Estimating risks for matching factors in case-control studies

Matching for factors such as age and sex is a convenient method for minimiz ing confounding in case-control studies, but it does not allow inferences a bout the effects of the matching factors unless case ascertainment is virtu ally complete and the distribution of the matching factors in the source po pulation is known. When this is so, the effect of a particular factor can b e estimated by comparing the population distribution of that factor with wh at is observed in the case series. Such a comparison, however, may itself b e confounded by other factors that are related to both the matching factors and the disease under investigation. This article proposes a method for ev aluating matching factors as risk factors, which uses information on the di stribution of potential confounders in the reference series and exposure re lative risk estimates to adjust the person-time proportionality constant in a Poisson regression model. The method is particularly suited to data sets in which many of the elementary matching strata contain few or no cases an d/or controls. It makes use of standard analytic procedures, but requires t he estimation of an additional variance-covariance component for the estima ted Poisson regression coefficients. Further factors that may confound the relationship between exposure and disease are easily accommodated. The meth od is demonstrated in two examples: a matched case-control study of drugs i n relation to the rare blood dyscrasia, agranulocytosis,that was conducted in Europe and Israel, and a case-control study of ovarian cancer in Austral ia. (C) 2000 Elsevier Science Inc. All rights reserved.