Many areas of clinical informatics research rely on accurate and complete race and ethnicity (RE) patient information, such as estimating disease risk, assessing quality and performance metrics, and identifying health disparities. Structured data in the electronic health record (EHR) is an easily accessible source for patient-level information, however RE information is often missing or inaccurate in structured EHR data. Furthermore, current federal standards on RE categories have been acknowledged as inadequate, and in need of more granular realizations. While more difficult to extract data from, clinical notes provide a rich, nuanced and subjective source of information that can be leveraged to increase granularity and/or recover RE information missing in structured data. State-of-the-art clinical natural language processing (NLP) approaches can enable researchers to extract RE information from clinical notes, however, NLP has also been shown to inherit, exacerbate, and create new biased and harmful associations, especially in modern deep learning approaches. This thesis explores the relationships between direct and indirect explicit mentions of RE and RE inferences in clinical text annotated by humans, and leverages an approach to audit deep NLP models for their learned associations. We develop gold-standard annotations for information related to RE (RE indicators) and RE labels. We leverage four RE indicators: country of origin, spoken language, direct race, and direct ethnicity mention. We find high agreement between annotators for RE label assignments, and that sentences assigned RE categories have drastically different distributions of RE indicators. Furthermore, we find high agreement between structured and unstructured sources of RE information, and that unstructured data can be used to recover missing RE information in structured data. Leveraging the gold-standard RE annotations, we train a model to label sentences with RE information and audit the model to examine the alignment between salient features and RE indicators. While our models attain good classification performance, this does not translate into high overlap with RE indicators. We find evidence for learned associations that are benign mistakes, helpful but not strictly correct, and potentially harmful mistakes if not addressed by future work.

published_or_final_version / Medicine / Master / Doctor of Medicine

Nine physiological measures were evaluated pre- to post-intervention on subjects participating in a university health promotion program over a seven-month period. Frequency of program attendance and choice of activity were also assessed. Of the 88 employees initially screened, most of the subjects were staff members (n=82, 93%),with a majority being female (n=68, 77%). Significant differences in physiological measures were found pre- to post-intervention between "higher" and "lower" cardiovascular risk participants, primarily due to the type of activity chosen. .The results indicate that health promotion programs at a university are an effective way to have an impact on employees in reducing their cardiovascular risk factors.

This dissertation had three primary aims. The first aim was to systematically review evidence documenting incidence of sexually transmitted infections (STI) during pregnancy. Eighteen papers were included in the final review which reported incidence of five STIs: chlamydia, gonorrhea, human papillomavirus (HPV), herpes simplex virus type 2 (HSV-2) and human immunodeficiency virus (HIV). The review found that there are very limited data on incidence of STIs during pregnancy and even fewer data comparing risk between pregnant and non-pregnant women. Although data are limited, studies suggest that women continue to acquire STIs during pregnancy, with incidence varying by type of infection, population of interest and geographic setting. Highest incidence was found for HPV and chlamydia although some studies of chlamydia showed low proportions of pregnant women infected. Studies in which partners were known to be infected with HSV-2 and HIV showed higher rates of acquisition in pregnant women compared to studies where partner status was not known. The second aim of this dissertation was to describe the impact of pregnancy on behavioral risk factors and vaginal practices that are associated with increased risk of STI acquisition. Data for this and the following aim came from the Methods for Improving Reproductive Health in Africa (MIRA) study, a randomized clinical trial conducted in South Africa and Zimbabwe 2003-2006. The analysis for the second aim included women in the MIRA trial who had a pregnancy during follow-up. Pregnancy was found to decrease sexual activity, particularly in the third trimester, but women were more likely to report sex without condoms while pregnant. There were lower reports by women during pregnancy of other risk factors for STI acquisition, including anal sex, concurrent sexual relationships and new sex partners. Vaginal wiping and insertion of material into the vagina, potentially important mechanisms for STI acquisition, were also less common during pregnancy. The data from this aim present a complicated picture of risk for STIs during pregnancy as a result of increased unprotected sex but decreased frequency of other known behavioral risk factors. The third and final aim of the dissertation was to measure incidence of four STIs in pregnant and non-pregnant women and to evaluate whether women are at greater risk during pregnancy for acquiring four STIs: chlamydia, gonorrhea, trichomoniasis and HIV. This analysis included 4,549 women 18-50 years of age, 17% (N=766) of whom had a pregnancy during follow-up. In general, women continued to be sexually active but reported less overall sex than non-pregnant women. Report of condom use was lower during pregnancy as were other types of high risk sexual behaviors, such as multiple sexual partners, sex in exchange for drugs or money and anal sex, as well vaginal practices. STI incidence was measured during pregnancy and it was found that women continued to acquire STIs when pregnant. In addition, during periods when women became pregnant, they appeared to be a high risk for acquiring chlamydia, trichomoniasis and HIV. Finally, in examining the association between pregnancy status and STI risk, we found that in multivariable models adjusted for demographic and time-varying self-reported behavioral risk factors and vaginal practices, pregnancy was not associated with increased STI risk. However in visit intervals when women became pregnant, they appeared to be at higher risk for contracting chlamydia compared to non-pregnant periods.

