When controversial medical studies are released, research organizations need to do a better job of clarifying the results with the media. Two recent media crises involving the New England Journal of Medicine (NEJM) and the American Psychological Association drive this point home with tough lessons learned.
Last month, NEJM admitted it published misleading results of a study, released in February, which cast physicians in a racist and sexist light and offered ways it should improve its communication to the public. The study claimed that the "race and sex of a patient independently influence how physicians manage chest pain" and triggered a national tidal wave of media coverage from most major newspapers and news programs like "Nightline" and CNN.
The media reported that blacks and women are "40 percent less likely" to be referred for cardiac testing than white men and implied that conscious or unconscious racial and gender biases affect the level of care American physicians provide. NEJM says a more accurate assessment of the findings, however, is that white women, white men and black men tend to be referred at the same rate and that only black women have a lower rate of referral.
The reporting was flawed, in large part, because of how the information was presented to the media, according to a media analysis headed by Dr. Gregory Curfman, NEJM's deputy editor. The most significant problem had to do with the way in which the authors chose to summarize their findings using an odds ratio instead of a risk ratio. Odds ratios are rarely used and are harder for the media to understand. In this case, the results were overstated and led to the media - and even the Surgeon General - equating odds ratios with probability.
Although NEJM published an article that clarified the study results in July, the clarifications did not generate nearly the media coverage the initial study did, which means the damage done to the image of physicians went largely uncorrected. Clearly, NEJM should have done more to set the record straight with the media through press releases and press conferences. To prevent future misunderstandings, it has developed guidelines to ensure media accuracy that include:
- Reporting absolute event rates;
- Converting odds ratios to risk ratios;
- Ensuring that comparisons are appropriate; and
- Questioning the desired outcome.
Implied Endorsements
The American Psychological Association is still rebounding from stinging accusations that it endorses sexual abuse because of an article printed in one of its journals last year. The article used a broad definition of sexual abuse and found that it is not as harmful as generally believed. The general media didn't notice the article until May 1999. But when the media did cover it, the APA found itself on the defense. Some in the advocacy and medical communities like the Family Research Council and Dr. Laura Schlessinger (of the popular nationally syndicated Dr. Laura Show) accused the APA of endorsing sexual abuse and questioned the integrity of its studies. These claims eventually caught the attention of House Majority Whip Tom DeLay, (R-Texas), who demanded an explanation.
This was a media crisis of nightmarish proportions. The APA immediately cranked out a press release and a letter to DeLay that strongly denied its endorsement of sexual abuse and clarified its position on the issue - publishing a study should not assume endorsement of the findings. The APA, which publishes thousands of studies targeted to researchers, never intended the study to be interpreted by public policy organizations, says Rhea Farberman, the APA's director of public affairs.
The crisis that ensued caused the APA to reevaluate its peer review process and the overall implications of its published studies. In addition to condemning sexual abuse of children, the APA's damage control involved:
- Asking APA journal editors to consider the social policy implications of articles on controversial topics; and
- Preparing a friend of the court brief that can be adapted for use by courts to refute any claims that the APA condones sexual relations between children and adults. (NEJM, Dr. Gregory Curfman, 617/734-9800; APA, Rhea Farberman, 202/336-5700)