Practicing internists accurately assess the methodological rigor of clinical trials but discount the findings of those that are industry-funded.

As clinicians peruse published clinical research, what factors affect the likelihood that they will believe the results? To address this question, investigators enlisted 269 board-certified internists (among 503 who were approached) to read abstracts of hypothetical clinical drug trials and to provide their assessment of the trial methods, their confidence in the results, and their willingness to prescribe the drugs. Each participant evaluated three scenarios — involving imaginary drugs for the treatment of hyperlipidemia, diabetes, and angina — with randomly selected combinations of two variables: methodological rigor (low, medium, or high) and disclosure of funding status (none, funding by NIH, or funding by a pharmaceutical company with financial involvement of the lead author).

A strong, direct association was found between methodological rigor and physician perception of study quality. The physicians also had more confidence in — and were more willing to prescribe the study medications from — the studies with stronger methods. By contrast, participants were less likely to consider studies to be rigorous, to have confidence in them, and to prescribe a medication based on their results if pharmaceutical companies funded them. Of the three types of disclosure, NIH funding elicited the greatest confidence and was the most likely to change practice.

Comment: These internists demonstrated a keen understanding of study methods, aptly discriminating among different studies based on design. They were also alert to funding, tending to give more credit to NIH-funded studies than to industry-funded ones. The author of an accompanying editorial decries this tendency, but events of the past have apparently given rise to skepticism among doctors. Pharmaceutical companies might instill more trust by more-systematic efforts to share data and methods.

Source:Journal Watch Cardiology