AN IMPUTATION METHOD FOR NON-IGNORABLE MISSING DATA IN STUDIES OF BLOOD-PRESSURE

In studies with repeated measures of blood pressure (BP), particularly in trials of hypertension prevention, BP measurements often become ce nsored once a participant commences antihypertensive medication. When prescribed by non-study physicians under uncontrolled conditions, the missing data mechanism is non-ignorable and may bias the BP effects of interest. I propose a method that models the distribution of BPs meas ured by non-study physicians and their relation to study BPs using ran dom effects models. If treated for hypertension, I assume that BP meas ured outside the study is greater than a clinical cutpoint, such as di astolic BP greater than or equal to 90 mmHg. I then compute estimates for the missing study BPs conditional on previously observed study BPs and treatment for hypertension. Multiple imputation is used to model the variability of the BP values and adjust the standard error estimat es of the parameters. Examples are given using simulated data and data from the weight loss intervention of phase I of the Trials of Hyperte nsion Prevention. (C) 1997 by John Wiley & Sons, Ltd.