Principled Approaches to Missing Data in Epidemiologic Studies

Abstract

Principled methods to appropriately analyze missing data have long existed; however, broad implementation of these methods remains challenging. In this and companion papers, we discuss issues of missing data in the epidemiologic literature. We provide details regarding missing data mechanisms and nomenclature and motivate principled analyses through a detailed comparison of multiple imputation and inverse probability weighting. We do so in the setting of a masked data-analytic challenge with missing data induced by known mechanisms to data from the Collaborative Perinatal Project, a multisite US study conducted from 1959 to 1974. We illustrate the deleterious effects of missing data with naïve methods and show how principled methods can sometimes mitigate such effects. For example when data were missing at random, naïve methods showed a spurious protective effect of smoking on spontaneous abortion, odds ratio (OR) of 0.43 (95% confidence interval, CI: 0.19, 0.93) while implementing principled methods multiple imputation (OR = 1.30, CI: 0.95, 1.77) or augmented inverse probability weighting (OR = 1.40, CI: 1.00, 1.97) provided estimates closer to the "true" full data effect (OR = 1.31, CI: 1.05, 1.64). We call for greater acknowledgement of and attention to missing data and for the broad use of principled missing data methods in epidemiologic research.