Patient Care

What Are Best Practices for Patients Discharged against Medical Advice?


Case No. 1

A 41-year-old woman with a history of asthma presents to the emergency department (ED) with shortness of breath and wheezing. She is diagnosed with a mild asthma exacerbation. After three albuterol nebulizer treatments, she still has wheezing on physical examination but appears comfortable and has no oxygen requirement. She has a primary medical doctor at the hospital and follows up with her regularly.

Image Credit:

Image Credit:

The hospitalist recommends that she stay in the hospital for further treatment, but the patient says she has a nebulizer machine at home and asks to be discharged. In addition, she is worried about her frail elderly mother, for whom she is the primary caretaker. The hospitalist acknowledges her concerns but continues to recommend that she remain in the hospital for additional care and monitoring. She becomes visibly upset and insists that she must return home. She asks for prescriptions for albuterol and prednisone and is discharged against medical advice (AMA).

Case No. 2

A 52-year-old man with a history of hypertension and diabetes presents to the ED with left foot pain. He frequently presents with this complaint but often leaves AMA before treatment is completed. He has no known physical address or telephone number and has no known outpatient healthcare providers. Physical examination reveals several ulcers on the dorsum of the foot, one with purulent drainage, and generalized lower extremity pallor. His left leg is cool to the touch, and vascular surgery is consulted for suspected limb-threatening ischemia; IV antibiotics are started for suspected osteomyelitis.

During the interview, he states that he wishes to leave the hospital because he has “things to take care of.” The hospitalist recommends that he remain in the hospital for limb-preserving surgery and antibiotics. He then explains that he is homeless and needs to return to his shelter to keep his bed. He is able to articulate the risks of premature discharge and the medical concerns, and it is determined that he has the capacity to participate in discharge planning. The hospitalist therefore discharges him AMA.


AMA discharges represent 1%–2% of all inpatient discharges.¹,² Despite being a small percentage of total discharges, these patients have disproportionately high healthcare costs. One study reported that healthcare costs among these patients were 56% higher than expected.² Furthermore, AMA patients suffer higher than expected rates of morbidity, mortality, and hospital readmission.

For example, in one case-control study in an urban teaching hospital, patients discharged AMA from the general medicine service had a 21% 15-day readmission rate compared to a 3% readmission rate among age, gender, and diagnosis-matched controls.3,4,5

Additionally, history of AMA discharge appears to confer risk of increased future utilization of healthcare resources. In a cohort study of hospital admissions among HIV-infected patients with high rates of intravenous drug abuse, patients discharged AMA (13% of the cohort) were not only more likely to be readmitted within 30 days for a related diagnosis (odds ratio = 5.0) but also were more likely to have increased length of stay during the year following the index admission.6

These studies highlight the barriers to safe and effective transitions of care for this vulnerable population and demonstrate the increased burden that this population places on the health system.

Several retrospective studies have identified psychosocial and demographic risk factors associated with AMA discharge. These include younger age, male sex, substance abuse, lack of a primary care physician or health insurance, and history of previous AMA discharge.1,3,7,8 Insurance status is also associated with AMA discharge, with increased odds of AMA discharge among Medicare and Medicaid patients and patients without health insurance.9,10

Of note, one study found that race did not act as an independent predictor of AMA discharge when adjusted for age, gender, and socioeconomic factors.11

The AMA population is clinically heterogeneous. Among patients with pneumonia, for example, Saitz et al showed that a patient’s documented clinical severity did not independently predict AMA discharge, suggesting that there is great clinical heterogeneity even among AMA patients with similar admission diagnoses.12

These studies highlight the clinical and demographic heterogeneity within this population, suggesting that patients discharged AMA require individualized attention from hospitalists and other healthcare providers.

