By Elizabeth Bahm
Medill School of Journalism, Northwestern University
Journalists have new tools to decipher conflicts of interest in medical research, but there are still many obstacles to obtaining information, according to panelists at the Health Journalism 2010 conference in Chicago.
Steps have been taken on many fronts, from greater transparency on the part of researchers to stricter standards by editors of several major medical journals, but speakers at the April 23 panel, "Spotting conflicts of interest in medical research," agreed that journalists still face challenges in identifying bias.
Research has confirmed empirical evidence of bias, said Lisa Bero, a faculty member of the Institute for Health Policy Studies at the University of California, San Francisco. This bias has a "multiplier effect" as the initial bias in a study can influence later literature and even guidelines.
While more researchers are including disclosures, they may not include useful information. Bero said that disclosures need to summarize potential conflicts, include a decision on whether a conflict of interest exists, and provide a management strategy. She suggested that journalists ask targeted questions about the impact of sponsors on research, and check the incomplete, but growing UCSF drug industry document archive.
"There is a tremendous amount of manipulation going on," said Catherine DeAngelis, editor-in-chief of the Journal of the American Medical Association.
Since the late 1980s, a shift in direction at pharmaceutical companies has led to marketing divisions taking the reins from scientific ones, said DeAngelis. Since that time, the number of articles involving a conflict of interest has steadily risen to the point where there are now about 600 published annually.
DeAngelis said that the expansion of clinical trial registration is a new trend allowing some control for editors seeking to determine whether research is biased. She is part of a group of editors who have agreed not to publish trials unless they were registered in places such as clinicaltrials.gov. Still, with 15,000 biomedical journals out and only certain major journals holding to these protocols, there are still many studies published without useful data on potential conflicts.
John Fauber, a health reporter at the Milwaukee Journal Sentinel, described his experiences in reporting on conflicts of interest in research at the University of Wisconsin. Fauber described these types of stories as a "very ripe area," and the other panelists agreed that sources of bias such as researchers employed by pharmaceutical company speakers' bureaus or compensation in the form of stock options can easily lead to ideas for stories that patients have a right to know about.
While there are challenges in finding sufficient information on conflicts of interest, panelists also made it clear that even partial progress towards transparency still represents a significant advance for patients' sense of confidence.
"Even if it gives us another 20, 25 percent assurance, as a physician, I'll take it," DeAngelis said.
