Are research studies sponsored by industry more likely to have favorable results and conclusions than studies without industry sponsorship?
Drug and medical device studies sponsored by the manufacturers tend to have more favorable effectiveness and harm findings and more favorable conclusions than studies with other funding sources. The source of funding should be considered when evaluating the implications of any clinical trial. (Strength of Recommendation: C, based on consensus, disease-oriented evidence, usual practice, expert opinion, or case series.)
Because industry-funded research has the potential to change practice guidelines, mechanisms to reduce bias, such as mandatory study registries and increasingly stringent limitations on conflicts of interest, have been instituted.1 Yet, industry-sponsored research still may have bias in areas such as study design, subject selection, and data analysis. In this Cochrane review, investigators assessed a total of 48 papers that evaluated original research studies for evidence of bias. The main outcomes of interest were favorable results and favorable conclusions in industry- vs. non–industry-sponsored studies.
When investigators evaluated the results of industry- vs. non–industry-sponsored studies, 14 papers that covered 1,588 studies were pooled for analysis. Industry-sponsored studies more often had favorable effectiveness results compared with non–industry-sponsored studies (relative risk [RR] = 1.24; 95% confidence interval [CI], 1.14 to 1.35). Three papers that evaluated the harm results of 561 studies also found that more favorable harm results were reported by industry-sponsored studies (RR = 1.87; 95% CI, 1.54 to 2.27).
Three papers that evaluated 151 drug studies looked at effectiveness results of trials sponsored by the test treatment manufacturer vs. those sponsored by the manufacturer of the comparator treatment. In two of the papers, studies sponsored by the manufacturer of the test treatment were more likely to favor that treatment (RR = 4.64; 95% CI, 2.08 to 10.32).
When investigators examined the conclusions of industry-sponsored vs. non–industry-sponsored studies, 21 papers that evaluated 120 medical device studies and 3,821 drug studies were pooled for analysis. The industry-sponsored studies had favorable conclusions more often than the non–industry-sponsored studies (RR = 1.31; 95% CI, 1.20 to 1.44). Similar results were seen within studies sponsored by test treatment companies vs. comparator treatment companies. Study conclusions were more likely to be favorable when the manufacturer of the test treatment was the sponsor than when a comparator was the sponsor (RR = 5.90; 95% CI, 2.79 to 12.49).
Industry sponsorship also appeared to influence how study results were interpreted. Five papers found less concordance between the study effectiveness results and conclusions among industry-sponsored studies (RR = 0.84; 95% CI, 0.70 to 1.01). For example, industry-sponsored studies were more likely than non–industry-sponsored studies to conclude that corticosteroids were safe.2 One possible reason for more favorable results among industry-sponsored studies is a high rate of builtin biases. Nine papers assessed potential risks of bias, including sequence generation, allocation concealment, blinding, and loss to follow-up. Four of the nine papers found no difference between industry- and non–industry-sponsored studies, and the remaining five papers found less risk of bias among industry-sponsored studies.
Given the expense of conducting clinical trials, industry sponsorship will likely continue to be a major source of funding for medical research. Practicing physicians should be cautious when evaluating the results of industry-sponsored studies and should also seek data from studies that are not sponsored by industry. These results highlight the need for objective, non–industry-sponsored summaries of recent research that help identify an overly positive spin of trial results by authors.