The effects of the order of multiple imputation in subset analysis examining the association between body mass index (BMI) and transrectal ultrasound prostate weight

Subset analysis is a comprehensive method for examining associations within a data set. This approach may be substantially affected by missing data. Multiple imputation is one approach for handling missing data. This work will explore how subset analysis and multiple imputation can provide benefit in investigating clinically relevant associations. Furthermore, the order in which these methods are performed and its effects will be examined. With this data set, we observe association between body mass index (BMI) and prostate weight as measured by transrectal ultrasound. We show that the order of the procedures affects the inferences.