Many studies have shown that short sleep duration and/or poor sleep quality is associated with increasing rates of cardiovascular (CVD) mortality and morbidity. One hypothesized explanation for this association has been that sleep loss is a type of chronic stress that induces dysregulation of biological systems that ultimately increase the risk of CVD. One biological system that has been thought to link sleep loss and CVD is the hypothalamus-pituitary-adrenal (HPA) axis. A number of studies using small or convenience samples have addressed the effects of sleep deprivation on cortisol. Only a few studies have examined the association of habitual short sleep duration and/or poor sleep quality with changes in the diurnal cortisol in population based-samples; those studies vary in their methodology and in findings. Another biological system that has been thought to link sleep loss and CVD is the autonomic nervous system (ANS), through overactivation of the sympathetic nervous system (SNS) and/or probably a withdrawal of the parasympathetic nervous system. Experimental studies have shown an association between the sleep stages and markers of the sympathetic system. However, very few studies of habitual sleep duration/sleep quality and ANS markers have been conducted. Even fewer studies have examined the association of habitual sleep duration and/or sleep quality and ANS responses to a stress challenge in a population-based sample. The findings again have been inconsistent probably due to the use of different methodology and different samples. This dissertation used measures of salivary diurnal cortisol as well as cortisol responses to a stress challenge protocol to assess the relationship of habitual sleep duration and/or sleep quality with diurnal cortisol profile in natural conditions and in response to a stress challenge protocol in a laboratory setting. Diurnal cortisol was assessed from up to 16 samples of salivary cortisol for two days. Cortisol responses to a stress challenge were assessed from four salivary samples taken during the stress challenge that was performed in a different day than the diurnal cortisol collection. To examine the relationship of habitual sleep duration and/or sleep quality and markers of the ANS, this dissertation used continuous cardiovascular measures (heart rate and heart rate variability) and four salivary amylase samples obtained during the stress challenge. The stress challenge included mental stress and orthostatic stress. Sleep duration and sleep efficiency (an objective measure of sleep quality) were assessed from 7-day actigraphy and sleep diaries. Insomnia symptoms (a subjective measure of sleep quality) were also assessed using a questionnaire that included the Women’s Health Initiative Insomnia rating scale (WHIIRS). We used mixed models so as to account for the repeated measures of diurnal salivary cortisol levels as well as the responses (reactivity and recovery) to the stress challenge tests. Chapter 1 presents an introduction to this dissertation discussing the relationship between short sleep duration and/or poor sleep quality and CVD morbidity and mortality. Chapter 2 presents a systematic literature review of studies of the association between habitual sleep duration and/or sleep efficiency and markers of neuro-endocrine systems: HPA and ANS. These are plausible mechanisms that link short and/or poor sleep to CVD morbidity and mortality. Chapter 3 presents our analyses of the relationship between short sleep duration and/or poor sleep quality and features of the diurnal cortisol. We hypothesized that those participants whose slept < 6 hours per night or whose sleep efficiency was < 85% would have higher cortisol levels on awakening, flatter cortisol awakening responses (CAR), and higher evening cortisol levels than participants who slept longer or slept better. We found that short sleepers had higher evening cortisol than the longer sleepers and that this association persisted after the adjustment for several known confounders. In chapter 4, we examined how the same groups of participants responded in terms of hormones (cortisol and amylase) and cardiovascular indices (heart rate (HR) and HR variability (HRV)) to a stress challenge test. We hypothesized that those participants who slept for a shorter time or whose sleep was of poorer quality would have more exaggerated responses to and less recovery from a stress challenge test than participants who slept longer or slept better. We found that participants with insomnia had exaggerated high frequency-HRV (HF-HRV) orthostatic reactivity. In an extended analysis, we found that participants who slept less than 7 hours/night had exaggerated heart rate reactivity to a mental stress test compared to participants who slept 7 or more hours/night, but this association was attenuated after adjustment for naps. Paradoxically, we also found that participants who slept less than 7 hours had higher HF-HRV recovery from mental stress compared to longer sleepers (≥7 hours). Short sleep duration or low sleep efficiency was not associated with cortisol or amylase responses to the stress challenge protocol. These findings suggest that sustained high evening cortisol levels and cardiovascular responses to a stress challenge may be among the mechanisms linking short/poor sleep and CV disease.

