Increased mortality in digoxin-treated subjects has been demonstrated in patients with recent myocardial infarction. Those with congestive heart failure (CHF) due to causes other than myocardial infarction seem to be free from this effect. No information is currently available concerning mortality in elderly people who are frequently prescribed digitalis even in the absence of CHF. The aim of this study was to investigate whether subjects improperly receiving digoxin were worse off than those not receiving this drug. This analysis is a part of CASTEL, a population-based prospective study that has enrolled a cohort of 2,254 subjects aged > or = 65 years. CHF was diagnosed in 187 subjects and atrial fibrillation (AF) in 90. The remaining 1,977 were free from CHF and in sinus rhythm, but 447 were treated with digitalis. Cumulative mortality and morbid events by digitalis treatment were calculated in all these categories. Among subjects free from CHF and AF (improper use), all-cause and cardiovascular mortality was significantly higher among those taking digitalis than in those who did not. Non-fatal events including CHF were also more apparent in the former than in the latter. Cox analysis confirmed digitalis as a predictor of mortality in these subjects. No effect of digitalis on survival was found in patients with CHF or AF (proper use). In elderly subjects without atrial fibrillation or CHF, the use of digitalis worsens morbidity and mortality.