Novel Methodologies in Assessing Benefits of Quality Improvement Initiatives

By Rebecca Vigen, M.D.

Assistant Professor of Internal Medicine

Quality improvement (QI) in health care is a rapidly evolving field with the goal of standardizing processes and structures to reduce variation and improve patient health outcomes. The Quality of Care and Outcomes Research Council dedicated a session at #AHA23 focused on achieving these objectives. My talk was one of four in this session that focused on novel methodologies for evaluating quality improvement interventions.

Traditional QI methodologies, including pre-post studies that compare differences in outcomes before and after an intervention, are limited by their inability to control for secular trends and are frequently not generalizable to other health care systems. I discussed alternative trial designs that can overcome some of these limitations.

Analyzing pre-post studies with an interrupted time series analysis and using a control group can better account for secular trends when evaluating a pre-post study. Another methodological improvement is to assess quality improvement interventions in randomized trials, a design which balances risk factors between the treated and untreated groups. The electronic medical record can be used to identify and even randomize patients. For many interventions where a patient-level intervention might result in contamination between control and intervention groups, a cluster randomized trial can be used. Additionally, for interventions in which the health care system intends to deploy the project in many units, the more complex stepped-wedge design may be appropriate.

Overall, these alternative methodologies have numerous advantages over traditional approaches. Incorporating such strategies in study design will provide a higher degree of confidence when assessing putative benefits of QI initiatives.