Weighing Clinical Trial Evidence with a Thumb on the Scale?
Posted by Osagie Obasogie November 2, 2009
Biopolitical Times
Several studies have shown that clinical trials sponsored by pharmaceutical companies and other for-profit entities tend to come to conclusions that favor the sponsoring body. While this association has been clear for some time, its cause has been less apparent. A number of factors could be involved; industry-sponsored clinical trials could simply produce fewer adverse events or test drugs that are more effective.
However, a recent study published in JAMA cuts through many of these possibly confounding factors to once again highlight the troublesome association between industry funding and clinical trial conclusions. Using logistic regressions to analyze the relationship between funding source and outcomes in 370 randomized drug trials (adjusting for many relevant factors such as sample size, quality of methods, etc.), the authors come to a startling yet statistically significant finding: the experimental drug was recommended in 16% of trials funded by nonprofits while being recommended in 51% of trials funded by for-profits. The authors did not mince words in their conclusion:
However, a recent study published in JAMA cuts through many of these possibly confounding factors to once again highlight the troublesome association between industry funding and clinical trial conclusions. Using logistic regressions to analyze the relationship between funding source and outcomes in 370 randomized drug trials (adjusting for many relevant factors such as sample size, quality of methods, etc.), the authors come to a startling yet statistically significant finding: the experimental drug was recommended in 16% of trials funded by nonprofits while being recommended in 51% of trials funded by for-profits. The authors did not mince words in their conclusion:
Conclusions in trials funded by for-profit organizations may be more positive due to biased interpretation of trial results. Readers should carefully evaluate whether conclusions in randomized trials are supported by data.
