Trends and Outcomes in the Hospitalization of Older Americans for Cardiac Conduction Disorders or Arrhythmias, 1991–1998

OBJECTIVE: To identify epidemiological trends and measure outcomes in elderly patients hospitalized for cardiac conduction disorders or arrhythmias. DESIGN: Review of the standard 5% samples of the Medicare Provider Analysis and Review Files to characterize 144,512 discharges from 1991 through 1998 in which the principal diagnosis was a conduction disorder or arrhythmia, using the corresponding Enrollment Databases for denominator data. SETTING: Short-stay hospitals in the United States. PARTICIPANTS: Medicare beneficiaries age 65 and older in the standard 5% sample. MEASUREMENTS: Diagnosis-specific trends and rates; discharges by year; cumulative age-, race-, and sex-specific discharge rates; mean length of stay in hospital and in intensive care; mean Medicare reimbursement to the hospital; case-fatality rate in hospital; discharge destinations of patients discharged alive. RESULTS: Annual hospitalizations for sinoatrial node dysfunction, atrial flutter, atrial fibrillation, or ventricular fibrillation increased more rapidly than did the elderly Medicare beneficiary population. Hospitalizations with a principal diagnosis of ventricular extrasystoles or asystole showed steep secular declines. Discharge rates for sinoatrial node dysfunction, a group of rhythms with a nonsinus pacemaker, atrial fibrillation, Mobitz I, or complete atrioventricular block all increased steeply and continuously with patient age. In contrast, discharge rates for atrial flutter or ventricular tachycardia or fibrillation peaked among 75- to 84-year-old patients. White men were at uniquely high risk of hospitalization for atrial flutter or ventricular tachycardia or fibrillation, and, among the white majority, men had higher discharge rates than women for nine of the 11 commonest rubrics. Whites, particularly white women, had the highest discharge rates for atrial fibrillation. Blacks, especially black women, were at disproportionate risk for hospitalization for the group of nonsinus pacemaker rhythms. Diagnosis-specific mean resource costs were strongly correlated with each other and with mean Medicare reimbursement but not with case-fatality rate. CONCLUSION: Medicare claims data demonstrated striking differences among and within diagnoses of heart blocks or arrhythmias in terms of the populations at greatest risk for hospitalization. This variation should be explored further to generate and test hypotheses about differential causation or delivery of care.