United Press International - July 24, 2000

WASHINGTON, July 24 (UPI)-- A new study found that two of the world's most prestigious medical journals publish relatively few studies each year on disease prevention.

      But the editors of the two journals studied questioned the authors' methods and said they would publish more prevention articles if they received more of them in the first place.

      In the article, Dr. Steven H. Woolf, and Robert E. Johnson reviewed 1, 159 clinical articles published in either The New England Journal of Medicine or the Journal of the American Medical Association during 1998. They divided the articles up into 12 areas, including basic science, epidemiology, primary prevention, diagnosis, and treatment.

      The authors found that, of the 1,159 articles, only 71, or 6 percent, dealt with the primary prevention of disease. Only 29 articles (3 percent) dealt with screening for disease. They also found that not all of the articles in those two areas were of great use to practicing physicians. For example, four of the articles on primary prevention of tobacco use dealt with reducing exposure to environmental tobacco smoke, and with anti-smoking legislation and advertising. No article addressed how the clinician could help prevent youth from starting smoking.

      "Given the prominence of journals as a leading information source for physicians and clinical trainees, the attention they give to primary prevention signals its importance in patient care," the authors wrote. "Clinicians rely on journals to disseminate research advances. If this knowledge is unreported in widely read journals, most clinicians and trainees are left uninformed about the science base for primary prevention and screening."

      The article was published in the American Journal of Preventive Medicine.

      JAMA editor Dr. Catherine D. DeAngelis, said she was baffled by the researchers' methods.

      "They arbitrarily chose some method of categorizing articles from NEJM and JAMA, and then they counted them up," she said. "I don't know what criteria was used to categorize the articles. So we had X number of articles in dermatology; so what?"

      If the researchers wanted to prove that there were too few prevention and screening articles published in these two journals, she said, then they should also have tested their categories against two other journals on either end of the spectrum - for example, Science magazine, which publishes entirely basic research and no prevention or screening studies; and the American Journal of Public Health, which publishes lots of prevention and screening studies - and see how they came out. And they should have gotten more than just the authors to assign the categories, she added.

      "You've got to have a model set up that can be used by a number of individuals, so that ten people reading the same articles would categorize them the same way."

      DeAngelis noted that JAMA receives about 4,000 article submissions a year and publishes about 10 percent of them; most of the articles they receive are not on prevention.

      "I'm an academic general pediatrician, and I'm extremely interested in clinical relevancy and prevention," she said. "But there isn't much out there in terms of articles that show, `Here is what you do for prevention, and it works.' You do what you can with the information you have."

      New England Journal editor-in-chief Dr. Jeffrey M. Drazen, said that his publication also tends to favor good prevention studies.

      "About 80 percent of our readers are general internists or internal medicine subspecialists," he said. "Internists are very interested in prevention, so when we see a good prevention study, we have a bias toward it."

      But few such studies ever come into the journal, Drazen said. He noted that the study showed that NEJM published a total of 31 articles on prevention and screening. Since the journal generally publishes about 5 percent of the articles that are submitted to it, it would logically follow that NEJM should receive about 600 submissions related to prevention and screening each year - or about 10 such studies every week.

      "I've been here six weeks, so I should have seen about 60 of these papers if they come in evenly," he said. "I don't think I've even seen one. So I think we're accepting a much higher proportion of prevention and screening articles than we are of other stuff."

      Wolf, professor of family medicine at the Medical College of Virginia, in Richmond, didn't dispute the editors' criticisms. "All those points are well-taken," he said. "Our intent was to get the ball rolling to develop a methodology so that further analyses could be done."

      On the other hand, he continued, "the results we did get are nonetheless pretty striking. Why is it that such a small proportion of articles deal with these issues?"

      If it's because there aren't many submissions, "we have to explore why," Woolf said. "Is it because primary prevention research is of insufficient quality, or does it have more to do with the 1/8lack of3/8 interest by the National Institutes of Health and other organizations in funding research into these issues?"

      He noted that the NIH is organized into disease-specific institutes, so research there tends to be oriented around specific diseases.

      "But some prevention measures, such as eating a healthy diet, are not disease-specific interventions. A healthy diet may prevent both a heart attack and a stroke, but when you play the funding game, you have to make it look like it relates to a certain disease to get an institute's interest."

      Dr. Douglas B. Kamerow, is the director of the Center for Practice and Technology Assessment at the federal government's Agency for Healthcare Research and Quality (AHRQ), in Rockville, Md. Kamerow said he was not surprised by the study's results.

      "It reflects the fact that there are generally more articles about treatment and fewer articles about prevention and behavior change."

      AHRQ would like to see more prevention and screening articles published in the major journals, he said. The agency has a project called "Translating Research Into Practice" in which it sponsors research aimed at making clinicians and hospitals change their behavior - like getting more doctors to ask patients whether they smoke.

      But changing anyone's behavior, whether clinician or patient, is difficult, he added. "If it were easy, it would have been done a long time ago."

(C) 2000 UPI All Rights Reserved.