ABSTRACT
PURPOSE: The purpose of this study is to enhance the quality of data by performing a methodological assessment of medical records reviews. METHODS: We reviewed the articles published in the Journal of the Korean Society of Emergency Medicine between January 2001 and December 2003 that used a retrospective medical records review as the study method. We assessed data collector's training, descriptions of inclusion/ exclusion criteria, definitions of important variables, use of standardized case record forms, monitoring the data collectors' performance, blind data collecting, inter-rater reliability/ test of inter-rater agreement, intra-rater reliability/test of intra-rater agreement, selection bias from consent, and rules regarding management of missing data. RESULTS: There were 111 articles that used retrospective medical records reviews during the study period. In 111 (100%) articles, inclusion/exclusion criteria were described, in 98 (88.3%), important variables were defined, and in 4 (3.6%), standardized case record forms were used. However no articles addressed the other items on the checklist. CONCLUSION: Study conductors should design studies to enhance the quality of data, and detailed descriptions are necessary to improve the reproducibility of the study.