Heterogeneity, Glycemic Control & Macro-Vascular Complications in Type 2 Diabetes

描述(由申请人提供)： 在研究纵向数据产生的复杂异质性(指差异效应或相关性)方面，我在稳健的统计方法方面拥有丰富的经验。我最近的工作表明，这种方法有可能促进对2型糖尿病(T2 DM)大血管并发症/心血管疾病(CVD)的研究。我的职业目标是在与设计和测试干预措施相关的统计方法方面发展一份独立的研究生涯，旨在减少T2 DM患者与糖尿病相关的心血管疾病。这项提议的主要目的是支持一项综合培训和研究计划，以实现这一目标。我的职业发展目标是：1)深入了解T2 DM的诊断、病因、病理生理学、治疗方法和相关的药理学理论；2)深入了解T2 DM相关的心血管疾病，尤其是其病因、诊断和预防理论；3)从卫生服务研究的角度深入了解可能影响提供者实施循证指南的因素以及患者对治疗和依从性的行为；4)提高我的统计技能，以开发有效和稳健的估计方法；5)增强我与临床医生、临床研究人员、资助机构和卫生政策分析师沟通定量方法和分析结果的技能。这些目标将通过关于糖尿病和心血管疾病研究的正式课程、二元指导、咨询、研讨会和会议以及贝叶斯统计方法来实现。对于拟议的研究，我将应用一般潜变量建模技术来检验在2型糖尿病患者的纵向临床队列中，由于对高血糖、血脂异常和高血压的药物调整而导致的血糖控制和心血管疾病之间的区别关系。为了确定预防心血管疾病的有效策略，我将进一步发展我的统计专业知识，通过改善血糖控制、高胰岛素血症或胰岛素敏感性来有效估计不同药物对心血管疾病风险的影响。这项研究将利用我在退伍军人管理局目前的研究环境中获得的丰富的纵向数据集，以及我导师的经验。这项研究的结果将提供进一步的见解，说明为什么该领域还没有就“严格的血糖控制”和心血管疾病之间的联系得出明确的结论，心血管疾病是2型糖尿病患者发病率和死亡率的主要原因。具体地说，结果将告诉我们，心血管疾病干预策略应该如何在不同特征的个人之间有所不同。

DESCRIPTION (provided by applicant): I have extensive experience in robust statistical methods to study complex heterogeneities (referring to differential effects or association) arising from longitudinal data. My recent work suggested the potential of such methods for advancing the research on macrovascular complications/cardiovascular diseases (CVD) in type 2 diabetes (T2DM). My career goal is to develop an independent research career in statistical methods pertinent to the design and testing of interventions aimed at reducing diabetes-related CVD for patients with T2DM. The primary purpose of this proposal is to support an integrated program of training and research to achieve this goal. My career development objectives are to: 1) acquire an in-depth understanding of T2DM regarding its diagnosis, etiology, pathophysiology, and treatment methods and the associated pharmacological theories;2) acquire an in-depth understanding of T2DM related CVD, especially regarding their etiology, diagnosis, and prevention theories; 3) gain insights from the health services research perspective about factors that may influence providers' implementation of evidence-based guidelines and patients' choices of treatment and adherence behaviors; 4) enhance my statistical skills for developing efficient and robust estimation methods; and 5) enhance my skills for communicating quantitative methods and analytic results with clinicians, clinical researchers, funding agencies, and health policy analysts. The objectives will be met through formal coursework, dyadic mentoring, consultation, seminars and conferences on diabetes and CVD research as well as in Bayesian statistics methodology. For the research proposed, I will apply the general latent variable modeling technique to examine the differential relationship between glycemic control and CVD due to medication adjustments for hyperglycemia, dyslipidemia, and hypertension in a longitudinal clinical cohort of type 2 diabetics. For the purpose of identifying effective strategies for prevention of CVD, I will further develop my statistical expertise to obtain efficient estimates of the differential medication effects on CVD risk, which are mediated through improving glycemic control, hyperinsulinemia, or insulin sensitivity. This research will take advantage of my access to a rich set of longitudinal data available in my current research environment in the VA and the experience of my mentors. Findings of the study will provide further insights about why the field has not yet reached a firm conclusion about the association between "tight glucose control" and CVD, the leading cause of morbidity and mortality in type 2 diabetics. Specifically, the results will inform how CVD intervention strategies should vary across individuals with different characteristics.