Re-examining threats to the reliability and validity of putative brain-behavior relationships: new guidelines for assessing the effect of patients lost to follow-up.

It often happens in behavioral and biomedical research that subjects in prospective, multiple assessment investigations, including clinical trials, are lost to follow-up evaluations. The purpose of this report is to outline a model that will enable the investigators to determine the extent to which results based upon the maintained cohort can be generalized to the attrited cohort, or those subjects lost to follow-up. While our proposed model derives from a specific application pertaining to changes in personality and affect behaviors following left and right hemisphere stroke, it should apply, with appropriate study-specific modifications, to a wide range of follow-up research designs in neuropsychology, behavioral science more generally, and other areas of biomedical research.