The analysis found that in about 1 in 5 cases, shortcomings in the quality of care and other factors beyond patients’ control explain why they leave the hospital before completing recommended treatment.
Clinicians may be quick to blame patients for so-called discharges against medical advice (AMA), which comprise up to 2% of hospital admissions and are associated with an increased risk of mortality and readmission. But “we as providers are very much involved in the reasons why these patients left,” Kushinga Bvute, MD, MPH, a second-year internal medicine resident at Florida Atlantic University, Boca Raton, who led the new study, told this news organization. Dr. Bvute and her colleagues presented their findings April 6 at the Society of General Internal Medicine (SGIM) 2022 Annual Meeting, Orlando, Florida.
Dr. Bvute and her colleagues reviewed the records of 548 AMA discharges – out of a total of 354,767 discharges – from Boca Raton Regional Hospital from January 2020 to January 2021. In 44% of cases, patients cited their own reasons for leaving. But in nearly 20% of AMA discharges, the researchers identified factors linked to treatment.
Hospital-related reasons patients cited for leaving AMA were general wait times (3.5%), provider wait times (2.6%), provider care (2.9%), the hospital environment (2.7%), wanting a private room (2%), and seeking medical care elsewhere (6.2%).
Patient-related factors were refusing treatment (27%), feeling better (3.5%), addiction problems (2.9%), financial complications (2.9%), and dependent care (2.4%). Ten (1.8%) eloped, according to the researchers.
Nearly 60% of patients who were discharged AMA were men, with a mean age of 56 years (standard deviation, 19.13). The average stay was 1.64 days.
In roughly one-third of cases, there was no documented reason for the departure – underscoring the need for better reporting, according to the researchers.
To address AMA discharges, hospitals “need to focus on factors they influence, such as high-quality patient care, the hospital environment, and provider-patient relationships,” the researchers report.
New procedures needed
The hospital is working on procedures to ensure that reasons for AMA discharges are documented. The administration also is implementing preventive steps, such as communicating with patients about the risks of leaving and providing discharge plans to reduce the likelihood that a patient will return, Dr. Bvute told this news organization.
Dr. Bvute said the findings should encourage individual clinicians to “remove any stereotypes that sometimes come attached to having those three letters on your charts.”
Data were collected during the COVID-19 pandemic, but Dr. Bvute does not believe that fear of coronavirus exposure drove many patients to leave the hospital prematurely.
The study is notable for approaching AMA discharges from a quality improvement perspective, David Alfandre, MD, MPH, a health care ethicist at the VA National Center for Ethics in Health Care, Washington, D.C., said in an interview.
Dr. Alfandre, who was not involved in the study, said it reflects growing recognition that hospitals can take steps to reduce adverse outcomes associated with AMA discharges. “It’s starting to shift the conversation to saying, this isn’t just the patient’s problem, but this is the health care provider’s problem,” he said.
Dr. Alfandre co-authored a 2021 analysis showing that hospital characteristics account for 7.3% of variation in the probability of a patient being discharged AMA. However, research is needed to identify effective interventions besides the established use of buprenorphine and naloxone for patients with opioid use disorder. “I think everybody recognizes the quality of communication is poor, but that doesn’t really help us operationalize that to know what to do,” he said.
Emily Holmes, MD, MPH, medical director of the Changing Health Outcomes Through Integrated Care Excellence Program at IU Health, Indianapolis, cautioned that data may be biased because defining AMA discharge can be subjective.
Reasons are not consistently documented and can be difficult to capture because they are often multifactorial, Dr. Holmes said. “For example, long wait times are more problematic when a patient is worried about finances and care for a child,” she said.
But Dr. Holmes, who was not involved in the study, said it does encourage clinicians “to think about what we can do systematically to reduce AMA discharges.”
Dr. Bvute, Dr. Alfandre, and Dr. Holmes reported no relevant financial relationships.