Ng, Mei. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves xvi-xlviii). / Abstracts in English and Chinese. / ACKNOWLEDGEMENTS --- p.i / ABSTRACT / In English --- p.ii / In Chinese --- p.iv / LIST OF TABLES --- p.vi / LIST OF FIGURE --- p.viii / ABBREVIATIONS / For Units --- p.ix / For Prefixes of the International System of Units --- p.ix / For Terms Commonly Used --- p.X / Chapter CHAPTER 1 --- Overview of Childhood Obstructive Sleep Apnoea Syndrome (OSAS) / Chapter 1.1 --- Prevalence --- p.1 / Chapter 1.2 --- Clinical Features --- p.3 / Chapter 1.3 --- Definitions and Cutoffs --- p.4 / Chapter 1.4 --- Pathophysiology --- p.6 / Chapter 1.5 --- Risk Factors / Chapter 1.5.1 --- Gender --- p.8 / Chapter 1.5.2 --- Obesity --- p.9 / Chapter 1.5.3 --- Adenotonsillar Hypertrophy --- p.10 / Chapter 1.5.4 --- Genetic --- p.11 / Chapter 1.5.5 --- Atopic Diseases --- p.12 / Chapter 1.6 --- Complications / Chapter 1.6.1 --- Neurobehavioural Deficits --- p.13 / Chapter 1.6.2 --- Growth Defects --- p.14 / Chapter 1.6.3 --- Metabolic Disorders --- p.16 / Chapter 1.6.4 --- Systemic inflammation --- p.17 / Chapter 1.6.5 --- Cardiovascular Consequences --- p.19 / Chapter 1.7 --- Diagnosis --- p.20 / Chapter 1.8 --- Treatment / Chapter 1.8.1 --- Surgical Treatment --- p.22 / Chapter 1.8.2 --- Continuous Positive Airway Pressure (CPAP) --- p.24 / Chapter 1.8.3 --- Corticosteroids --- p.24 / Chapter 1.8.4 --- Leukotriene Receptor Antagonist --- p.25 / Chapter 1.8.5 --- Oral Appliances --- p.26 / Chapter 1.8.6 --- Weight Control --- p.27 / Chapter CHAPTER 2 --- OSAS and Cardiovascular Complications in Adults / Chapter 2.1 --- Mechanism / Chapter 2.1.1 --- Acute Cardiovascular Responses --- p.28 / Chapter 2.1.2 --- Chronic Cardiovascular Responses --- p.29 / Chapter 2.2 --- Hypertension / Chapter 2.2.1 --- Epidemiological and Clinical Data --- p.31 / Chapter 2.2.2 --- Characteristics --- p.32 / Chapter 2.2.3 --- Mechanisms --- p.33 / Chapter 2.2.4 --- Treatment --- p.34 / Chapter 2.3 --- Heart Failure --- p.35 / Chapter 2.4 --- Stroke --- p.37 / Chapter 2.5 --- Cardiac Arrhythmias --- p.39 / Chapter 2.6 --- Myocardial Ischemia and Vascular Disease --- p.41 / Chapter 2.7 --- Pulmonary Hypertension --- p.43 / Chapter CHAPTER 3 --- OSAS and cardiovascular complication in children / Chapter 3.1 --- Blood Pressure --- p.45 / Chapter 3.2 --- Ventricular Hypertrophy and Dysfunctions --- p.48 / Chapter 3.3 --- Heart Rate Variability --- p.50 / Chapter 3.4 --- Arterial Tone --- p.51 / Chapter 3.5 --- Endothelial Function --- p.51 / Chapter CHAPTER 4 --- Longitudinal follow-up study of children with OSAS - a cardiovascular perspective / Chapter 4.1 --- Introduction --- p.53 / Chapter 4.2 --- Methods / Chapter 4.2.1 --- Subjects and Study Design --- p.57 / Chapter 4.2.2 --- Polysomnography --- p.59 / Chapter 4.2.3 --- Ambulatory Blood Pressure Measurement --- p.61 / Chapter 4.2.4 --- Statistical Analysis --- p.62 / Chapter 4.3 --- Results / Chapter 4.3.1 --- Subject Characteristics --- p.64 / Chapter 4.3.2 --- Blood Pressure During Wakefulness --- p.71 / Chapter 4.3.3 --- Blood Pressure During Sleep --- p.76 / Chapter 4.3.4 --- Nocturnal Blood Pressure Dipping --- p.83 / Chapter 4.3.5 --- Blood Profile --- p.86 / Chapter 4.4 --- Discussion --- p.87 / Chapter 4.5 --- Conclusion --- p.99 / Reference List --- p.xvi

Au, Chun Ting. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves xxvii-lv). / Abstracts in English and Chinese. / ACKNOWLEDGEMENTS --- p.i / ABSTRACT / In English --- p.ii / In Chinese --- p.v / LIST OF TABLES --- p.vii / ABBREVIATIONS / For Units --- p.ix / For Prefixes of the international system of units --- p.ix / For Terms commonly used in the report --- p.x / STATEMENT OF WORK DONE --- p.xvi / Chapter CHAPTER 1 --- Overview of Childhood Obstructive Sleep Apnea Syndrome (OSAS) / Chapter 1.1. --- Clinical Features of Childhood OSAS --- p.1 / Chapter 1.2. --- Definition of Childhood OSAS --- p.2 / Chapter 1.3. --- Prevalence of Childhood OSAS --- p.3 / Chapter 1.4. --- Pathophysiology --- p.4 / Chapter 1.5. --- Risk Factors --- p.6 / Chapter 