Skip to main content
Fig. 1 | BMC Medical Research Methodology

Fig. 1

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

Fig. 1

Exposure pattern in a hypothetical dynamic population of 80,000 patients. Patients receiving drug treatment with a cycle of 7 days are divided into 8 Subgroups A to H where 1 period is defined as 1 day. Subgroup A represents stoppers that stop treatment, Subgroup H represents starters that start treatment, and Subgroups B to G represent different exposure patterns of those being treated at the case period. The bolded outline indicates that 21 control periods can be divided into 3 cycles with the same exposure pattern. N: size of each subgroup; c0: exposure status at the case period; cm (m = 1, 2, −--, 21): exposure status at the m-th control period; Tx: treatment

Back to article page