Patients describe numerous motivations for leaving the hospital prematurely, including needing to pick up public-assistance checks, personal financial issues, and familial obligations.13 Interestingly, in the cohort of HIV patients referenced above, discharge on the day welfare checks were distributed was an independent predictor of AMA discharge.6 In focus groups composed of patients discharged AMA and their treating nurses and physicians, several themes were identified as potential contributors to AMA discharge, including drug addiction, pain management issues, external obligations, wait time, the physician’s bedside manner, being in a teaching hospital, and communication issues.14

Clearly, patients have a diversity of reasons for requesting to be discharged AMA, and further research is necessary to define clear and potentially modifiable risk factors.


The clinical scenarios outlined above present two patients with very different clinical presentations and outpatient support systems as well as demonstrate the great variability in clinical risk at the time of discharge AMA. These examples emphasize the importance of an individualized approach to care for each patient.

In Case No. 1, the patient is admitted with a mild asthma exacerbation with persistent bronchospasm, though she clinically appears well and has reliable follow-up. In contrast, in Case No. 2, the patient has life-threatening disease and no established primary care physician or mechanism for outpatient care. These examples demonstrate extremes on the clinical and psychosocial spectrum of patients requesting an “early” discharge and suggest that no two patients at risk of AMA discharge are the same. Patient 1 could likely be safely managed at home with close outpatient follow-up, while Patient 2 presents a high-risk scenario with very few safe outpatient treatment options.

We suggest that an individualized approach be taken for each patient, with attention to both clinical and psychosocial risk. In clinically low-risk cases (e.g., Case No. 1), an approach that prioritizes shared decision making and coordination with the outpatient care team may be preferable to an AMA discharge, particularly given the often adversarial nature of the later.2 In such cases, a collaborative approach may provide greater opportunity for harm reduction, provision of appropriate prescriptions, and follow-up appointments. In clinically high-risk patients such as Case No. 2, however, premature discharge is clearly inappropriate. Even in such clinically high-risk cases, however, we argue that a collaborative strategy aimed at identifying and addressing the patient’s psychosocial concerns is appropriate, as such an approach promotes shared decision making, builds trust between the patient and the care team, and therefore may facilitate improved follow-up at the time of discharge. Research is needed to formally assess the optimal approach for this patient population, including impact on rates of AMA discharge and the quality of post-discharge follow-up.

At present, the decision to classify a discharge as AMA falls solely on the treating provider, and we suspect that there is great variability in practice patterns, particularly as there are few established professional society practice guidelines regarding this difficult issue. As with all discharges from the hospital, the burden falls on the provider to engage the patient in shared decision making and ensure that the patient has the capacity to understand the risks and benefits of the proposed treatment plan. It is in this spirit that simply “filling out an AMA form” does not provide legal protection to a physician who does not adequately explain the full risks and benefits of refusal of inpatient treatment.2,15

We propose that a high-quality AMA discharge be defined as a discharge in which the patient is informed of the clinical team’s determination that further hospitalization is required but elects to leave the hospital, and it includes a clear discussion of the risks of outpatient treatment, a determination of capacity, and an exploration of safe alternative care plans that could satisfy both the patient’s medical and social needs. This definition places the burden on hospitalists and other providers to fully explore the motivations behind a patient’s request to leave the hospital and treats psychosocial motivators for premature discharge as variables in the complex risk-benefit analysis that underlies the informed consent discussion prior to AMA discharge.

Furthermore, AMA discharge does not obviate a physician’s responsibility to advocate for a patient’s well-being, and therefore an AMA discharge should be accompanied by reasonable efforts to coordinate a patient’s ongoing outpatient care. Of note, this approach is consistent with previous reviews and attempts to balance the physician’s duty to honor a patient’s autonomy with the responsibility to protect the patient from harm.2,16


Patients discharged AMA are a diverse population at markedly increased risk of morbidity, readmissions, and subsequent healthcare cost. We argue that in all cases of a potential premature discharge, a collaborative and patient-centered approach is crucial. Such an approach allows the provider to identify and address the patient’s concerns regarding further inpatient care, to explore possible safe outpatient treatment options, to document patient capacity, and to provide appropriate harm-reduction measures such as prescriptions.