1.6. --- Diagnosis --- p.10 / Chapter 1.7. --- Treatment / Chapter 1.7.1. --- Tonsillectomy and Adenoidectomy (T&A) --- p.12 / Chapter 1.7.2. --- Continuous Positive Airway Pressure (CPAP) --- p.14 / Chapter 1.7.3. --- Corticosteroids --- p.15 / Chapter 1.7.4. --- Leukotriene Receptor Antagonist --- p.16 / Chapter 1.8. --- Complications of Childhood OSAS / Chapter 1.8.1. --- Growth Failure --- p.17 / Chapter 1.8.2. --- Neurocognitive Abnormalities --- p.19 / Chapter 1.8.3. --- Cardiovascular Abnormalities --- p.20 / Chapter CHAPTER 2 --- Cardiovascular Complications of OSAS in Adults (Literature Review) / Chapter 2.1. --- Acute Effects of OSAS on Cardiovascular System --- p.21 / Chapter 2.2. --- Chronic Effects of OSAS on Cardiovascular System --- p.23 / Chapter 2.3. --- Hypertension --- p.24 / Chapter 2.4. --- Heart Failure --- p.28 / Chapter 2.5. --- Pulmonary Hypertension --- p.30 / Chapter 2.6. --- Arrhythmias --- p.31 / Chapter 2.7. --- Cardiac Ischemia and Vascular Disease --- p.33 / Chapter 2.8. --- Stroke --- p.34 / Chapter CHAPTER 3 --- Cardiovascular Complications of Childhood OSAS (Literature Review) / Chapter 3.1. --- Blood Pressure --- p.37 / Chapter 3.2. --- Ventricular Structure and Function --- p.40 / Chapter 3.3. --- Arterial Distensibility --- p.42 / Chapter 3.4. --- Heart Rate Variability --- p.42 / Chapter CHAPTER 4 --- Ambulatory Blood Pressure in Children with OSAS / Chapter 4.1. --- Introduction --- p.44 / Chapter 4.2. --- Methods / Chapter 4.2.1. --- Subjects and Study Design --- p.46 / Chapter 4.2.2. --- Polysomnography (PSG) --- p.47 / Chapter 4.2.3. --- Ambulatory Blood Pressure Measurement (ABPM) --- p.49 / Chapter 4.2.4. --- Statistical Analysis --- p.50 / Chapter 4.3. --- Results / Chapter 4.3.1. --- Subject Characteristics --- p.52 / Chapter 4.3.2. --- Blood Pressure during Wakefulness --- p.55 / Chapter 4.3.3. --- Blood Pressure during Sleep --- p.57 / Chapter 4.4. --- Discussion --- p.62 / Chapter 4.5. --- Conclusion --- p.70 / Chapter CHAPTER 5 --- Cardiac Remodeling and Dysfunction in Children with OSAS / Chapter 5.1. --- Introduction --- p.71 / Chapter 5.2. --- Methods / Chapter 5.2.1. --- Subjects and Study Design --- p.72 / Chapter 5.2.2. --- Polysomnography (PSG) --- p.74 / Chapter 5.2.3. --- Conventional Echocardiography --- p.75 / Chapter 5.2.4. --- Tissue Doppler Imaging --- p.76 / Chapter 5.2.5. --- Statistical Analysis --- p.77 / Chapter 5.3. --- Results / Chapter 5.3.1. --- Study Population --- p.79 / Chapter 5.3.2. --- Polysomnographic Findings --- p.79 / Chapter 5.3.3. --- Echocardiographic Findings / Chapter 5.3.3.1. --- Right Ventricle --- p.81 / Chapter 5.3.3.2. --- Left Ventricle --- p.83 / Chapter 5.3.4. --- Treatment Effect --- p.86 / Chapter 5.4. --- Discussion --- p.90 / Chapter 5.5. --- Conclusion --- p.95 / Chapter CHAPTER 6 --- Conclusion --- p.96 / APPENDIX I Hong Kong Children Sleep Questionnaire (Chinese) --- p.xvii / APPENDIX II Hong Kong Children Sleep Questionnaire (English) --- p.xxii / REFERENCES --- p.xxvii

目的：針對心血管初級預防，世界各國均推薦某一年齡段人群全部測量膽固醇以估算心血管病發病風險。此舉耗費高且非必須，本研究旨在建立並驗證一個新型的选择性膽固醇篩查模型，用以篩查需藥物治療之高危人群，并在成本效益方面與其它篩查模型相比較。 / 方法：本模型具體采用兩步法：首先利用一個足夠高的假設膽固醇值代入心血管病風險預測方程，用以系統性的高估絶大多數人的心血管病風險；其次只有假設心血管病風險高於推薦治療閾值時，該個體才需要測量膽固醇，並進行實際心血管病風險分析。 / 英国健康调查和中国营养与健康调查是本次研究的合适数据。我們首先探索最優的假設膽固醇值，尋找到最後膽固醇值之後，我們將繼續測試我們的新型膽固醇篩查模型，在不同的治療閾值下，表現是否穩定。我們以靈敏度，特異度和徐篩查人群為指標，比較我們模型與全民篩查模型和英國NICE 選擇篩查模型相比較。之後我們估算在中英人群中應用該篩查模型，所需耗費的成本和可預防心血管事件數。 / 结果：與全名篩查模型相比，我們的模型靈敏度相若但可以節省80%左右的篩查費用。模型的靈敏度主要取決於所採用的假設膽固醇值，與所用風險預測方程，治療閾值和人群心血管風險分佈無關。當以均數加2 倍標準差作為假設膽固醇值時，靈敏度可達到97.5%左右，特異度可以達到90%左右，符合預期。模型應用於中國人群得到的結果類似。值得註意的是，在中國人群中，即使不測量膽固醇，模型靈敏度亦接近95%。此外，將膽固醇篩查項目限制于男性50-84歲，女性60-84 歲年齡段可以進一步減少篩檢費用。在人群影響方面，我們模型可預防心血管事件數比全名篩查模型略少，但成本大大降低。英國NICE 模型適用於某些特定情況，但並非全部。 / 結論：我們的新型篩查模型靈敏度與全民篩查模型相若，但可以節省大量篩查費用。在资源匮乏地区，可考虑在某一特定年龄段运用我们的模型已达到进一步减少费用的效果。如果本研究结果得到进一步数据证实，對於中國人群而言，膽固醇測量可能並非心血管風險評估所必須。 / Objectives / Since the mid 1990s, most guidelines on primary prevention of cardiovascular disease (CVD) have recommended regular cholesterol measurement for all adults or those above a certain age (which is known as mass screening). Cholesterol measurement comprises a large cost of CVD prevention and is not necessarily required in those who do not need drug intervention. In order to reduce this cost, we have developed a new selective cholesterol screening model in order to determine whose cholesterol should be measured for drug prevention. The model was evaluated and compared with other widely adopted models in basic model performance as well as cost effectiveness. / Methods / The new model has two steps. In the first step, we purposely over-estimated the majority of respondents’ CVD risk by substituting a sufficiently high hypothetical cholesterol value in the risk estimation. We then recommend cholesterol measurement only to those with the estimated CVD risk above a predetermined risk threshold for drug treatment. In the second step, the CVD risk is re-estimated based on the individual’s real cholesterol consentration. Those with a risk above the treatment threshold are recommended for drug treatment. / We evaluated the performance of our two-step model with data from the Health Survey for England and re-evaluated it with data from the China Nutrition and Health Survey 2002. By varying the hypothetical cholesterol values and treatment thresholds in CVD risk, we assessed the sensitivity, specificity and proportion of the population who need to measure cholesterol and compared it with the US mass screening model and the UK NICE selective screening model. We further compared the costs and CVD events avoided in the compared screening programmes. We also examined how the age restriction should be set in cholesterol screening programmes. / Results / As compared to mass screening, our new model can achieve a high sensitivity and save some 80% the cost of cholesterol measurements. The sensitivity depends mainly on the hypothetical cholesterol level used and seems independent of population’s CVD risk, treatment cut-off values and risk prediction model. The model performed well in almost all the conditions tested. When the hypothetical cholesterol was set at MEAN+2SD, the resulting sensitivity of our selective screening model was almost always above 95% and close to the expected 97.5%. The sensitivity was only compromised slightly if cholesterol is not measured at all for the Chinese population. Furthermore, in order to save more costs, cholesterol measurement could be better restricted to men aged 50-84 and women 60-84 years regardless of the screening model used. In CVD events prevented, mass screening is always the best but our model can prevent almost as many. In costs, mass screening is always the most expensive but our model can save all or most of the cost. The NICE selective model can perform as well as our model only when it is used in an appropriate manner and in certain circumstances. / Conclusion / Our new cholesterol screening model has a high sensitivity which is comparable to that of universal screening programs but can save most of the cost on cholesterol measurements. In where resources are particular sparse, our model can also perform well by applying it only to certain age groups, which will further save cholesterol measurement costs. Cholesterol measurement could even be completely avoided for the Chinese population if our findings can be re-confirmed correct with more updated data. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Hu, Xuefeng. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 114-121). / Abstract also in Chinese. / Abstract (in English) --- p.i / Abstract (in Chinese) --- p.iv / Acknowledgements --- p.vi / Abbreviations used in the thesis --- p.viii / List of Tables --- p.xvi / List of Figures --- p.xviii / List of Boxes --- p.xix / Chapter 1. --- Introduction --- p.1 / Chapter 1.1 --- The burden of cardiovascular disease --- p.1 / Chapter 1.2 --- Primary prevention of CVD --- p.2 / Chapter 1.3 --- The high-risk individual strategy for CVD primary prevention --- p.3 / Chapter 1.3.1 --- The high risk individual strategy is effective --- p.4 / Chapter 1.3.2 --- The high risk individual strategy is cost-effective --- p.4 / Chapter 1.4 --- Who should be treated with drugs? --- p.5 / Chapter 1.4.1 --- The single risk factor strategy --- p.5 / Chapter 1.4.2 --- The overall CVD risk strategy --- p.7 / Chapter 1.4.3 --- Scope of CVD primary prevention --- p.8 / Chapter 1.5 --- Methods for assessing the CVD risk --- p.9 / Chapter 1.6 --- Current strategies for cholesterol measurements --- p.10 / Chapter 1.6.1 --- United States National Cholesterol Education Program --- p.13 / Chapter 1.6.2 --- American Heart Association CVD and Stroke prevention guideline --- p.14 / Chapter 1.6.3 --- The U.S. Preventive Services Task Force guideline --- p.15 / Chapter 1.6.4 --- New Zealand guideline 2003 --- p.16 / Chapter 1.6.5 --- Australian guideline 2009 --- p.17 / Chapter 1.6.6 --- The Joint British Society guideline-2 --- p.17 / Chapter 1.6.7 --- UK Department of Health guideline on vascular check --- p.18 / Chapter 1.6.8 --- China Blood Lipid Modification Guideline 2007 --- p.18 / Chapter 1.6.9 --- Summary of the reviewed guidelines --- p.19 / Chapter 1.7 --- Rationale for a selective screening model --- p.20 / Chapter 1.8 --- The UK NICE model --- p.22 / Chapter 1.9 --- Objectives of this study --- p.24 / Chapter 2 --- Methods --- p.25 / Chapter 2.1 --- The new cholesterol screening model --- p.25 / Chapter 2.2 --- Framework for evaluating the new screening model --- p.27 / Chapter 2.3 --- Indexes for evaluating the basic performance of screening models --- p.28 / Chapter 2.3.1 --- Sensitivity, specificity and % need cholesterol measurement --- p.28 / Chapter 2.3.2 --- Sensitivity analysis for model performance --- p.29 / Chapter 2.3.2.1 --- Using different hypothetical cholesterol values --- p.29 / Chapter 2.3.2.2 --- Using different treatment cut-off thresholds --- p.30 / Chapter 2.3.2.3 --- Using different populations --- p.30 / Chapter 2.3.2.4 --- Using different risk equations --- p.31 / Chapter 2.4 --- Data --- p.31 / Chapter 2.4.1 --- The Health Survey for England --- p.31 / Chapter 2.4.1.1 --- Background and aim of the survey --- p.31 / Chapter 2.4.1.2 --- Survey design --- p.32 / Chapter 2.4.1.2.1 --- Sampling Frame --- p.32 / Chapter 2.4.1.2.2 --- Weighting variables --- p.33 / Chapter 2.4.1.3 --- Data collection --- p.33 / Chapter 2.4.1.3.1 --- Blood cholesterol --- p.34 / Chapter 2.4.1.3.2 --- Blood pressure --- p.34 / Chapter 2.4.1.3.3 --- Smoking --- p.34 / Chapter 2.4.1.3.4 --- History of CVD and diabetes --- p.34 / Chapter 2.4.1.3.5 --- Treatment history --- p.35 / Chapter 2.4.2 --- The 2002 China National Nutrition and Health Survey --- p.35 / Chapter 2.4.2.1 --- Survey design --- p.36 / Chapter 2.4.2.2 --- Data collection --- p.36 / Chapter 2.4.2.2.1 --- Blood pressure --- p.36 / Chapter 2.4.2.2.2 --- Blood cholesterol --- p.38 / Chapter 2.4.2.2.3 --- Smoking --- p.38 / Chapter 2.4.2.2.4 --- History of CVD, diabetes and drug treatment --- p.38 / Chapter 2.4.3 --- Subjects eligible for analysis in this study --- p.38 / Chapter 2.5 --- CVD risk prediction --- p.43 / Chapter 2.5.1 --- The Framingham risk equation for the UK population --- p.43 / Chapter 2.5.2 --- The Asian equation for the Chinese population --- p.44 / Chapter 2.5.3 --- Adjusting for cholesterol and blood pressure --- p.45 / Chapter 2.5.4 --- Deriving the hypothetical cholesterol --- p.46 / Chapter 2.6 --- Identifying the appropriate age ranges for cholesterol measurement --- p.47 / Chapter 2.7 --- Comparing various screening models and options --- p.47 / Chapter 2.7.1 --- Compared screening models and options --- p.47 / Chapter 2.7.1 --- Indices for the performance of the screening options --- p.49 / Chapter 2.7.2 --- Costs of different screening options --- p.50 / Chapter 2.7.2.1 --- Components of screening cost from societal perspective --- p.50 / Chapter 2.7.2.1.1 --- Cost for inviting people for data collection --- p.50 / Chapter 2.7.2.1.2 --- Cost for the full risk assessment --- p.51 / Chapter 2.7.2.1.3 --- Treatment cost --- p.51 / Chapter 2.7.2.1.4 --- Cost saved for avoided CVD events --- p.52 / Chapter 2.7.2.2 --- Components of screening cost from health system’s