …………………………………  I don’t know whether I like a new outfit until I find out what my friends think…..……………………….……….…..  It’s very important to me to do my homework well…..  Very often I forget work I am supposed to do.…….…….   Strongly             Disagree                 Agree                     Strongly Disagree                                                                             Agree  ....................................................................................   ....................................................................................   ....................................................................................   ....................................................................................    Strongly             Disagree                 Agree                     Strongly Disagree                                                                             Agree   ....................................................................................    ....................................................................................  ....................................................................................  ....................................................................................   RS: ______________ Youth Survey:  VERSION 1 Jan. 28, 2012 149     Choose the rating you feel best describes each of your parents   Relationship with my mother Almost never or never true  Seldom true  Sometimes true  Often true  Almost always or always true  1. My mother is always telling me how I should behave ……..…...............................................................................        1. 2. My mother trusts me………………………………………………..……….................................................................................        2. 3. My mother wants to control whatever I do………….…………..…..............................................................................         3. 4. My mother shows interest and support for me…………….….…..............................................................................         4. 5. My mother is always trying to change me………….……….…...…..............................................................................         5.  Almost never or never true Seldom true Sometimes true Often true Almost always or always true 6. My mother insists that I must do exactly as I am told……..…................................................................................         6. 7. My mother feels affection for me………………………………..……................................................................................         7. 8. My mother approves of me and the things I do………….………...............................................................................         8. 9. My relationship with my mother is getting better…………….…..............................................................................         9.    Relationship with my father Almost never or never true  Seldom true  Sometimes true  Often true  Almost always or always true  1. My father is always telling me how I should behave ………..…..............................................................................         1. 2. My father trusts me…………………………………………….………...…................................................................................         2. 3. My father wants to control whatever I do…………………..…....….............................................................................         3. 4. My father shows interest and support for me……….…….....…...............................................................................         4. 5. My father is always trying to change me………………….……….................................................................................         5.  Almost never or never true Seldom true Sometimes true Often true Almost always or always true 6. My father insists that I must do exactly as I am told………….….............................................................................         6. 7. My father feels affection for me………………………………….….…................................................................................        7. 8. My father approves of me and the things I do…………....………..............................................................................         8. 9. My relationship with my father is getting better………….……................................................................................         9. For Teens:  Relationship with your Parents   150 Appendix D Principal Component Analysis of the Psychosocial Maturity Inventory Table 22 Component Loadings and Eigenvalues for PCA of 20-item PMI   Components PMI Item 1 2 1. If you haven’t been chosen as the leader, you shouldn’t suggest how things should be done .45 .09 2. In a group I prefer to let other people make the decisions .59 .12 3. You can’t be expected to make a success of yourself if you had a bad childhood .32 .27 4. When things have gone wrong from me, it is usually because of something I couldn’t do anything about .42 -.06 5. Hard work is never fun .41 .09 6. I find it hard to stick to anything that takes a long time to do .65 -.02 7. No one should expect you to do work that you don’t like .43 -.04 8. I get upset if I am not immediately successful in learning something new .29 .11 9. Luck decides most things that happen to me  .54 .23 10. The main reason I’m not more successful is that I have bad luck .53 .25 11. When things go well for me, it is usually not because of anything I myself actually did .55 .36 12. I feel very uncomfortable if I disagree with what my friends think .43 .48 13. I seldom get behind in my work .04 -.33 14. I tend to go from one thing to another before finishing any one of them .61 -.25 15. I often don’t finish work I start .65 -.53 16. I often leave my homework unfinished if there are a lot of good TV shows on that evening .55 -.39 17. It is best to agree with others, rather than say what you think, if it will keep the peace .44 .39 18. I don’t know whether I like a new outfit until I find out what my friends thing .43 .21 19. It’s very important to me to do my homework well .37 -.38 20. Very often I forget work I am supposed to do .54 -.49  Eigenvalue 4.65 1.79 Percentage of variance 23.27 8.94 n = 183 Note. Factor loadings > .40 are in boldface. Only components supported by the parallel analysis are shown.          