Publication Bias in Reporting Drug Efficacy

Most of us rely on the pharmaceutical industry to some extent for our health and well-being, whether it’s for an occasional round of antibiotics, a flu vaccination, or medication regularly taken for a chronic condition. The industry is regularly under fire for inflating drug prices, misleading or inappropriate advertisements, and concentrating research efforts on drugs that will elicit the highest profit, rather than on lifesaving treatments for rarer conditions. It’s easy to understand the disgruntled customer’s reaction to an exorbitantly priced pill for a necessary treatment, but it’s also not hard to rationalize that the drug companies must make sound economic decisions in order to successfully develop new lifesaving products, especially in light of the nature of the business, which is that a significant amount of the research and development budget is spent on drugs that simply don’t work as anticipated, or whose risks don’t outweigh the potential benefits.

It’s an ethically charged situation: the pharmaceutical industry does indeed profit from the illnesses and maladies of the general public, and they often hold the key to well-being, and indeed, survival, which means that the price on that key can be very dear. However, profitability in the industry attracts some of the best and brightest minds, without which, the key might never be discovered. Admittedly, there must exist a balance, and drug companies do negotiate pricing with governments and other agencies to ensure that lifesaving treatments are often available when they are needed, even to those who can’t afford the market value.

One of the crucial factors in profitability is marketability of the product. In the case of prescription drugs, the marketability depends upon the perception of efficacy by the prescribing entities, or the physicians. The medical community relies in large part on published results of clinical trials, so the nature and comprehensiveness of the body of published work about a new drug is critical to its success in the marketplace. Herein lies an ethical conflict for the drug companies: publishing perceptually negative results may negatively affect the likelihood that a drug is regularly chosen to treat a particular malady; however, not publishing available clinical results may artificially inflate the positive perception of a less-than-ideally-effective treatment.

A recent study published in the New England Journal of Medicine indicates that there may indeed exist a publication bias toward perceptually positive clinical trials. The study examined 74 industry studies reported to the FDA on twelve antidepressant drugs. The FDA evaluates all studies as positive, negative, or neither clearly positive nor clearly negative; according to the FDA evaluations, 51% of the conducted trials were positive. A thorough literature search revealed that 31% of the conducted trials never made it to publication; an analysis of the published studies revealed a 94% positive result.

The intention of this study was not to determine wrongdoing by the pharmaceutical companies or the publishing entities, but rather to determine whether a bias existed. As the results indicate that a bias does exist toward publishing positive data, or in making questionable studies appear positive, it is clear that more attention needs to be given to all clinical studies, and that the medical community may need to consider the positive bias in making treatment decisions.

Reference

Turner, E.H., Matthews, A.M., Linardatos, E., Tell, R.A., Rosenthal, R. (2008). Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy. New England Journal of Medicine, 358(3), 252-260. DOI: 10.1056/NEJMsa065779