Multiple imputation of missing blood pressure covariates in survival analysis

This paper studies a non-response problem in survival analysis where the oc currence of missing data in the risk factor is related to mortality. In a s tudy to determine the influence of blood pressure on survival in the very o ld (85+ years), blood pressure measurements are missing in about 12.5 per c ent of the sample. The available data suggest that the process that created the missing data depends jointly on survival and the unknown blood pressur e, thereby distorting the relation of interest. Multiple imputation is used to impute missing blood pressure and then analyse the data under a variety of non-response models. One special modelling problem is treated in detail ; the construction of a predictive model for drawing imputations if the num ber of variables is large. Risk estimates for these data appear robust to e ven large departures from the simplest non-response model, and are similar to those derived under deletion of the incomplete records. Copyright (C) 19 99 John Wiley & Sons, Ltd.