Background. With increasing personal affluence and changes in lifestyle, there is rising prevalence of type 2 diabetes mellitus in Hong Kong. Approximately 60% of diabetic subjects in Hong Kong are asymptomatic and previously undiagnosed. Since diabetes carries significant mortality and morbidity risk, it is important to diagnose these subjects early for intervention. There are many known factors associated with development of type 2 diabetes. Some are remediable such as obesity, dyslipidaemia, hypertension, while some are non-remediable such as age and past history of gestational diabetes. Identifying high-risk subjects will increase the yield and cost-effectiveness of screening program for diabetes and related risk factors and provide useful epidemiological information on the natural history of these diseases. / Methods. I used data from several cross-sectional and prospective studies of which I was the principal investigator or one of the co-investigators to test these hypotheses. The studies include mainly the following: (1) A public utility company workforce survey on cardiovascular risk factors in 1990 (n=1513). (2) Chinese subjects with risk factors for diabetes who underwent 75 gram oral glucose tolerance test (OGTT) screening at the Prince of Wales Hospital (PWH) between 1988 and 1995 (n=3718). (3) The 'United Christian Nethersole Community Health Service' (UCNCHS) primary health screening program database in 1997 (n=17764). / Objectives & hypothesis. I aimed to study the various factors associated with the development of type 2 diabetes in Hong Kong Chinese. With this information, I can design a screening method to early identify those subjects who are at high-risk for diabetes. I hypothesize the following: (1) Many risk factors for diabetes in Caucasians are also applicable to Hong Kong Chinese. (2) The presence of multiple factors increases the risk of diabetes in a linear fashion. (3) Chinese subjects are at risk of developing diabetes at a lower threshold of obesity, which is one of the most important risk factors for type 2 diabetes. / Results. Based on a cohort of 1513 asymptomatic subjects from a workforce survey, those in the top quartile of body mass index (BMI), as compared to those in the lowest quartile, had a 4 to 10-fold increased risk of diabetes and a 2.5 to 5-fold increased risk of impaired glucose tolerance (IGT) for men and women. (Abstract shortened by UMI.) / Ko Tin Choi. / "May 2005." / Source: Dissertation Abstracts International, Volume: 67-01, Section: B, page: 0173. / Thesis (M.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 264-283). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / School code: 1307.

Fok, Tsz Nam = 香港二型糖尿病併發症風險評估模型 / 霍梓楠. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (p. 71-72). / Abstracts in English and Chinese. / Fok, Tsz Nam = Xianggang er xing tang niao bing bing fa zheng feng xian ping gu mo xing / Huo Zinan. / Abstract --- p.i / 概要 --- p.iii / Acknowledgements --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Dataset Information --- p.3 / Chapter 3 --- Background and Literature Review --- p.9 / Chapter 3.1 --- The Idea of Risk Model --- p.9 / Chapter 3.2 --- Discrimination Problem --- p.10 / Chapter 3.3 --- Receiver Operating Characteristic (ROC) Curve --- p.11 / Chapter 3.4 --- Summary Indices of the ROC Curve --- p.13 / Chapter 3.4.1 --- Area Under ROC Curve (AROC) --- p.14 / Chapter 3.4.2 --- Maximum Vertical Distance --- p.16 / Chapter 3.5 --- Discrimination Performance in Prognostic Model --- p.18 / Chapter 3.5.1 --- Survival Data --- p.18 / Chapter 3.5.2 --- Survival Function --- p.20 / Chapter 3.5.3 --- Time-dependent ROC Curve for Censored Data --- p.21 / Chapter 3.6 --- Earlier Work on Diabetic Complications Risk Models --- p.22 / Chapter 3.6.1 --- Maximization of the AROC --- p.28 / Chapter 4 --- Model Development --- p.29 / Chapter 4.1 --- Overview --- p.29 / Chapter 4.2 --- Estimating the ROC curve and the AROC --- p.29 / Chapter 4.3 --- Choosing Suitable Risk Factors --- p.30 / Chapter 4.4 --- Mixing Risk Factors and Optimizing Coefficients --- p.31 / Chapter 4.5 --- Validation of Risk Equations Using Test Set --- p.33 / Chapter 5 --- Results and Validation --- p.34 / Chapter 5.1 --- Performance of the Risk Factor Candidates --- p.34 / Chapter 5.2 --- Estimation of the Coefficients --- p.37 / Chapter 5.3 --- Checking the Uniqueness of the Solution --- p.41 / Chapter 5.4 --- Validation Using Test Set --- p.46 / Chapter 5.5 --- Comparison of AROC-optimized and MVD-optimized Risk Equations --- p.56 / Chapter 6 --- Comparison of our Results with Earlier Work --- p.58 / Chapter 7 --- "Discussion, Outstanding Issues and Future Works" --- p.66 / Chapter 7.1 --- Comparison Between the AROC and the MVD --- p.66 / Chapter 7.2 --- Applications of Risk Models --- p.68 / Chapter 7.3 --- Limitations of the study --- p.69 / Chapter 7.4 --- Outstanding Issues and Future Works --- p.69 / Chapter 7.4.1 --- The Estimation of Error Due to Sampling Variance --- p.69 / Chapter 7.4.2 --- Time-dependent Coefficients --- p.69 / Chapter 7.4.3 --- Extending the Idea to other Datasets --- p.70 / Chapter 7.5 --- Conclusion --- p.70 / Bibliography --- p.71

Wu Man Sze Anise. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (leaves 72-84). / Abstracts in English and Chinese. / ACKNOWLEDGEMENTS --- p.ii / ABSTRACT --- p.iii / TABLE OF CONTENTS --- p.vi / LIST OF TABLES --- p.viii / Chapter CHAPTER 1 --- INTRODUCTION --- p.1 / Health Habits --- p.2 / Health Habits and Health Status of Elderly Adults --- p.2 / Health Habits among Chinese Elders with Diabetes --- p.3 / Measurement of Health Habits --- p.5 / Perceived Control --- p.6 / Perceived Control over Health --- p.6 / Health Locus of Control and Health Value --- p.7 / Self-efficacy --- p.15 / Previous Research Focus of Effect of Perceived Control on Chronic Illness --- p.16 / Perceived Control and Chinese Culture --- p.19 / Chapter CHAPTER 2 --- MAIN STUDY --- p.20 / Purpose of the Present Study --- p.20 / Method --- p.22 / Participants --- p.22 / Instruments --- p.24 / Health Habits --- p.24 / Health Locus of Control --- p.25 / Health Value --- p.26 / Self-efficacy --- p.26 / Health Status --- p.26 / Blood Glucose Control --- p.27 / Demographics --- p.27 / Procedure --- p.28 / Data Analysis --- p.28 / Results --- p.30 / Results for the First Wave Data --- p.30 / Analysis of Demographic Information --- p.30 / Analyses of Blood Glucose --- p.33 / Factor Analyses of Health Habits --- p.35 / Analysis of Demographic/ Medical Effects on Two Health Habits Dimensions --- p.36 / Correlations among Psychological Variables and Health Habits at Time1 --- p.38 / Regression Analyses of General Health Habits at Time1 --- p.40 / Regression Analyses of Diabetes-related and Preventive Health Habits at Time1 --- p.42 / Results for Second Wave Data --- p.45 / Correlations of Health Habits/Health Status at --- p.45 / Time 1 and Health Status at Time2 / Changes across Time --- p.46 / Stepwise Regression Analysis of Health Habits at Time2 --- p.47 / Discussion --- p.50 / Predictors of Health Habits at Time1 --- p.50 / Predictive Power of Psychological Constructs over --- p.57 / Habits at Time2 / Glycemic Control --- p.58 / Demographic and Medical Influences on Health --- p.60 / Status and Psychological Variables / Limitations --- p.61 / Research and Practical Implications --- p.63 / Conclusion / REFERENCES --- p.70 / FOOTNOTES --- p.72 / APPENDIX A Health Habits Checklist --- p.85 / APPENDIX B Self-efficacy (Toward Health Habits) and Health Value Scales --- p.103 / APPENDIX C Multidimensional Health Locus of Control Scales --- p.104