“Against Medical Advice” Discharges: Disparities in Post-Discharge Care and Outcomes

Introduction: Against medical advice (AMA) discharges, which occur when a patient leaves the hospital prior to their medical team’s recommendation to do so, are frequent and associated with poor outcomes. Some prior qualitative evidence suggests that clinicians may feel less obligated to offer post-discharge care (e.g. referrals for outpatient follow-up) to patients who are discharged AMA than non-AMA. We sought to characterize patterns of discharge referrals for patients discharged AMA versus non-AMA after exacerbations of chronic obstructive pulmonary disease.

Methods: Using electronic health record data from 15,594 admissions, we compared referrals for post-discharge care and risk of 30-day re-admission between non-AMA versus AMA discharges.

Results: Compared to non-AMA discharges, patients discharged AMA were less likely to receive any referral for post-discharge care (26.5 vs 11.5%, p< 0.001) and were 43% more likely to be re-admitted at 30-days (p < 0.001).

Discussion: Our results raise important questions as to why patients discharged AMA receive discharge referrals less often than non-AMA discharges. For instance, this finding might reflect that clinicians cannot feasibly facilitate referrals if AMA discharges occur too quickly. Alternatively, as some prior evidence suggests, clinicians’ biases related to AMA discharges may contribute to sub-standard discharge practices. In considering the potential role of clinician bias, we emphasize the need for attention to the ethical principles of ‘respect for persons’ and ‘beneficence’ with AMA discharges—even if those discharges do not seem medically ideal compared to continued admission. Ultimately, we argue for a need to develop strategies to reduce adverse outcomes among those discharged AMA.

Aparna Balasubramanian – Division of Pulmonary and Critical Care Medicine – Johns Hopkins University; Meredith McCormack – Division of Pulmonary and Critical Care Medicine – Johns Hopkins Univeristy; Jeremy Sugarman – Berman Institute of Bioethics – Johns Hopkins Univeristy