In the Systolic Hypertension in the Elderly Program (SHEP) trial (1985-1990), active treatment reduced the incidence of cardiovascular events, but not that of dementia and disability, as compared with placebo. This study aims to evaluate if assessment of cognitive and functional outcomes was biased by differential dropout. Characteristics of subjects who did or did not participate in follow-up cognitive and functional evaluations were compared. The relative risks of incident cognitive impairment and disability were assessed in the two treatment groups, with the use of the reported findings and under the assumption that the proportions of cognitive and functional impairment among dropouts increased. Assignment to the placebo group and the occurrence of cardiovascular events independently predicted missed assessments. From the reported findings, the risk of cognitive and functional impairment was similar between the two treatment groups. However, when 20-30% and 40-80% of the subjects who missed the assessment were assumed to be cognitively and, respectively, functionally impaired, assignment to active treatment reduced the risk of these outcomes. In the SHEP, the cognitive and functional evaluations were biased toward the null effect by differential dropout. This might have obscured the appraisal of a protective effect of treatment on the cognitive and functional decline of older hypertensive adults.