MATCHED-PAIR CASE-CONTROL STUDIES WHEN RISK-FACTORS ARE CORRELATED WITHIN THE PAIRS

Background. if pair members are independent, simple matched-pair case- control studies are known to yield consistent estimates of the populat ion odds ratio. If pair members are not independent, this is not neces sarily true, It has been shown previously that the usual matched-pair estimate remains consistent if the exposure of interest is correlated within the pairs. However, the effect of correlation of unmeasured ris k factors within the pairs has not been studied.Methods. We examine th e effect of within-pair correlation of unmeasured risk factors indepen dent of the measured exposure. This is done within the context of a si mple matched-pair case-control study. We compare the large-sample expe ctation of the usual matched-pair estimate to the population odds rati o. Results. We show that the usual estimate may be inconsistent in the presence of this correlation. However, if the disease is rare, the ma gnitude of the bias will be negligible. Conclusions. Correlation of un measured risk factors independent of the measured exposure is not a pr actical problem in this setting.