Abstract

BACKGROUND:

Approximately 1% of patients leave hospitals against medical advice, but the clinical significance of premature hospital discharge is unknown, particularly after admission for acute myocardial infarction (AMI).

METHODS:

We used California hospital discharge data (1998-2000) to compare readmissions and mortality among patients admitted for AMI who were discharged against medical advice with those who weren't. Effects were adjusted for age, race, income, comorbidity, insurance, and hospital characteristics. We also examined whether the effects of premature hospital discharge were partly explained by lower rates of coronary revascularization.

RESULTS:

There were 1079 patients (1.1% of the sample) with AMI on admission who left against medical advice. Compared with those who didn't leave against medical advice, these patients were younger, more often male, low income, black, insured through Medicaid or uninsured, and had less physical comorbidity, but greater mental health comorbidity. Their mean length of stay was shorter (4 vs 8 days) than those who stayed. They were transferred less often. They received fewer cardiac procedures, including coronary revascularization. In multivariate analyses, they had 60% higher risk for death or re-admission for AMI or unstable angina up to 2 years postdischarge than patients with standard discharge (hazard ratio 1.59; 95% confidence interval, 1.43-1.77). Adjustment for revascularization attenuated, but did not eliminate, this risk (hazard ratio 1.39; 95% confidence interval, 1.25-1.55).

CONCLUSIONS:

Discharge against medical advice after AMI is associated with appreciable morbidity and mortality. These results should be used to manage AMI patients contemplating such discharge.