The quality of reporting of case control studies published in selected Sri Lankan medical journals

Abstract

BACKGROUND: Incomplete and inadequate reporting of research hampers the assessment of strengths and weaknesses of the studies published in the medical literature. OBJECTIVE: To assess the quality of reporting of case control studies published in selected Sri Lankan medical journals METHODS: Nine medical journals were hand searched for case control studies published since 1980 and all (18) which appeared in the 287 journal issues were included for assessment. A checklist was developed based on STROBE guidelines. Each study was assessed by two independent observers for presence of criteria in the checklist developed. RESULTS: Of the 18 articles, 10 (65.5%) clearly and correctly stated the study design, 16 (89%) the objectives, 11 (61%) described the periods of recruitment, 14 (77.7%) reported comparability of diagnosis of cases and controls clearly, 15 (83.3%) defined exposure and confounding variables clearly, 11 (61%) described comparability of assessment of exposure variables among cases and controls, 11 (61%) data collection methods and 12 (66.6%) sources of data. With regard to statistical issues, 3 (16.6%) included method of sample size determination, 15 (83.3%) presented unadjusted estimates of effects and 6 (33.3%) adjusted estimates and 10 (55.5%) the precision of the estimates. Of the 7 (39%) matched case control studies and only one applied matched analysis. Five (27.7%) studies have attempted to address potential sources of bias but none described the generalizability of the findings. CONCLUSION: None of the studies had fulfilled all criteria recommended by STROBE guidelines. Awareness on the STROBE statement and more attention to details may improve the quality of reporting.