News Media Often Omit Potential Sources of Bias in Medical Research
Laurie Barclay, MD
October 1, 2008 — When reporting on medical research, news media often omit potential sources of bias, according to the results of a study reported in the October 1 issue of JAMA.
"The news media [are] not doing as good a job as [they] could be of informing readers of an important source of bias in medical research," lead author Michael Hochman, MD, a resident physician at the Cambridge Health Alliance and Harvard Medical School in Cambridge, Massachusetts, told Medscape Medical News. "News articles frequently refer to medications by their proprietary brand names, rather than the generic names.... The vast majority of news organizations do not have formal written policies stating that all stories about medical research must indicate how the research was funded or that the articles must refer to medications by their generic names."
Although news media are an important source of information regarding medical research for many patients, and even for some clinicians, little was previously known about how often news articles report funding of drug research by pharmaceutical companies or about how often news articles refer to drugs by their generic vs their brand names.
The goals of this study were to evaluate the reporting of pharmaceutical company funding and use of generic medication names in lay news articles about pharmaceutical research and to determine the views and policies of newspaper editors concerning these issues.
The investigators reviewed US news articles from newspaper and online sources about all pharmaceutical company–funded medication studies published in the 5 most prominent general medical journals (JAMA, the New England Journal of Medicine, The Lancet, the Archives of Internal Medicine, and the Annals of Internal Medicine) between April 1, 2004, and April 30, 2008. Editors of the 100 most widely circulated newspapers in the
Of the 306 news articles thus identified (of which 175 were from newspapers and 131 were from online sources), 130 (42%; 95% confidence interval [CI], 37% – 48%) did not report that the research was funded by a pharmaceutical company. This percentage did not differ significantly between print or online articles. Even when this information was reported, it was seldom positioned prominently in the text, suggesting that those readers who rely on the news media as their primary or sole source of information about medical research may remain unaware of how the research was funded.
Of the 306 articles identified, 277 reported on drugs that have both generic and brand names, and 186 of these articles (67%; 95% CI, 61% – 73%) referred to the study medications by their brand names at least half of the time.
"When patients and doctors read about brand name medications in the lay media, they immediately think these drugs are better than the generic options, [because] the news articles are often upbeat, interesting, and very favorable to new developments," Dr. Hochman said. "We doctors frequently prescribe expensive brand name drugs even though we know that less expensive and better tested generics are often at least as good, if not better.... For this reason, it is critical that news articles identify sources of bias in medical research, for example, company funding, that might cause readers to think about the results in a different light."
Of the 93 newspaper editors who responded to the survey, 82 (88%; 95% CI, 80% – 94%) stated that articles in their publication always or often indicated when studies had received company funding, and 71 of 92 (77%; 95% CI, 67% – 85%) also stated that articles in their publication always or often referred to medications by their generic names.
Despite this perception, only 3 (3%) of 92 newspapers (95% CI, 1% – 9%) had written policies requiring reporting of pharmaceutical company funding of medical studies. Only 2 (2%) of 93 (95% CI, 1% – 8%) newspapers had written policies requiring that medications be referred to by their generic names.
Study limitations include that some of the editors surveyed worked for publications not included in the analysis of news articles; the analysis only covered studies published in the 5 highest-impact medical journals; only journalists from the most widely circulated publications were surveyed, limiting generalizability; and these findings could not be extrapolated to news media reports from television and radio sources.
"I think the study used adequate methods," Michael Steinman, MD, an assistant professor of medicine in geriatrics at the San Francisco VA Medical Center and University of California–San Francisco, told Medscape Medical News when asked for independent comment. Dr. Steinman has published significant research on the role of the pharmaceutical industry in medicine, and his work was cited in Dr. Hochman's study.
"Studies funded by commercial entities are much more likely to reach favorable conclusions than those sponsored by [the National Institutes of Health] or nonprofit foundations," Dr. Steinman said. "Thus, information about the funding source is essential for interpreting a study's findings — not because industry support automatically invalidates a study but because the presence of financial conflicts of interest should lead the reader to take an extra careful look to ensure that the methods and conclusions were fair and reasonable."