Comparative Effectiveness of Treatment Combinations Post MI in the Elderly

University Of Iowa, Iowa City IA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Based on randomized controlled trial (RCT) evidence establishing the individual benefit of beta- blockers, renin-angiotensin system antagonists (ACE/ARBs),, HMG-CoA reductase inhibitors (statins), and antiplatelets, clinical guidelines recommend using each of these treatments indefinitely for secondary acute myocardial infarction (AMI) prevention. However, the benefits and risks of these treatments in combination are unclear and evidence describing the effects of these treatments for the elderly patients and patients with multiple comorbidities is sketchy. Wide treatment variation and the observed treatment risk paradox suggests that clinicians are hesitant to follow guidelines for many of their patients. Given the barriers to conducting additional RCTs, the analysis of observational data has been suggested to fill this evidence gap. Different estimation methods are available to analyze observational data that are derived from distinct assumptions and yield estimates with distinct interpretations. Making proper inferences from analysis of observational data requires an analytical framework that enables validation of the assumptions underlying these methods and recognizes the proper inferential context of estimates across estimation methods. The AHRQ Comparative Effectiveness Portfolio, as described in PA-09-070, has both clinical and methodological goals. Our proposed research uses an advanced study design to estimate the comparative effectiveness of alterative treatment combinations post-AMI. Our estimates will fill knowledge gaps in the care of post-AMI patients that will probably never be filled using RCT methods. The study design applies innovative methodological approaches to Medicare Part D data along with the use of primary data collection via chart abstraction to validate estimation assumptions. We will estimate the comparative treatment effectiveness of treatment combinations for secondary prevention post-AMI using both risk-adjustment and instrumental variable approaches in light of recent methodological insights into the correct interpretations of estimates from these methods. We will exploit the large number of Medicare patients from the CMS Chronic Condition Data Warehouse (CCW) to estimate the effects of specific treatments within treatment combinations, as well as estimating these effects in important patient subgroups. In addition, using chart abstraction data we will interpret our estimates in light of potential biases and provide bounds of treatment effects. This research will give clinicians estimates of the benefits and risks associated with each treatment combination by patient age and comorbidity status, and will also provide evidence for policy-makers to assess whether changes in treatment rates across treatment combinations are warranted.