Sources of bias in retrospective cohort mortality studies: a note on treatment of subjects lost to follow-up.

The three important sources of bias in retrospective cohort mortality studies are: the healthy worker confounding bias, the lost to follow-up bias, and bias due to methods of follow-up that result in underascertainment of deaths. This paper presents how the treatment of the lost to follow-up impacts ultimately on the apparent forces of mortality in a cohort. The findings are discussed in the context of the other sources of bias. The treatment of subjects lost to follow-up as lost at the time of loss offers the best estimate of expected mortality and should be the preferred approach.