Enhanced Data to Accelerate Complex Patient Comparative Effectiveness Research, 2006-2009 [United States]
- Chrischilles, Elizabeth A. (University of Iowa. College of Public Health. Department of Epidemiology)
- Schneider, Kathleen (Schneider Research Assoicates, and Buccaneer, a General Dynamics Company)
- O'Donnell, Brian (Buccaneer, a General Dynamics Company)
- Lessman, Gregory (Buccaneer, a General Dynamics Company)
- Gryzlak, Brian (University of Iowa. College of Public Health. Department of Epidemiology)
- Wilwert, June (Buccaneer, a General Dynamics Company)
- Brooks, John (University of Iowa. College of Pharmacy)
- Robinson, Jennifer (University of Iowa. College of Public Health. Department of Epidemiology)
- Lund, Brian (University of Iowa. College of Public Health. Department of Epidemiology)
- Wright, Kara (University of Iowa. College of Public Health. Department of Epidemiology)
- Letuchy, Elena (University of Iowa. College of Public Health. Department of Epidemiology)
- Rudzianski, Nicholas (University of Iowa. College of Public Health. Department of Epidemiology)
United States Department of Health and Human Services. Agency for Healthcare Research and Quality
- Award Number: R24HS019440
acute myocardial infarction (AMI); beneficiaries; chronic illnesses; comorbidity; health; health care; medical care; medical records; Medicare; patient care; patients; research; statin; stroke
Purpose: Develop an easy-to-use data product to facilitate comparative effectiveness research involving complex patients.
Scope: Claims data can be difficult to use, requiring experience to most appropriately aggregate to the patient level and to create meaningful variables such as treatments, covariates, and endpoints. Easy to use data products will accelerate meaningful comparative effectiveness research (CER).
Methods: This project used data from the Medicare Chronic Condition Data Warehouse for patients hospitalized with acute myocardial infarction (AMI) or stroke in 2007 with two-year follow-up and one-year pre-admission baseline. The project joined over 100 raw data files per condition to create research-ready person- and service-level analytic files, code templates, and macros while at the same time adding uniformity in measures of comorbid conditions and other covariates. The data product was tested in a project on statin effectiveness in older patients with multiple comorbidities.
Results: A programmer/analyst with no administrative claims data experience was able to use the data product to create an analytic dataset with minimal support aside from the documentation provided. Analytic dataset creation used the conditions, procedures, and timeline macros provided. The data structure created for AMI adapted successfully for stroke. Complexity increased and statin treatment decreased with age. The two-year survival benefit of statins post-AMI increased with age.
Conclusion: Claims data can be made more user-friendly for CER research on complex conditions. The data product should be expanded by refreshing the cohort and increasing follow-up. Action is warranted to increase the rate of statin use among the oldest patients.
Data Access: These data are not available from ICPSR. The data cannot be made publicly available. Data are stored on University of Iowa College of Public Health secure servers, and may be used only for projects covered within the aims of the original research protocol and Centers for Medicare and Medicaid Services (CMS)-approved data use agreement. Data sharing is allowed only for research protocols approved under data re-use requests by the CMS privacy board. The CMS process for data re-use requests is described at Research Data Assistance Center (ResDac). Please note that as of May 2013, the DUA covering this work is set to expire February 1, 2014. Thereafter, per the terms of the DUA, datasets created for this project may not be available.
User guides are available from ICPSR for detailed descriptions of the data products, including a user guide for Acute Myocardial Infarction (AMI) Analytic Files and a user guide for Stroke and Transient Ischemic Attack (TIA) Analytic Files. Data dictionaries are available upon request. Please contact Nick Rudzianski (email@example.com or 319-335-9783) for more information.
Comparative effectiveness research (CER) using observational data sets is one way to harness very large sample sizes required to answer questions about subgroups of complex patients. The complex cardiovascular disease (CVD) patient is of particular concern. Over 38 million American adults over age 60 are living with CVD. Despite recent declines, CVD remains the leading cause of death in the United States. In 2006, coronary heart disease (CHD) caused one of every six deaths, and stroke was the third leading cause of death in both men and women. CVD is the leading cause of death after age 65, with 67 percent of CVD events in this age group. Several evidence-based drug therapies have been shown to reduce CVD risk, including statin therapy. Physicians frequently do not adhere to national cholesterol guidelines and they are often reluctant to aggressively treat the elderly with CVD, especially those =75 years of age. However, the degree to which co-existing conditions contribute to gaps in guideline adherence is not known. Moreover, virtually no clinical trial data are available from individuals =80 years of age for evidence-based guidelines in this age group.
The Enhanced Data to Accelerate Complex Patient Comparative Effectiveness Research project was funded through an Agency for Healthcare Research and Quality (AHRQ) R24 grant (R24HS01944D-01) in order to address these important gaps that remain in the literature. To accelerate the use of Medicare data by analysts, the University of Iowa (UI) and Buccaneer, A General Dynamics Company have produced a data product based on Medicare administrative data contained in the Centers for Medicare and Medicaid Service (CMS) Chronic Condition Data Warehouse (CCW). This data product, developed as a collaboration between UI CER scientists and Buccaneer methodologists, consists of an easy to use suite of analytical files and accompanying pre-coded variables and algorithms that compress over 100 raw Medicare data files into streamlined longitudinal person-level and summary service-level analytical files. These files are subject to the same data use agreement (DUA) requirements with CMS that the agency has for all research identifiable Medicare administrative files.
To create the data product, data for Medicare beneficiaries hospitalized with acute myocardial infarction (AMI), and/or treated in the Emergencey Department (ED) with stroke or transient ischemic attack (TIA) in 2007 were manipulated to produce analytic files for comparative effectiveness studies in patients with multiple chronic illnesses (referred to as complex patients). It is the hope of the investigators/developers that access to researcher-ready data files will enhance CER capacity. This expanded data capacity is housed in the UI Health Effectiveness Research Center and has been designed to serve as the porch for Iowa Clinical Translational Research Awards (CTSA) researchers, trainees, and community partners to engage in CER.
The current documentation describes the data cycle for patients with AMI (n=167,420), or stroke (n=271,314) and provides users with information that may be helpful in understanding and using the CER Accelerate AMI/stroke data files, analytic code templates, and corresponding documentation to achieve a wide range of research objectives. A separate document summarizes similar data files created for Medicare beneficiaries hospitalized with stroke or transient ischemic attacks (TIA) in 2007.
MethodsPresence of Common Scales: Charlson and Elixhauser comorbidity indices
AbstractDatasets: DS1: Enhanced Data to Accelerate Complex Patient Comparative Effectiveness Research, 2006-2009 [United States]
2006-01-01 / 2009-12-31Time Period: Sun Jan 01 00:00:00 EST 2006--Thu Dec 31 00:00:00 EST 2009
Is version of
Update Metadata: 2020-11-18 | Issue Number: 7 | Registration Date: 2015-06-16