Automated Post-discharge Phone Call System Identifies Vascular Patients At Risk For Unplanned ED Visits And Readmissions After Hospital Discharge
Austin Holmes, MD, Samantha Stradleigh, MD, Janet Wells, BSN, Misty Humphries, MD, Matthew Mell, MD.
UC Davis, Sacramento, CA, USA.
OBJECTIVES: We sought to evaluate the implementation of an automated patient contact system after hospital discharge for vascular patients.
METHODS: All patients hospitalized on the vascular surgery service from May 2018 through July 2019 were included. Patients discharged to home received an automated phone call from a third-party vendor (CipherHealth, New York, NY). Patients with concerns could opt to speak with a central triage nurse. Unresolved concerns by the triage nurse were escalated to the vascular clinic for resolution. Primary outcome was unplanned readmission or ED visit within 30 days of discharge (ED-READMIT). Data were obtained from the CipherHealth database and EMR. Patients were offered a 5-point Likert survey to evaluate the automated system at the first post-hospital visit.
RESULTS: The cohort comprised 324 patients (mean age 65±15; 62% male) who received an automated call a median 1 (IQR 1-5) day after discharge. Of these, 12.7% (41/324) of patients hung up, 76.0% (246/324) did not ask for additional assistance, 17.0% (56/324) were escalated to the central triage nurse, and 6.8% (22/324) required a second escalation to the vascular clinic. ED-READMIT occurred in 16.0% (52/324) of patients. Request for triage nurse signaled a significant increase in ED-READMIT (25.6% vs. 13%, p=0.008). Further escalation to our clinic showed no difference in ED-READMIT (25.6% vs 27.5%, p=0.8), although it was associated with a non-significant shorter interval from discharge to clinic visit (26 ± 25 days vs. 37 ± 50 days, p=0.06). Surveys were completed by 108 (30%) patients. Of these, 72% agreed or strongly agreed that automated calls were easy to use; however, patient experiencing ED-READMIT scored lower than those who did not (2.4 ± 0.9 vs. 3.1 ± 0.9, p = 0.02).
CONCLUSIONS: Patients who were escalated to a central triage nurse had the highest likelihood of unplanned readmission or ED visit. These findings suggest that an automated system of post-discharge screening has promise for facilitating more timely identification of patients who need more intensive follow-up care. This offers an opportunity to engineer improved care systems to reduce 30 day ED-READMIT without increasing the demands on clinic staff for post-discharge follow-up calls.
Back to 2020 Abstracts