Most trials in primary care stand the test of time

According to a new paper in Family Practice, published by Oxford University Press, while medical practice is often undermined by subsequent investigation, randomized trials relevant to primary care generally hold up over time.  

Medical reversal describes a problem whereby new research leads doctors to stop using a popular medication, procedure or diagnostic test that had been implemented without a robust evidence base. Vinayak Prasad, a health researcher and associate professor at the University of California, San Francisco, has found that up to 46% of original studies on already adopted medical practices led to a reversal or shift in evidence of effect.

Evidence-based medicine is built on the premise that doctors can be confident when their decisions are grounded in high quality research. But decisions supported by robust evidence from randomized controlled trials can be reversed. For example, although aspirin is prescribed commonly to prevent cardiovascular disease, new studies indicate this treatment is unlikely to be effective.

Concerned about this practice, researchers here studied the extent to which evidence from randomized control trials relevant to primary care were contradicted in subsequent research. The authors studied 408 randomized controlled trials relevant to primary care published from 2002 to 2005. Over 12-17 years of follow up time, the researchers found only 35 occurrences of evidence reversal, or about two a year. The findings of about nine in ten of such randomized control trials were not reversed.

“Conclusions from randomized trials relevant to primary care that also meet criteria for validity are stable over time,” said Christian Ruchon, the lead author of this study.

The article, “Evidence reversals in primary care research: A study of randomized controlled trials” is available (at midnight on September 23, 2021) at: https://doi.org/10.1093/fampra/cmab104.

Direct correspondence to: 
Christian Ruchon
Department of Family Medicine, McGill University
858, chemin de la Côte-des-Neiges
Montreal, H3S 1Z1, CANADA
christian.ruchon@mail.mcgill.ca

To request a copy of the study, please contact:
Daniel Luzer 
daniel.luzer@oup.com