1025 — Effect of Peer Support on Emergency Room usage in Relation to Primary Care in a High Risk Veteran Population

Lead/Presenter: Chantele Mitchell-Miland, COIN - Pittsburgh/Philadelphia
All Authors: Mitchell-Miland CE (Pittsburgh VA) Chinman M (Pittsburgh VA) McCarthy S (Pittsburgh VA) Shutt R (University of Massachussetts Boston, Bedford VA) Hanusa B (Pittsburgh VA) Kelly M (Pittsburgh VA) Schultz M (Bedford VA) Shah M (Pittsburgh VA)Ellison M (Bedford VA)

Objectives:
The use of emergency room services for medical care can be an indicator of patient medical concerns or inappropriate uses of the emergency room for non-emergent care. Average costs for emergency room care are at least double compared to primary care. Models of care are moving toward reducing emergency room visits, when possible, to offset costs and provide more continuous care. We examined the effect of peer support services on emergency room usage (ERU) in relation to primary care (PCU) over time, in a chronically-ill, high-risk population of Veterans.

Methods:
Formerly homeless Veterans with a history of mental illness and substance use problems who received housing assistance and support from HUD-VASH were recruited at two sites into a randomized trial of peer support compared to usual care. Service utilization data from medical records were used to calculate a ratio of ERU to total ERU and PCU visits for four time periods (T0 = 90 days prior, T1 = 1st half of the study, T2 = second half of study and T3 = 90 days post). Lower ratios indicated greater PCU in relation to ERU. Generalized linear mixed models were used to assess differences in the ratios across time.

Results:
The total sample was 167 (Intervention = 86, Control = 81). There was a significant main effect for time (p < 0.01). For both groups, ratios increased from baseline (T0) to T1 (p < 0.01) and T2 (p < 0.01), with the increase being highest at T2. For the intervention group only, there was a significant drop in the ratio (p = 0.03) at T3 compared to T0.

Implications:
During the study, there was a temporary increase in ERU in both groups. Study participation might have heightened participant's awareness for emergency treatment for both groups. However, participants who received focused care from a peer specialist had less ERU compared to baseline than controls after study completion.

Impacts:
Understanding patterns of ERU in relation to PCU in high-risk Veteran populations may provide insight into identifying better ways of increasing PCU and decreasing ERU, leading to more continuous care, and reducing costs.