2024 15th Annual NUBE Conference

Background: Homelessness is a risk factor for Emergency Department (ED) utilization.(1) Additionally, people who are unhoused are more likely to have a mental health diagnosis compared to the general population. Among persons with mental illness, homelessness increases ED utilization.(2) In most healthcare systems, frequent ED users are more likely to have a mental health diagnosis.(3)

In context of the above discussion, it is reasonable to suggest that access to shelter impacts ED utilization. During cold weather months, the incentive to avoid cold exposure may increase ED use, especially if there is scarcity of shelters or other housing resources outside of the hospital setting. However, there is limited research — particularly in the United States — describing this phenomenon. In October 2023, the City of Detroit made a significant change to its shelter access policy: intakes were limited to Monday through Friday 8AM-6PM, and walk-ins were no longer allowed. We hypothesized that the reduction in shelter access would lead to an increase in ED visits due to homelessness.

Methods: A preliminary analysis of ED visits at an urban Level I Trauma Center in the Midwest (90,000+ visits per year) after the change in shelter access revealed an increase in total visits related to homelessness in 10/2023-12/2023 as compared to the previous year. To quantify the overall impact of this change, we conducted a retrospective chart review of ED visits during cold weather months by unhoused individuals before and after change in shelter access occurred (10/2022-3/2023 and 10/2023-3/2024). Unhoused individuals were identified through multiple avenues using homelessness-related ICD-10 codes and ED chief complaints. Additional variables collected included: total ED visits and ED length of stay, and presence/absence of psychotic disorders, bipolar disorders, alcohol use disorders and substance use disorders (also using ICD-10 codes).

Results: Initial analysis in this review demonstrated a 10% increase in the total number of ED visits by unhoused individuals between the two time periods. There was also a 4% increase in the number of different patients who made ED visits between the two time periods. Not all data has been abstracted or analyzed (expected complete date: 9/30/2024) though the next steps include: calculating ED LOS (average and total) for these groups, ED disposition, frequency of psychosis/bipolar disorders/substance use disorders and alcohol use disorders in each population. We hypothesize that the reduction in shelter access led to a disproportionate increase in ED visits by those with severe mental illness and/or substance use disorder diagnoses who are unhoused.

Conclusions: The ED perceived an increase in presentations and boarders after shelter access changed. This study will quantify if there was an actual change and, if so, what risks factors/vulnerable populations were affected. This study contributes to the understanding of how social determinants of health, such as housing, interface with the healthcare system. Our findings draw attention to and reflect a potential need to revisit shelter accessibility in urban areas, as a vulnerable population is being affected in a manner that is increasing the strain on the healthcare system.

Learning Objectives:

Audience members will be able to describe one example of how social determinants of health (e.g. shelter availability) affect emergency department utilization.

Audience members will be able to describe how unhoused individuals with severe mental illness and/or substance use disorders are affected by shelter availability.

Audience members will be able to describe how access to shelter affects length of stay (LOS) and frequency of emergency department encounters.