• A J Singer, C S Homan, M J Stark, M C Werblud, H C Thode, and J E Hollander.
• Department of Emergency Medicine, State University of New York, Stony Brook 11794-7400, USA. asinger@epo.som.sunysb.edu
• Acad Emerg Med. 1997 Dec 1;4(12):1153-8.

IntroductionAs the specialty of emergency medicine (EM) matures, its journals should be publishing research of a quality similar to that which appears in other premier journals.ObjectiveTo compare the types of original research published in 4 EM vs 3 non-EM journals.MethodsRetrospective review of all 1995 articles published in Academic Emergency Medicine, American Journal of Emergency Medicine, Annals of Emergency Medicine, Journal of Emergency Medicine, Annals of Internal Medicine, JAMA, and New England Journal of Medicine. Research articles were classified as longitudinal vs cross-sectional, prospective vs retrospective, and interventional vs observational. Other characteristics noted were number of subjects, randomization, blinding, control, and power calculations. Journals were reviewed by 4 investigators who received specific training in research classification, adhering to previously reported criteria for retrospective reviews. Interobserver reliability was independently validated.ResultsThe authors reviewed 3,524 articles, of which 874 (24.8%) were original research. Compared with research reported in non-EM journals, EM journals contained fewer longitudinal studies (40.5% vs 60.4%, p < 0.0001) and fewer prospective studies (70.8% vs 78.7%, p = 0.008). Fewer EM journals had studies that were blinded (13.7% vs 18.9%, p = 0.047) or controlled (36.3% vs 50.0%, p = 0.003). Studies reported in EM journals had fewer subjects (138 vs 300, p < 0.001). Research reports in EM journals were less likely to have been funded, even after adjustment for the differences in study designs (adjusted odds ratio 7.0, 95% CI 5.1-9.7).ConclusionSignificant differences in types of research published in EM and non-EN journals were identified.

15 keywords...

hide…