perspective --- p.52 / Chapter 2.7.3 --- Number of CVD events avoidable --- p.53 / Chapter 2.8 --- Statistical analysis --- p.54 / Chapter 2.8.1 --- Descriptive analysis --- p.54 / Chapter 2.8.2 --- Cross-tabulation analysis --- p.54 / Chapter 2.8.3 --- Survey data analysis --- p.54 / Chapter 3 --- Results --- p.57 / Chapter 3.1 --- Description of data --- p.57 / Chapter 3.1.1 --- The UK population --- p.57 / Chapter 3.1.1.1 --- Sumamry of CVD risk and risk factors --- p.57 / Chapter 3.1.1.2 --- Distribution of age --- p.57 / Chapter 3.1.1.3 --- Distribution of blood pressure and blood cholesterol --- p.58 / Chapter 3.1.1.4 --- Distribution of the predicted 10-year CVD risk --- p.62 / Chapter 3.1.1.5 --- Relation between the risk threshold and age --- p.63 / Chapter 3.1.2 --- The Chinese population --- p.65 / Chapter 3.1.2.1 --- Summary of CVD risk and risk factors --- p.65 / Chapter 3.1.2.2 --- Distribution of age --- p.65 / Chapter 3.1.2.3 --- Distribution of blood pressure and blood cholesterol --- p.66 / Chapter 3.1.2.4 --- Distribution of the predicted 10-year CVD risk --- p.69 / Chapter 3.1.2.5 --- Relation between the risk threshold and age --- p.70 / Chapter 3.2 --- Performance of our new screening model --- p.72 / Chapter 3.2.1 --- Performance according to cholesterol values in the UK population --- p.72 / Chapter 3.2.2 --- Performance according to treatment cut-offs in the UK population --- p.73 / Chapter 3.2.3 --- Performance according to cholesterol values in the Chinese population --- p.73 / Chapter 3.2.4 --- Performance according to the risk cut-offs in the Chinese population --- p.74 / Chapter 3.2.4 --- Performance using different risk equations --- p.76 / Chapter 3.3 --- Comparison with other existing screening models --- p.77 / Chapter 3.3.1 --- Performance of the 3 models within an age-restricted UK population --- p.79 / Chapter 3.3.2 --- Performance of the 3 models within an age-restricted Chinese population --- p.81 / Chapter 3.3.3 --- Performance of the 3 models in the entire UK population --- p.83 / Chapter 3.3.4 --- Performance of the 3 models in the entire Chinese population --- p.84 / Chapter 3.3.5 --- Costs of various screening options --- p.87 / Chapter 3.3.6 --- Number of CVD events avoidable of the screening programmes --- p.92 / Chapter 4 --- Discussion --- p.96 / Chapter 4.1.1 --- Performance at different hypothetical cholesterol values --- p.96 / Chapter 4.1.2 --- Performance at various treatment cut-off thresholds --- p.97 / Chapter 4.1.3 --- Performance with different risk equations --- p.98 / Chapter 4.1.4 --- Performance in different populations --- p.99 / Chapter 4.1.5 --- Performance with different survival functions --- p.99 / Chapter 4.2 --- Further modifications of the model --- p.100 / Chapter 4.2.1 --- A model without any cholesterol measurement --- p.100 / Chapter 4.2.2 --- Age restriction for selective models --- p.102 / Chapter 4.2.3 --- Our model with potential personalized treatment cut-off --- p.103 / Chapter 4.2.4 --- Three key things to ensure model performance in other population --- p.104 / Chapter 4.3 --- CVD events preventable --- p.105 / Chapter 4.3.1 --- Importance of age restriction --- p.105 / Chapter 4.3.2 --- Limitations of the NICE model --- p.106 / Chapter 4.4 --- Costs of different screening models --- p.107 / Chapter 4.4.1 --- Cost from different perspectives --- p.107 / Chapter 4.4.2 --- Cholesterol measurement cost and routine data collection --- p.108 / Chapter 4.4.3 --- Cost components --- p.109 / Chapter 4.4.4 --- Ways to reduce cholesterol measurement costs --- p.109 / Chapter 4.4.5 --- Costs and gain of the missing 2.5% high risk individuals --- p.109 / Chapter 4.5 --- Strengths and limitations of this study --- p.110 / Chapter 4.6 --- Recommendations --- p.113 / References --- p.114

