Skip to main content
Fig. 2 | BMC Medical Research Methodology

Fig. 2

From: A novel weighting method to remove bias from within-subject exposure dependency in case-crossover studies

Fig. 2

Expected frequency of cases by subgroup in Fig. 1. (a) 140 cases who had the outcome at the case period involving one binary exposure variable only, where the incidence rate per period at unexposed period (r0) is assumed to be 0.001 and the rate ratio of the exposure (RR) is assumed to be 4. (b) 184 cases who had the outcome at the case period involving one binary exposure variable and one binary time-varying confounder where the incidence rate per period at unexposed period without confounder (r0) is assumed to be 0.001, the rate ratio of the exposure (RR) is assumed to be 4, and that of the time-varying confounder (RRz) is assumed to be 2. The proportion of the time-varying confounder at unexposed periods (f0) is 0.2 and that at exposed periods (f1) in the population 0.4. The status of the confounder at control periods is randomly assigned . n: frequency of cases belonging to each ID_Subgroup in Fig. 1; c0: exposure status at the case period; z0: status of time-varying confounder at the case period

Back to article page