Digoxin intoxication has been reported to be a common adverse drug reaction with an in-hospital incidence of 6% to 23% and an associated mortality rate as high as 41%. A retrospective review was conducted to assess the accuracy of diagnosis, the morbidity and mortality of digoxin intoxication, and its incidence in hospitalized patients with heart failure. We reviewed the medical records of 219 patients discharged with the diagnosis of digoxin intoxication between 1980 and 1988. Patients were classified as follows: (1) Definite intoxication--patients with symptoms and/or arrhythmias suggestive of digoxin intoxication that resolved after discontinuation of digoxin; (2) possible intoxication--patients with symptoms and/or arrhythmias suggestive of digoxin intoxication in the absence of documented resolution after discontinuation of digoxin, or the presence of other clinical illnesses that could possibly account for those findings; (3) no intoxication--patients whose symptoms or ECG abnormalities were clearly explained by other associated clinical illnesses and persisted after withdrawal of digoxin. We identified only 43 patients (20%) with definite intoxication. The majority of patients discharged with the diagnosis of digoxin intoxication (133 or 60%) were classified as possibly digoxin intoxicated, and 43 patients (20%) had no clinical evidence to support this diagnosis. To estimate the incidence of digoxin intoxication, we also reviewed the medical records of 994 patients admitted in 1987 with heart failure. Of these, 563 were receiving digoxin and in 27 the diagnosis of digoxin intoxication was made by their clinicians. Our review showed that only four were definitely intoxicated (0.8%), and the diagnosis could not be excluded in another 16 (4%).(ABSTRACT TRUNCATED AT 250 WORDS)