Project Summary

This research project is in progress. PCORI will post the research findings on this page within 90 days after the results are final.

One of PCORI’s goals is to improve the methods that researchers use for patient-centered outcomes research. PCORI funds methods projects like this one to better understand and advance the use of research methods that improve the strength and quality of comparative effectiveness research.

What is the project about?

A randomized trial is one of the best ways to learn if one treatment works better than another. In these trials, researchers assign patients to different treatments by chance. But it’s not always practical or ethical to assign patients to a treatment by chance. When randomized trials aren’t possible, researchers can use observational studies to learn how well treatments work. In observational studies, researchers observe what happens when patients and their doctors choose the treatments. But it can be hard for researchers to say for sure how well a treatment worked in an observational study.

To address this issue, researchers can use propensity score matching to determine how well treatments work in observational studies. Based on patients’ traits, these methods match patients who choose to take a new treatment with those who choose to take the usual treatment. In this way, these methods help an observational study mimic a randomized trial.

In this study, the research team is developing and testing new propensity score matching methods for use in observational studies with large data sets. The team is also developing new methods to match patients by their traits and the treatments they choose.

To address this issue, researchers can use propensity score matching to determine how well treatments work in observational studies. Based on patients’ traits, these methods match patients who choose to take a new treatment with those who choose to take the usual treatment. In this way, these methods help an observational study mimic a randomized trial.

In this study, the research team is developing and testing new propensity score matching methods for use in observational studies with large data sets. The team is also developing new methods to match patients by their traits and the treatments they choose.

How can this project help improve research methods?

Results may help researchers use propensity score matching methods with large data sets.

What is the research team doing?

The research team is developing new methods to

  • Improve matching for observational studies with large amounts of data
  • Estimate how well a treatment works for different groups of patients
  • Estimate how well a treatment works when patients have many treatment options

The research team is testing these methods using simulated data and mental health data from the Veterans Health Administration. Finally, the team is developing free software and online video tutorials to make the new methods widely available to other researchers.

Research methods at a glance

Design Element Description

Goal

  • Develop new statistical methods that encompass and improve fundamental aspects of matching with large data sets.
  • Develop a new matching approach to estimate heterogeneity of treatment effect by building optimal matched samples for specific patients that allow for a determination of which treatment is best for that patient’s medical condition.
  • Develop new matching methods for estimating treatment effects for interventions that have many options to allow assessment of treatment provider and setting quality.

Approach

Propensity score matching; simulation

Project Information

Jose R. Zubizarreta, PhD
Harvard Medical School
$1,048,691

Key Dates

November 2019
January 2024
2019

Study Registration Information

Tags

Award Type
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 4, 2022