151  Table 23  Component Loadings and Eigenvalues for PCA of the 18-item PMI   Components PMI Item 1 2 1. If you haven’t been chosen as the leader, you shouldn’t suggest how things should be done .45 .12 2. In a group I prefer to let other people make the decisions .59 .13 3. You can’t be expected to make a success of yourself if you had a bad childhood .32 .27 4. When things have gone wrong from me, it is usually because of something I couldn’t do anything about .42 -.10 5. Hard work is never fun .41 .13 6. I find it hard to stick to anything that takes a long time to do .65 -.03 7. No one should expect you to do work that you don’t like .43 -.07 9. Luck decides most things that happen to me  .54 .23 10. The main reason I’m not more successful is that I have bad luck .54 .25 11. When things go well for me, it is usually not because of anything I myself actually did .55 .39 12. I feel very uncomfortable if I disagree with what my friends think .43 .47 14. I tend to go from one thing to another before finishing any one of them .60 -.26 15. I often don’t finish work I start .65 -.53 16. I often leave my homework unfinished if there are a lot of good TV shows on that evening .55 -.43 17. It is best to agree with others, rather than say what you think, if it will keep the peace .44 .43 18. I don’t know whether I like a new outfit until I find out what my friends thing .43 .21 19. It’s very important to me to do my homework well .37 -.37 20. Very often I forget work I am supposed to do .55 -.47  Eigenvalue 4.58 1.74 Percentage of variance 25.46 9.66 n = 183 Note. Factor loadings > .40 are in boldface. Only components supported by the parallel analysis are shown.        152 Appendix E Revised Am I ON TRAC for Adult Care Questionnaire 1. I can describe my health condition to others  Strongly disagree  Disagree  Agree  Strongly Agree  2. I have a family doctor  Strongly disagree  Disagree  Agree  Strongly Agree  3. I know what my long-term health problems might be  Strongly disagree  Disagree  Agree  Strongly agree  4. I know what patient confidentiality means  Strongly disagree  Disagree  Agree  Strongly agree  5. I understand the risks and benefits of health care treatments before consenting to those treatments  Strongly disagree  Disagree  Agree  Strongly agree  6. I know how to get my medical records  Strongly disagree  Disagree  Agree  Strongly agree  7. I know the names of my medications  Strongly disagree  Disagree  Agree  Strongly disagree  I do not take any medications 8. I know what each of my medications are for  Strongly disagree  Disagree  Agree  Strongly agree  I do not take any medications  9. I know the side effects of the medications I take  Strongly disagree  Disagree  Agree  Strongly agree  I do not take any medications  10. I can get to my clinic appointments on my own  Strongly disagree  Disagree  Agree  Strongly agree  11. I know how my condition might affect my sexual health  Strongly disagree  Disagree  Agree  Strongly agree  12. I know how to prevent sexual health risks such as pregnancy and sexually transmitted infections (STIs)  Strongly disagree  Disagree  Agree  Strongly agree  13. My family supports me in managing my health  Strongly disagree  Disagree  Agree  Strongly agree  14. I know how my health condition might limit my career choices  Strongly disagree  Disagree  Agree  Strongly agree   153 15. I know how my health condition affects my physical activities  Strongly disagree  Disagree  Agree  Strongly disagree  16. I know how alcohol, drugs and tobacco can affect my medications  Strongly disagree  Disagree  Agree  Strongly agree  I do not take any medications  17. I visit my family doctor when I need to (For example: to have check-ups, get birth control, or if I have the flu)  Never  Rarely  Sometimes  Often  Always  18. I meet with my health care providers on my own  Never  Rarely  Sometimes  Often  Always  19. I participate in clubs, groups, sports or activities  Never  Rarely  Sometimes  Often  Always  20. I talk to my friend(s) about my problems or worries  Never  Rarely  Sometimes  Often  Always  21. I participate in activities/exercise to stay healthy  Never  Rarely  Sometimes  Often  Always 22. I ask health care providers questions about my health at my visits  Never  Rarely  Sometimes  Often  Always  23. I take my medications on my own  Never  Rarely  Sometimes  Often  Always  I do not take any medications  24. When my symptoms are getting worse I contact the clinic for help  Never  Rarely  Sometimes  Often  Always  25. I think about what I would like to do after high school  Never  Rarely  Sometimes  Often  Always     154 Appendix F Scoring Guide for Revised Am I ON TRAC for Adult Care Questionnaire  Knowledge Scale - Items 1 to 16 Scoring: strongly disagree = 1, disagree = 2, agree = 3, strongly agree = 4 Items 2 and 13 are not included in the score. Sum item scores.  Scores range from 16 to 64. *If youth indicate they are not taking any medications items 7, 8, and 9 are not scored. Scores range from 13 to 52.  Behaviour Index – Items 17 to 25 Scoring: never = 1, rarely = 2, sometimes = 3, often = 4, always = 5 Raw Score - sum item scores.  Score range from 9 to 45 *If youth indicate they are not taking any medications item 23 is not scored. Scores range from 8 to 40. Cut-off Score – items are given a 0 if threshold is not met and 1 if the threshold is met Item thresholds: 17 – Often, 18 – Sometimes, 20 – Sometimes, 21 – Sometimes, 22 – Often, 23 – Often, 24 – Often, 25 – Often Threshold scores are summed.  Youth need a cut-off score of at least 8 or 9 to be deemed ready for transition to adult care. * If youth indicate they are not taking any medications item 23 is not scored and youth need a cut-off score of at least 7 or 8 to be deemed ready for transition to adult care.  

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.24.1-0073464/manifest

Comment

Related Items