手法を比較

	リスク調整済み症例クロスオーバーデザイン	リスク調整コホート研究
分野	疫学	疫学
系統	Process / pipeline	Process / pipeline
提唱年	1991 (base design); risk-adjustment extensions from mid-1990s onward	Mid–late 20th century (risk-adjusted cohort designs systematized by 1970s–1990s)
提唱者	Malcolm Maclure (case-crossover base); extensions incorporating covariate risk adjustment developed in subsequent pharmacoepidemiology literature	Evolution of cohort study methodology; risk adjustment formalized through work of Rothman, Greenland, and others in epidemiology, 20th century
種類	Observational analytic epidemiological design	Observational epidemiological study design with statistical confounding control
原典	Maclure, M. (1991). The case-crossover design: a method for studying transient effects on the risk of acute events. American Journal of Epidemiology, 133(2), 144–153.	Rothman, K. J., Greenland, S., & Lash, T. L. (2008). Modern Epidemiology (3rd ed.). Lippincott Williams & Wilkins. ISBN: 978-0781755641
別名	adjusted case-crossover study, covariate-adjusted case-crossover, risk-controlled case-crossover, case-crossover with risk adjustment	adjusted cohort study, covariate-adjusted cohort, risk-controlled prospective study, propensity-adjusted cohort
関連	4	4
概要	The risk-adjusted case-crossover design is a self-matched epidemiological method that compares a person's exposure during a brief hazard window immediately preceding an acute event to their exposure during one or more control windows from the same individual, while formally accounting for time-varying or time-fixed covariates that could confound the exposure-event relationship. By using each case as their own control, stable individual-level confounders are automatically cancelled, while covariate adjustment handles residual time-varying risks.	A risk-adjusted cohort study is an observational epidemiological design in which a defined group of individuals is followed over time to compare outcomes between exposed and unexposed subgroups, with statistical methods applied to control for measured confounders. Adjustment strategies — including multivariable regression, propensity score matching, inverse probability weighting, or standardization — are used to reduce bias and produce effect estimates that more closely approximate what would be observed in a randomized trial.