Does the medical literature contain the evidence to answer the questions of primary care physicians? Preliminary findings of a study

Proc Annu Symp Comput Appl Med Care. 1993:571-5.

Abstract

Objective: To assess the quality of information retrieved in MEDLINE searches performed to answer clinical questions of primary care physicians.

Setting: Ambulatory rural and nonrural practices.

Participants: Active, non-academic primary care physicians; professional medical librarians, a general internist.

Design: Descriptive study in 3 stages: 1) Office Interviews, to record questions that arise in primary care practice; 2) Online Searches, to locate information that might answer these questions; and 3) Critical Appraisal to determine the quality of the information contained in the articles retrieved.

Main results: Of 74 articles judged by clinicians to be relevant to their questions, 2 were critical reviews, 14 reported results of randomized controlled trials, and 4 reported results of cohort studies. The remainder contained weaker evidence: 1 reported results of a case-control study, 6 were longitudinal case series, 13 were cross-sectional descriptions of disease, 12 were cross-sectional descriptions of diagnostic testing, and 22 were review articles which did not meet criteria for a critical review.

Conclusion: Searching the medical literature to answer the questions of primary care physicians was successful in retrieving recent, relevant information, and often provided "a clear answer" to the question, but the cost of using bibliographic retrieval was high, and fewer than a third of the articles retrieved contained high quality evidence.

MeSH terms

  • Decision Making, Computer-Assisted*
  • Family Practice*
  • Humans
  • MEDLINE*