CARES: A Community-wide Collaboration Identifies Super-utilizers and Reduces Their 9-1-1 Call, Emergency Department, and Hospital Visit Rates.

Background: A subset of individuals who inefficiently and frequently use emergency department (ED) services are called "super-utilizers." Our healthcare system is fragmented and complex, making it difficult for providers to identify super-utilizers and address their wide range of health issues.

Objective: The objective of our study was to evaluate a novel community-wide collaboration program called CARES (Community Assistance Referral and Education Services) designed to identify super-utilizers through local partnering organizations. CARES assists patients in developing their personal health and wellness goals, and navigates them away from 9-1-1 calls, emergency room visits, and hospital admissions, and toward more appropriate resources over 90 days.

Methods: This was a retrospective observational analysis of the CARES program. Data were collected from March 2013 to December 2015. The study population included: enrolled adults with non-compliance of medication or treatment; behavioral health problems; multiple 9-1-1 responses in a short period of time; three or more ED visits within six months; patients with multiple hospital admissions. Adults who were outside of the study period or had missing outcome information were excluded. The primary outcomes of this study were the median rate of 9-1-1 calls/month/person, ED and hospital visits/month/person. Wilcoxon rank-sum tests were used to compare changes between pre- vs. post-enrollment for each subject.

Results: A total of 441 subjects were included in this study. The majority of patients (64%) were female, 64% were white, and the median (IQR) age was 48 (35-62) years old. A total of 51% were on Medicaid and 69% identified behavioral health issues as their barriers to optimal health care. Between pre- and post-enrollment periods, the median (IQR) monthly rate of 9-1-1 calls, ED visits, and hospital admissions significantly decreased by 0.26 (-0.06, 0.90), 0.25 (-0.08, 0.71), and 0.18 (0.04, 0.53) (p < 0.001 for all).

Conclusions: When health systems in a geographic area share data, they are better able to recognize patterns of overuse, and address them properly. This study demonstrated that a collaborative 90-day intervention identifying super-utilizers reduced the monthly rate of 9-1-1 calls, ED visits, and hospital admissions.