手法を比較

	適応型症例シリーズ	適応型診断精度研究
分野	疫学	疫学
系統	Process / pipeline	Process / pipeline
提唱年	Late 20th–early 21st century	2000s–2010s (adaptive designs codified for diagnostics ~2010s)
提唱者	Evolved from classical case series methodology combined with adaptive design principles (Chow & Chang, 2008; FDA adaptive design guidance)	Adaptation of STARD framework (Bossuyt et al.) combined with adaptive design principles (Jennison & Turnbull; FDA guidance)
種類	Observational study with adaptive monitoring	Adaptive observational/experimental study design
原典	Chow, S.-C., & Chang, M. (2008). Adaptive Design Methods in Clinical Trials. Chapman & Hall/CRC. ISBN: 978-1584887775	Bossuyt, P. M., Reitsma, J. B., Bruns, D. E., Gatsonis, C. A., Glasziou, P. P., Irwig, L., ... & Cohen, J. F. (2015). STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ, 351, h5527.
別名	adaptive case-series design, sequential adaptive case series, adaptive observational case series	adaptive DTA study, adaptive diagnostic test evaluation, adaptive test accuracy trial, adaptive STARD study
関連	5	6
概要	An adaptive case series is an observational study design that documents a consecutive group of patients with a shared condition or exposure while incorporating pre-specified rules for modifying data collection, monitoring, or stopping criteria as accumulating evidence warrants. It combines the descriptive richness of traditional case series with the prospective flexibility of adaptive design principles, enabling structured mid-course adjustments without compromising the integrity of the recorded clinical observations.	An adaptive diagnostic accuracy study evaluates how well an index test distinguishes between patients with and without a target condition, while incorporating pre-specified interim analyses that allow modifications — such as sample size re-estimation, threshold adjustment, or subgroup enrichment — based on accumulating data. This design improves efficiency and ethical conduct compared to fixed-sample diagnostic studies, particularly when prior prevalence or test performance data are uncertain.