PRM273 - METHODOLOGICAL CONSIDERATIONS ON THE RECOMMENDATIONS TO CONDUCT AND REPORT SYSTEMATIC REVIEWS IN MEDICAL LITERATURE: SYSTEMATIC REVIEW

Abstract

Several recommendations are available to help conducting and/or reporting a systematic review. This systematic review aims to identify and characterize the published recommendations to conduct and/or to report a systematic review of medical interventions. This systematic review followed the Cochrane Collaboration and the Preferred Reporting Items for Systematic Reviews and Meta-analyses guide. A search was carried out in MEDLINE and EMBASE databases, using “systematic review” indexed term. No language restrictions or publication dates were applied. The search comprised all recommendations to conduct and/or report a systematic review assessing a medical intervention. Data on methodology of each recommendation were extracted. Descriptive analysis was performed. Fifty recommendations were identified. Almost 60% of retrieved references were published in the last 4 years. Recommendations to both conduct and report a systematic review were issued in 42% studies. A total concordance on the methods to conduct and/or report a systematic review was not observed among the recommendations. Literature search and data assessment were recommended in more than 90% of recommendations. An a priori protocol was only recommended in 54% of recommendations. Forty-two percent of recommendations endorsed their methods to any subject related with medical interventions. Half of the studies did not specify the study designs to be included on systematic reviews. Despite the recommendations to conduct and/or report systematic reviews offer guidance on the systematic search and selection and analysis of data, they can differ on other aspects. Such heterogeneity can have methodological implications on the results and their interpretation.