Outcomes of Blood Pressure Management in Diabetes Patients with Comorbidities

描述（由申请人提供）：拟议研究的目标是调查糖尿病患者和并存健康状况的血压（BP）管理的预测因素、模式和结局，以便使用这些信息优化现有的BP指南。我们将使用一个关联的登记-索赔数据库来调查合并症的血压控制模式，评估血压和不良结局之间的关系，并为具有复杂健康状况的糖尿病患者制定血压管理指南原型。具体目标是：（1）评估2型糖尿病患者特定血管和非血管共病对血压模式随时间变化的影响。我们假设，血压的轨迹将是不同的糖尿病患者有特定的共存条件和不同的管理干预。（二）、评估糖尿病患者合并症负担、血压控制和血压水平变化与不良临床结局（卒中、缺血性心脏事件、住院和死亡）之间的相关性。我们假设，复杂的健康状况，而不是血压水平，将更有力地预测糖尿病患者的多种合并症和脆弱性的不良事件。然而，对于那些没有复杂健康状况的患者，我们假设血压水平将是不良结局的预测因子。(3)使用特定目标1和2的结果，开发指南原型，最大限度地降低复杂患者的不良事件概率。我们将使用关于特定合并症的BP轨迹的信息和来自预测不良临床结局的回归模型的数据，以生成将额外数据元素纳入复杂糖尿病患者现有BP指南的概率。这是一项分析性流行病学研究。它将使用包含糖尿病管理措施（HbA 1c，LDL水平，BP水平等）的临床糖尿病注册表的二级数据分析，这些数据将与健康维护组织和医疗保险的索赔相关联。医疗保险索赔可通过密歇根大学参与医师团体实践（PGP）医疗保险示范项目获得。该项目为UM分配了一群在UM接受护理的医疗保险患者;糖尿病登记处的大多数医疗保险患者都是分配的。参与者将是2005-2008年的约20，000例登记患者，其中约70%的患者在登记研究中有索赔和多次BP测量。本研究将利用这些关联登记数据，调查具有不同合并症负担的中老年糖尿病患者的血压水平。主要结局包括卒中、新发缺血性心脏事件、死亡和住院时间增加。作为糖尿病患者血压水平和临床结局的预测因子，最受关注的协变量是冠状动脉疾病、复杂的健康状况（定义为合并症负担）和年龄。本研究的结果将为糖尿病和复杂健康状况患者的血压管理提供实际的临床证据。

DESCRIPTION (provided by applicant): The goal of the proposed research is to investigate predictors, patterns and outcomes of blood pressure (BP) management among patients with diabetes and coexisting health conditions, in order to use this information to optimize existing BP guidelines. We will use a linked registry-claims database to investigate the pattern of BP control by comorbidities, assess the relation between BP and adverse outcomes, and develop guideline prototypes for BP management in diabetes patients with complex health status. The Specific Aims are: (1) To assess the effect of specific vascular and nonvascular comorbidities on BP patterns over time in patients with type 2 diabetes mellitus. We hypothesize that trajectories of BP will be different in diabetes patients who have specific coexisting conditions and different management interventions. (2). To evaluate the association between comorbidity burden, BP control and changes in BP level, and adverse clinical outcomes (stroke, ischemic cardiac events, hospitalization, and death) in diabetes patients. We hypothesize that complex health status, rather than BP level, will more strongly predict adverse events in diabetes patients with multiple comorbidities and vulnerabilities. For those patients without a complex health status, however, we hypothesize that BP level will be a predictor of adverse outcomes. (3) Using the results from Specific Aims 1 and 2, to develop guideline prototypes that minimize the probability of adverse events in complex patients. We will use information regarding the trajectories of BP for specific comorbidities and data from the regression models which predict adverse clinical outcomes, to generate the probabilities of incorporating additional data elements into existing BP guidelines for complex diabetes patients. This is an analytic epidemiology study. It will use secondary data analysis of a clinical diabetes registry containing measures of diabetes management (HbA1c, LDL level, BP level, among others) that will be linked to claims from a Health Maintenance Organization and Medicare. Medicare claims are available through the University of Michigan's participation in the Physician Group Practice (PGP) Medicare Demonstration Project. This project assigns UM a population of fee-for service Medicare patients cared for at UM; most Medicare patients in the diabetes registry are assigned. Participants will be the approximately 20,000 registry patients from 2005-2008, approximately 70% of whom will have claims and multiple BP measures on the registry. Using this linked registry data, this research will investigate blood pressure (BP) level among patients with diabetes from middle to advanced old age, with different comorbidity burdens. Major outcomes include stroke, new ischemic cardiac event, death, and increased hospitalization. Covariates of most interest as predictors of both BP level and clinical outcomes in diabetes are coronary artery disease, complex health status (defined as comorbidity burden) and age. Results of this research will provide practical clinical evidence for modifying BP management in patients with diabetes and complex health status.