Comparative studies are increasingly important as experts work to identify the most effective interventions for improving the quality of the U.S. health care system. The research standard for these studies is the randomized controlled trial (RCT), a model that can be impractical in terms of design, time, and cost. In this study, Harkness Fellow Atle Fretheim and colleagues examine the validity of a faster and less costly method for determining the effectiveness of a health care intervention.
What the Study Found
The researchers designed an intervention that used educational outreach visits and computerized reminders to encourage doctors to use older and less expensive antihypertensive drugs (i.e., diuretics) as a first-line therapy for patients instead of drugs that were newer and costlier but no more effective. The intervention proved successful. Results from the RCT indicated that use of diuretics doubled in the intervention group compared with the control group. The researchers then used an interrupted time-series analysis (ITS) to analyze data from just the intervention group and replicated the same findings. Using the ITS method, researchers collect data at multiple time points before and after an intervention to see if the intervention had an effect. This type of data is readily available because it is routinely collected in patient charts and for reimbursement purposes.
Randomized controlled trials can be difficult to conduct and their high cost and slow pace can impede their use in health systems research. Interrupted time-series analysis—which requires a smaller sample size, is less costly, and leads to faster results—was equally valid in determining the effectiveness of an intervention. The researchers note that their findings are based on just one case and will require replication in future comparative studies.