Methodology and analytic techniques used in clinical research: Associations with journal impact factor

OBJECTIVE:: To describe research methodology and statistical reporting of published articles in high-impact-factor general medical journals compared with moderate-impact-factor obstetrics and gynecology journals. METHODS:: A cross-sectional analysis was performed on 371 articles published from January to June 2006 in six journals (high-impact-factor group: Journal of the American Medical Association, The Lancet, the New England Journal of Medicine; moderate-impact-factor group-American Journal of Obstetrics & Gynecology, British Journal of Obstetrics and Gynaecology, and Obstetrics & Gynecology). Articles were classified by level of evidence. Data abstracted from each article included number of authors, clearly stated hypothesis, sample size/power calculations, statistical measures, and use of regression analysis. Univariable analyses were performed to evaluate differences between the high-impact-factor and moderate-impact-factor groups. RESULTS:: The majority of published reports were observational studies (50%), followed by randomized controlled trials ([RCTs] 24%), case reports (14%), systematic reviews (6%), case series (1%), and other study types (4%). Within the high-impact-factor group, 35% were RCTs compared with 12% in the moderate-impact-factor group (relative risk 2.9, 95% confidence interval 1.9-4.4). Recommended statistical reporting (eg, point estimates with measures of precision) was more common in the high-impact-factor group (P<.005). CONCLUSION:: The proportion of RCTs published among the high-impact-factor group was three times that of the moderate group. Efforts to provide the highest level of evidence and statistical reporting have the potential to improve the quality of reports in the medical literature available for clinical decision making.