Comments on: Missing Data Methods in Longitudinal Studies: A Review