New Tools Identify Patients at Risk of Overdose After Hospital Discharge

Research published in the Canadian Medical Association Journal highlights the potential of risk prediction tools to identify patients most vulnerable to overdose and death after being discharged from hospital against medical advice. This phenomenon, referred to as “before medically advised” (BMA) discharge, poses significant risks; patients who leave hospital prematurely are approximately twice as likely to die and about ten times more likely to suffer an illicit drug overdose within the first 30 days.

Each year, around 500,000 individuals in the United States and 30,000 in Canada opt for BMA discharges. Dr. Hiten Naik from the University of British Columbia in Vancouver, along with his co-authors, emphasizes that calculating a patient’s specific risk of death and drug overdose, combined with clinical judgment, can facilitate a constructive dialogue between clinicians and patients. This discussion may include evaluating the patient’s capacity to make informed decisions and exploring strategies to mitigate risks associated with BMA discharges.

Developing Risk Prediction Models

The researchers created two distinct risk prediction models. The first estimates the risk of death from any cause within 30 days following a BMA discharge. The second focuses on patients with a history of substance use, estimating their risk of illicit drug overdose in the same timeframe. By analyzing data from British Columbia, the study examined two cohorts: the first comprised 6,440 adults from the general population who initiated a BMA discharge, while the second included 4,466 individuals known to have a history of substance use.

In cohort A, the study revealed that death rates were lower than anticipated, with one death occurring for every 63 BMA discharges within the 30-day period. Strong predictors of mortality included multimorbidity, heart disease, and cancer. Conversely, cohort B highlighted that factors such as homelessness, reliance on income assistance, opioid use disorder, and previous drug overdoses were significant predictors of drug overdose following BMA discharge. The findings indicated that among patients with a history of substance use, illicit drug overdose occurred approximately once for every 19 BMA discharges, suggesting that this post-discharge period represents a critical opportunity for overdose prevention.

Implementing Risk Prediction in Healthcare

The authors of the study propose that hospitals and health systems could incorporate these risk prediction models into their protocols to better manage high-risk BMA discharges. Automation of alerts and automatic enrollment in support programs could enhance patient safety. Dr. Naik and his colleagues assert that these models provide a foundational step towards identifying at-risk patients, enabling them to receive the necessary support.

As healthcare systems continue to grapple with the complexities of patient care and discharge processes, these insights underscore the importance of utilizing data-driven tools to improve outcomes. The study advocates for a more proactive approach in addressing the needs of vulnerable populations, ultimately aiming to reduce the risks associated with BMA discharges.

This research serves as a pivotal contribution to the ongoing discourse surrounding patient safety and the critical importance of informed decision-making in healthcare settings. For further details, refer to the study titled “Predicting drug overdose and death after ‘before medically advised’ hospital discharge” in the Canadian Medical Association Journal, set for publication in 2025.