Sociopolitical and economic factors shape the lived experience of immigrants and subsequent US-born generations. Often marked by immigrant-related federal and subfederal (i.e., state, county, and city) government policies, but also inclusive of public sentiment toward immigrants, an anti-immigrant climate limits Latino immigrants’ and US-born Latinos’ access to pro-health resources and services, keeps them in a lower socioeconomic position, increases their exposure to interpersonal and structural discrimination, and directly and indirectly exposes them to acute and chronic stressors that can take a toll on their cardiovascular health. The objective of this dissertation is to examine the association between anti-immigrant climate, first defined using policies and then defined using anti-Latino immigrant sentiment, and a panel of traditional and non-traditional cardiovascular disease risk factors among immigrant and US-born Latino adults living in the United States. This dissertation is organized into five sections: 1) an introduction, 2) a systematic review and critical analysis of the literature on US federal and subfederal policies and physical and mental health outcomes among Latino adults, 3) an empirical study of subfederal immigrant-related policies enacted in 2007 and their association with a panel of cardiovascular disease risk factors among Latino adult participants in the National Health Interview Survey, 4) an empirical study of anti-Latino immigrant sentiment during the 2016 Presidential campaign and election and a panel of incident cardiovascular disease risk factors in a cohort of Latino participants of the Hispanic Community Health Survey/Study of Latinos, and 5) a discussion of findings and implications for future research. The systematic review did not identify any studies of immigrant-related policies and traditional cardiovascular disease health condition risk factors of obesity, hypertension, high cholesterol, or diabetes. Exclusionary policies were associated with poor mental health and poor self-rated health and no relationship between policies and adverse birth outcomes was observed. In the empirical study of subfederal immigrant-related policies, I did not observe a statistically significant association between exposure to exclusionary policy climates in 2007 and a greater increase in the prevalence of cardiovascular disease risk factors relative to exposure to neutral/inclusive policy climates. Although no statistically significant difference-in-differences were observed, Latinos living in exclusionary states had a statistically significant increase in the prevalence of high alcohol consumption one year after exposure, while the prevalence remained unchanged among Latinos living in neutral/inclusive states. This increase was reflective of increases among foreign-born Latinos, not US-born Latinos. In the empirical study of anti-Latino immigrant sentiment during the 2016 Presidential campaign and election, findings from models of high depressive symptoms suggested that among Latinos of Mexican and Central American background, the exposed were more likely to have incident high depressive symptoms than the unexposed. Findings also suggested an association between exposure status and incident current alcohol use, particularly among the foreign-born. An inverse association between exposure and risk of hypertension was observed, with further differences by duration of US residence. Patterns for alcohol consumption across both empirical studies suggest that future studies should continue to explore the effect of anti-immigrant climate on acute changes in alcohol consumption among Latinos in the US. Additionally, findings from the systematic review and the second empirical study also support the continued study of the relationship between anti-immigrant climate and mental health outcomes. As the sociopolitical climate of the US becomes increasingly harsh toward Latino immigrants and their families, studies should examine other health outcomes in order to understand which dimensions of health are affected by exposure to an anti-immigrant climate among one of the largest ethnic populations in the US.

Includes bibliographical references. / 1 v. (various foliations) : ill. (some col) ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (Ph.D.)--University of Adelaide, Dept. of Community Medicine, 1994