Correspondence

Race, Sex, and Referral for Cardiac Catheterization

N Engl J Med 1999; 341:2021-2022December 23, 1999

Article

To the Editor:

I have read the article by Schwartz et al. (July 22 issue),1 and I am convinced by their arguments that the article by Schulman et al. in the February 25 issue2 did not present the best description of the data. I further agree that the three interpretations suggested by Schwartz et al. are reasonable alternatives, and I also like the idea of reporting risk ratios rather than odds ratios. However, I find myself wondering about the elephant in the back yard. Whether catheterization should or should not be prescribed as the treatment of choice is an excellent question but completely beside the point for this study. What is obvious from the data is that black women are treated differently. This finding makes for a different generalization, one that cannot be attacked by either those who are politically correct or the critics of political correctness. I suggest another guideline, in addition to those suggested by Schwartz et al.: first, report what the data most clearly show.

John Grima, Ph.D.
McKay-Dee Hospital, Ogden, UT 84403

2 References

1
Schwartz LM, Woloshin S, Welch HG. Misunderstandings about the effects of race and sex on physicians' referrals for cardiac catheterization. N Engl J Med 1999;341:279-283
Full Text | Web of Science | Medline

2
Schulman KA, Berlin JA, Harless W, et al. The effect of race and sex on physicians' recommendations for cardiac catheterization. N Engl J Med 1999;340:618-626[Erratum, N Engl J Med 1999;340:1130.]
Full Text | Web of Science | Medline

To the Editor:

The analysis by Schwartz et al. of what went wrong with the distorted media coverage of the article by Schulman et al. displays a common misunderstanding among medical researchers of the way the media works.

The research finding appeared to be that race and sex were an element — perhaps more precisely, a rather small factor — in medical decision making with regard to referrals for diagnostic procedures.

This kind of finding might be of interest to clinicians, but in itself, it is unlikely to interest the media. What will grip them is the perspective that doctors are sexist or racist. It is the incessant search for the attention-grabbing headline that doctors must consider before releasing their research findings to journalists, who are prone to view the world in such a peculiarly distorted way.

Research reports in journals are presented in an environment in which precision and caution are valued, but these are not features of the reporter's universe. Instead, reporters are engaged in a desperate, never-ending battle to beat the competition in ratings and circulation. It is the fear that their rivals will get the story first that pushes correspondents to make hasty appraisals and to sensationalize stories.

The self-flagellatory conclusion of Schwartz et al. that the authors of research reports must accept the blame for what goes wrong in the coverage of their studies is misplaced. The media do not simply report what they are told, as Schwartz et al. state. In my experience as a doctor working closely with the press-relations staff at our hospital, no matter how careful you are, the headline writers always lose the precise message somewhere between the interview and the layout desk. Instead, the answer lies in the cultivation of longer-term relationships between journalists and doctors, so that over an extended period, the complexity of medicine can be imparted, rather than a hurried briefing between strangers, which is what currently passes for medical information in the media.

Raj Persaud, M.R.C.Psych.
Maudsley Hospital, London SE5 8AF, United Kingdom

Author/Editor Response

The authors reply:

To the Editor: We agree with Dr. Grima that in the study by Schulman et al., black women were treated differently; the two black actresses were referred for cardiac catheterization less often than the two black actors, the two white actresses, or the two white actors. The importance of highlighting this sort of interaction was one of the main points of our critique. We disagree, however, with Dr. Grima's generalization from the two black actresses in the study to all black women. The lower rate of referral for the black actresses primarily reflected the physicians' response to a single actress — the 70-year-old black woman. Although race, sex, and age bias are all possible explanations for the different referral rate, it is also important to acknowledge the possibility of other, more innocent explanations (e.g., the older woman may have been a less convincing actress, her body language may have had an effect, or the difference may simply have been a matter of chance).

As Dr. Persaud points out, scientists and journalists can drive each other crazy. They work at different paces and speak to different audiences. It is no surprise that their interactions are sometimes unsatisfactory. Throughout our critique, we suggested that the misreporting of the results of the study by Schulman et al. reflected problems at multiple levels, starting with how the authors chose to present their findings, including the editorial review process, and ending with the media's reporting. Dr. Persaud seems to imply that Schulman et al. should not bear any responsibility for what went wrong. We strongly disagree. Had they reported actual referral rates and (appropriately) tempered their conclusions and comments to the press, we think the miscommunication could largely have been avoided.

Finally, although we recognize the inherent conflict between the media's goal of accurate reporting and that of selling newspapers, we think it important to point out that a number of press organizations have made some effort to set the record straight. In response to our article and the statement by the editors of the Journal that the “evidence of racism and sexism in this study was overstated,”1 a number of newspapers have run brief reports (generally in the Corrections section) alerting readers to problems with their original coverage. Unfortunately, the corrections have been far less prominent than the original stories. We were especially disappointed that ABC's Nightline, which dedicated an entire show to the article by Schulman et al., has chosen not to correct its original story.

Steven Woloshin, M.D.
Lisa M. Schwartz, M.D.
H. Gilbert Welch, M.D., M.P.H.
Veterans Affairs Medical Center, White River Junction, VT 05009

1 References