Selection bias in cohort studies is generally less likely than in other studies, as selection for participation in the study generally precedes the development of the outcome of interest. However, losses to follow-up provide a mechanism by which selection bias may be introduced. For example, in the case of a perceived 'risky behaviour', it is plausible that individuals engaged in this may be less likely to remain in the study, meaning that these individuals (which may be more likely to experience the outcome) will be lost from the study. | Selection bias in cohort studies is generally less likely than in other studies, as selection for participation in the study generally precedes the development of the outcome of interest. However, losses to follow-up provide a mechanism by which selection bias may be introduced. For example, in the case of a perceived 'risky behaviour', it is plausible that individuals engaged in this may be less likely to remain in the study, meaning that these individuals (which may be more likely to experience the outcome) will be lost from the study. |