Strategies for Multiple Imputation in Longitudinal Studies

摘要

Multiple imputation is increasingly recommended in epidemiology to adjust for the bias and loss of informationnthat may occur in analyses restricted to study participants with complete data (‘‘complete-case analyses’’). However,nlittle guidance is available on applying the method, including which variables to include in the imputationnmodel and the number of imputations needed. Here, the authors used multiple imputation to analyze the prevalencenof wheeze among 81-month-old children in the Avon Longitudinal Study of Parents and Children (Avon,nUnited Kingdom; 1991–1999) and the association of wheeze with gender, maternal asthma, and maternal smoking.nThe authors examined how inclusion of different types of variables in the imputation model affected point estimatesnand precision, and assessed the impact of number of imputations on Monte Carlo variability. Inclusion of variablesnassociated with the outcome in the imputation model increased odds ratios and reduced standard errors. Whennonly 5 or 10 imputations were used, variability due to the imputation procedure was substantial enough to affectnconclusions. Careful preliminary analysis identified the scope for multiple imputation to reduce bias and improvenefficiency and provided guidance for building the imputation model. When data are missing, such preliminarynanalyses should be routinely undertaken and reported, regardless of whether multiple imputation is used in thenfinal analysis.