Enhanced Data to Accelerate Complex Patient Comparative Effectiveness Research

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): It is our premise that easy to use data products will accelerate meaningful comparative effectiveness research (CER). Claims data are extremely difficult to use, requiring extensive experience to most appropriately aggregate these data to the patient level. Specialized expertise is also required to create meaningful variables such as treatments, covariates, and endpoints from claims data. We propose to develop and disseminate a large easy to use suite of analytical files and pre-coded algorithms to study comparative effectiveness of secondary prevention strategies among complex patients with cardiovascular disease (CVD). Adherence to CVD practice guidelines declines with age and this may be explained by uncertainty over the effectiveness of various secondary prevention strategies for the oldest old and those with multiple comorbidities. This collaboration of the University of lowa Older Adults Center for Education and Research on Therapeutics (lowa CERT) and Buccaneer Computing Systems & Services, Inc (Buccaneer) proposes to use the Medicare Chronic Condition Warehouse (CCW) as our major input source to create a longitudinal cohort of 1.8 million Medicare beneficiaries admitted with AMI or stroke/TIA in 2007 and followed through 2008 for recurrent events, complications and death. The large sample ensures power to test comparative effectiveness of secondary prevention strategies in priority subgroups with complex conditions, e.g. people over 80 years old with diabetes. At least 2 SAS(r) analytic data files will contain different levels of detail and aggregation for the same CVD cohort. From the more than fifty raw claims, drug event, and demographic (enrollment) data files in CCW, we will process the files to join the information for each beneficiary in our cohort over time and across all care settings. The analytical files will consist of: 1) patient-level aggregated data file(s) that summarize across care settings the care received by each beneficiary, and from this file we will create 2) summary treatment and outcome data files, to allow for rapid querying and reporting (e.g., for feasibility analyses and preliminary data analysis).The code and algorithms to create the data product are expected to streamline the development of future data products in new clinical populations for future CER studies. The federal CERT and CTSA networks are natural communication assets to disseminate information about the data product. We will test the data product in a research project focused on the uncertainty around statin effectiveness and safety in the aging patient with multiple comorbidities. PUBLIC HEALTH RELEVANCE: Cardiovascular disease (CVD) is the leading cause of death and over 80% of this is in people age 565, yet evidence-based guidelines are not aggressively followed, with the effect of co-morbidities not known. We will develop an easy to use data product for comparative effectiveness studies of 1.8 million Medicare heart attack and stroke patients and support its use through a collaboration between the lowa Older Adults Center for Education and Research on Therapeutics and the Institute for Clinical and Translational Science.