Targeted Intervention Reduces Use in Frequent Users of Emergency Medical Services

Abstract

Some of the increase in EMS use is driven by a small group of frequent users. Age, housing status, mental illness, and substance abuse have been identified as contributors to frequent use. Creating EMS-based outreach team could be a cost-effective means of reducing frequent use of EMS. We hoped to determine if engagement by the San Francisco Fire Department's (SFFD) Homeless Outreach and Medical Emergency (HOME) Team is associated with decreased EMS use, and to identify predictors of responsiveness to HOME team intervention. Patients who called EMS four or more times per month were identified from SFFD records and through referrals from EMS providers and EDs. Intervention began with a traditional paramedic assessment. Once patients were determined to not have any immediate medical need, the HOME TEAM Interventional Technique was employed. This includes direct transport to medical detoxification services, San Francisco Sobering Center, Psychiatric Emergency Services at San Francisco General Hospital (SFGH), substance abuse treatment clinics, primary care clinics, and access to case management and housing. We retrospectively analyzed data collected by HOME Team regarding EMS utilization in the 7 months before and after their intervention. Change in mean EMS use was the primary outcome measure. Social and demographic characteristics were gathered from the Department of Public Health EMR. Of the 59 subjects included in the final analysis, 14 subjects (nonresponders) had increased EMS use (mean increase 9.71, STD 9.75), and 45 subjects (responders) had decreased use (mean decrease 16.29, STD 18.88). The mean decrease in EMS use for all subjects was 10.1 calls per 7 months period (95% CI 4.4 - 15.9). None of the predictors were statistically associated with nonresponsiveness. HOME team intervention was associated with decrease in overall EMS use. While the response was not uniform, none of the tested factors predicted nonresponse. However, sample size limitations and high prevalence of some factors may have limited this finding. EMS-based outreach programs have the potential to reduce EMS utilization by frequent users.TablePatient Demographic and Social Characteristics.CharacteristicN = 59Sex, Male68%Mean Age in years (IQR)55 (9)Not Housed38.9% (5 missing)Substance Abuse88.9% (5 missing)Psychiatric History83.0% (6 missing)No primary Care Doctor43.4% (6 missing) Open table in a new tab