Current Practice in Japan for the Prevention and Treatment of Missing Data in Confirmatory Clinical Trials: A Survey of Japanese and Foreign Pharmaceutical Manufacturers.

This study aims to survey the current practice in Japan for the prevention and treatment of missing data in clinical trials since the publication of regulatory guidelines on missing data issues. A web-based questionnaire was conducted among 65 member companies of the Japan Pharmaceutical Manufacturers Association in 2013. Responses were obtained on 187 clinical trials from 55 companies, including 42 based in Japan and 13 based in other countries. Missing data were most frequent in trials involving the central nervous system (65.2% had ≥10% missing data). Overall, last observation carried forward (LOCF) was the most popular method for handling missing data (45.0%), followed by mixed-effect models for repeated measures (15.5%), although this was used as frequently as LOCF imputation in central nervous system trials. Even after the publication of regulatory guidelines discouraging use of LOCF, LOCF imputation remains the most popular method for treating missing data among pharmaceutical manufacturers in Japan.