Purpose: To identify self-perceived problems in managing science information needs of U.S. primary practitioners and their opinion leaders.
Sample: A two-stage random sample of the American Medical Association's "Masterlist of Physicians" to identify 625 office-based physicians, including general practice, family practice, internal medicine, obstetrics and gynecology, and pediatrics, and 100 physician opinion leaders.
Data collection: A formal telephone survey was completed for 79% of practitioners and 90% of opinion leaders sampled, using a pretested instrument applied by trained telephone surveyors.
Results of data analysis: Both practitioners and opinion leaders claimed that published reviews were the most useful means of identifying their information needs. When asked about use of six recent clinical advances, that is, "markers," for example, hemoglobin-A1C for diabetic control, as many as one fifth to one half were not using or were not aware of such an advance. Less than 1 in 3 practitioners personally searched the literature when information was needed; 2 in 3 claimed literature volume was unmanageable; 9 of 10 practitioners and opinion leaders assessed the scientific value of literature obtained, primarily from their own experience, with less than 1 in 10 contacting research methods specialists. Respondents suggested various innovations to better manage their science information needs in the future.
Conclusions: Primary practitioners require substantial help in meeting current science information needs. Increase in such resources as "validated reviews" or "expert networks" might help meet these needs.