ADJUSTED ODDS RATIOS FOR CASE-CONTROL STUDIES WITH MISSING CONFOUNDERDATA IN CONTROLS

Nonexperimental studies using computerized databases often give rise t o missing or partially available information on con founders. A freque nt situation occurs when data on exposure are available for all subjec ts of a case-control study, but data on confounders are available only for the cases but not for the controls. In that situation, the fact o f confounding can be verified by assessing the association between exp osure and a confounder in the cases, but the data are insufficient to produce an adjusted estimate of the relative risk if confounding is fo und to be present. We propose simple conditions under which an adjuste d estimate of the relative risk can be obtained when data on a confoun der are available only for the cases, and we derive formulae for the e stimator and its confidence limits. The method requires an external es timate of the confounder prevalence or, additionally, of the confounde r-exposure odds ratio. We illustrate the technique with data from a ne sted case-control study of the risk of acute cardiac death associated with the use of bronchodilator drugs within a cohort of 12,301 asthmat ics, with smoking as the confounder of interest.