Uses and misuses of the STROBE statement: bibliographic study

Objectives Appropriate reporting is central to the application of findings fromresearch to clinical practice. The Strengthening the Reporting of ObservationalStudies in Epidemiology (STROBE) recommendations consist of a checklist of 22items that provide guidance on the reporting of cohort, case-control andcross-sectional studies, in order to facilitate critical appraisal andinterpretation of results. STROBE was published in October 2007 in severaljournals including The Lancet, BMJ, Annals of Internal Medicine and PLoSMedicine. Within the framework of the revision of the STROBE recommendations, theauthors examined the context and circumstances in which the STROBE statement was used in the past. Design The authors searched the Web of Science database inAugust 2010 for articles which cited STROBE and examined a random sample of 100articles using a standardised, piloted data extraction form. The use of STROBE inobservational studies and systematic reviews (including meta-analyses) wasclassified as appropriate or inappropriate. The use of STROBE to guide thereporting of observational studies was considered appropriate. Inappropriate usesincluded the use of STROBE as a tool to assess the methodological quality ofstudies or as a guideline on how to design and conduct studies. Results Theauthors identified 640 articles that cited STROBE. In the random sample of 100articles, about half were observational studies (32%) or systematic reviews(19%). Comments, editorials and letters accounted for 15%, methodologicalarticles for 8%, and recommendations and narrative reviews for 26% of articles.Of the 32 observational studies, 26 (81%) made appropriate use of STROBE, andthree uses (10%) were considered inappropriate. Among 19 systematic reviews, 10(53%) used STROBE inappropriately as a tool to assess study quality. Conclusions The STROBE reporting recommendations are frequently used inappropriately insystematic reviews and meta-analyses as an instrument to assess themethodological quality of observational studies.