Information excess in medicine. Overview, relevance to dermatology, and strategies for coping.

BACKGROUND: The amount of biomedical information contained in scientific journals has grown to the point that complete coverage of this material is impossible. The number of articles and journals being published has been growing at an exponential rate since 1750. Thirty-four thousand references from 4000 journals are added each month to the National Library of Medicine MEDLINE database from among the more than 100,000 scientific journals now published. This increase in scientific literature reflects not greater productivity but simply a larger number of scientists at work. Most articles written are eventually published in the medical literature, but a large number of scientific articles are of less than optimal quality, and most scientific articles that are published are never cited. Articles of higher quality and usefulness cluster in a few journals that enjoy great use in medical libraries and high scores on bibliometric scores such as the Science Citation Index.

OBSERVATIONS: To assess the reading habits of dermatologic trainees and evaluate how they cope with this information excess, a survey was distributed to 36 residents in three training programs. The average number of medical journal reading hours per month was 17; the number of hours devoted to reading issues of the Archives of Dermatology, the Journal of the American Academy of Dermatology, and the Journal of Investigative Dermatology were 3.2, 5.0, and 1.4, respectively; and the average resident read seven peer-reviewed journals, two to four dermatologic tabloids, and four peer-reviewed medical journals.

CONCLUSIONS: Critically and consistently reading a limited number of high-quality peer-reviewed journals as well as taking advantage of information available through computer networks and bibliographic and full-text databases is the best approach to coping with the volume of medical literature. Translating this information into a change in attitude and modification of physician behavior is best accomplished when local role models incorporate new ideas into their practice and teaching. Modifications that would, in fact, bring about truly useful changes, such as decreasing the number of new publications, melding journals already present into smaller numbers of publications rather than instituting new journals, altering the "publish-or-perish" rules, and writing more comprehensive articles rather than multiple small contributions, all await fundamental alterations in long-accepted policies in medicine.