NLM logo

National Information Center on Health Services Research and Health Care Technology (NICHSR)

HSRProj (Health Services Research Projects in Progress)

Information about ongoing health services research and public health projects


Understanding treatment effect estimates when treatment effects are heterogeneous for more than one outcome
Investigator (PI): Brooks, John
Performing Organization (PO): (Current): University of South Carolina, Arnold School of Public Health, Department of Health Services Policy and Management / (803) 777-1627
(Past): University of Iowa / (319) 335-3500
Supporting Agency (SA): Patient-Centered Outcomes Research Institute (PCORI)
Initial Year: 2014
Final Year: 2019
Record Source/Award ID: PCORI/ME-1303-6011
Funding: Total Award Amount: $1,003,163
Award Type: Contract
Award Information: PCORI: More information and project results (when completed)
Abstract: Helping patients make patient-centered treatment decisions requires treatment effect evidence that is aligned to the circumstances of individual patients. Because real-world treatment choices often affect several outcomes, this evidence must include relationships between treatment choice and the array of consequences that can result from this choice. Randomized controlled trials (RCT) are usually insufficient to supply this evidence as RCTs describe the treatment effects for a single outcome and may not reflect the circumstances of many patients in the real world. In addition, if treatment effects vary across patients, it is impractical and probably impossible to generate sufficient RCT evidence for all patients in all circumstances. Analysis of observational data has been suggested as an alternative to find estimates of treatment effects across patient circumstances and across outcomes. Observational databases often afford large sample sizes, providing power to estimate treatment effects across patient subsets for alternative outcomes. Treatment variation in observation databases stems from patients and providers making treatment choices instead of randomized treatment assignment. If the benefits and risks associated with a treatment vary across patients and treatment choice is based on this variation, this has real implications on the treatment effect evidence that can be ascertained from analyzing observational data. Failure to interpret evidence of treatment effects from observational data properly could lead to treatment and policy mistakes. We propose investigating the properties of statistical estimators for use with observational data when treatment effects vary across patients for more than one outcome. First, we will use simulation modeling to assess the properties of these estimators under various relationships between treatment benefit and risk. Second, we will assess the effects of renin-angiotensin system antagonists (ACE/ARBs) on benefits and risks for patients post-stroke that also have a history of chronic kidney disease (CKD) using Medicare claims data from the Chronic Condition Warehouse. Third, we will perform chart abstraction for a sample of patients post-stroke to assess estimator assumptions and we will interpret our estimates in light of the estimator properties we uncovered via simulation and assumption validity.
MeSH Terms:
  • Chronic Disease
  • Computer Simulation
  • Health Policy
  • Humans
  • Medicare
  • Models, Statistical
  • Outcome Assessment, Health Care
  • * Randomized Controlled Trials as Topic
  • Renal Insufficiency, Chronic /therapy
  • Renin-Angiotensin System
  • Research Design
  • Risk
  • Statistics as Topic
  • Treatment Outcome
  • United States
Country: United States || United States
State: South Carolina || Iowa
Zip Code: 29208 / 52242
UI: 20143590
Project Status: Completed
Record History: ('2019: Project extended to 2019',) ('2018: Project extended to 2018.',) (' 2017: Project extended to 2017.',)