Cause of death among 917 of the 959 subjects in a population-based incidence cohort with Alzheimer's disease (AD) during the years 1960 to 1984 was compared to that of an age- and gender-matched control group (n = 703). In general, a summary of the diagnostic codes entered anywhere on a death certificate suggests that control subjects had more cardiovascular disease and neoplasms than did patients, while patients more often had a diagnostic code of bronchitis/pneumonia (P < 0.01). By logistic regression, this difference remained statistically significant after adjusting for age and gender.