Barry HC, Ebell MH, Shaughnessy AF, Slawson DC, Nietzke F.
Department of Family Practice, College of Human Medicine, Michigan State University, East Lansing 48824-1315, USA. J Am Board Fam Pract 2001 Nov-Dec;14(6):437-42
BACKGROUND: Many physicians rely on the abstracts of research articles to guide their clinical decision making. This need for expediency is one basis for many journals to reformat their abstracts.
METHODS: To determine whether the format of medical abstracts affects physician decision making, we surveyed family physicians in Michigan, Pennsylvania, and Virginia. All participants were members of the American Academy of Family Physicians. The survey included three case scenarios (corneal abrasion, fibromyalgia, and hyperlipidemia) followed by structured and open-ended assessments of usual management.
After assessing their usual management in each scenario, the respondents were provided with an abstract of a valid research paper. The format of abstracts (unstructured, IMRAD [introduction, methods, results, and discussion], structured, and POEM [patient-oriented evidence that matters]) were randomly assigned. After reading the abstract, we assessed changes in management of the case scenario.
RESULTS: Two hundred eighty-nine family physicians responded to the survey. At baseline, 187 (65%) of physicians patched corneal abrasions. After reading the abstract, 142 (76%) would no longer use eye patches. Two hundred forty-five (83%) of physicians did not use the combination of fluoxetine and amitriptyline for managing fibromyalgia. After reading the abstract, 179 (73%) would use combination therapy. Two hundred thirty-four (84%) of physicians used “statins” when managing hyperlipidemia. After reading the abstract, 211 (90%) would continue using statins. The format of abstract had no significant effect on physicians’ decision making.
CONCLUSIONS: Whereas the format of abstract in this study had no effect on physician decision making, having valid information available in the context of a clinical scenario appeared to influence decisions.
PMID: 11757886 [PubMed – in process]