Further research into the current practice patterns of hospitalists and other providers is necessary to allow for the formulation and adoption of best practices and implementation of appropriate harm-reduction strategies. TH

Dr. Tummalapalli is an internal medicine resident in the department of medicine at Icahn School of Medicine at Mount Sinai in New York City. Dr. Goodman is a hospitalist in the division of hospital medicine, department of medicine, at the Icahn School of Medicine at Mount Sinai.

Key Points

  • Patients discharged against medical advice are a vulnerable patient population and suffer increased morbidity and mortality, healthcare costs, and rates of readmission.
  • AMA patients are clinically, demographically, and psychosocially heterogeneous.
  • Shared decision making between patients and providers may allow for harm reduction in discharge planning and may obviate the need to discharge certain patients AMA.
  • Further research is needed to better understand current practice patterns and to identify evidence-based strategies for safe discharge planning in the AMA population.


    1. Aliyu ZY. Discharge against medical advice: sociodemographic, clinical and financial perspectives. Int J Clin Pract. 2002;56(5):325-327.
    2. Kahle CH, Rubio ML, Santos RA. Discharges against medical advice: considerations for the hospitalist and the patient. Hospital Medicine Clinics. 2015;4(3):421-429.
    3. Baptist AP, Warrier I, Arora R, Ager J, Massanari RM. Hospitalized patients with asthma who leave against medical advice: characteristics, reasons, and outcomes. J Allergy Clin Immunol. 2007;119(4):924-929.
    4. Hwang SW, Li J, Gupta R, Chien V, Martin RE. What happens to patients who leave hospital against medical advice? CMAJ. 2003;168(4):417-420.
    5. Glasgow JM, Vaughn-Sarrazin MV, Kaboli PJ. Leaving against medical advice (AMA): risk of 30-day mortality and hospital readmission. J Gen Int Med. 2010;25(9):926-929.
    6. Anis AH, Sun H, Guh DP, Palepu A, Schechter MT, O’Shaughnessy MV. Leaving hospital against medical advice among HIV-positive patients. CMAJ. 2002;167(6):633-637.
    7. Jeremiah J, O’Sullivan P, Stein MD. Who leaves against medical advice? J Gen Int Med. 1995; 10(7);403-405.
    8. O’Hara D, Hart W, McDonald I. Leaving hospital against medical advice. J Qual Clin Pract.1996;16(3):157-164.
    9. Ibrahim SA, Kwoh CK, Krishnan E. Factors associated with patients who leave acute-care hospitals against medical advice. Am J Public Health. 2007;97(12): 2204-2208.
    10. Weingart SN, Davis RB, Phillips RS. Patients discharged against medical advice from a general medicine service. J Gen Intern Med. 1998;13(8):568-571.
    11. Franks P, Meldrum S, Fiscella K. Discharges against medical advice: are race/ethnicity predictors? J Gen Intern Med. 2006;21(9):955-960.
    12. Saitz R, Ghali WA, Moskowitz MA. Characteristics of patients with pneumonia who are discharged from hospitals against medical advice. Am J Med. 1999;107(5):507-509.
    13. Alfandre, DJ. “I’m going home”: discharges against medical advice. Mayo Clin Proc. 2009;84(3):255-260.
    14. Onukwugha E, Saunders E, Mullins CD, Pradel FG, Zuckerman M, Weir MR. Reasons for discharges against medical advice: a qualitative study. Qual Saf Health Care. 2010;19(5):420-424. doi: 10.1136/qshc.2009.036269.
    15. Battenfeld v. Gregory, 589 A.2d 1059, 1061 (N.J. Super. Ct. App. Div. 1991).
    16. Berger J. Discharge against medical advice: ethical considerations and professional obligations. J Hosp Med. 2008;3(5):403-408.

Next Article:

   Comments ()