RS15 - Improved Follow-Up After Hospital Discharge Does Not Impact 30-Day Readmissions After Vascular Procedures

Objectives: We sought to evaluate the integration of an automated patient contact system with a standardized process for in-patient discharge for successful transition to out-patient care.

Methods: All patients discharged from the vascular surgery service who underwent procedures from April 2021 through October 2021 with pre-discharge process improvement (PI) were included and compared with a historical cohort discharged from May 2018 through July 2019 (control). For both cohorts all patients received an automated phone call from a third-party vendor (CipherHealth, New York, NY), with an option to escalate to a clinical nurse. For the 2021 PI cohort, improvements in the discharge process were enacted including standardized in-person discharge instruction by a vascular nurse coordinator, and more timely follow up visit after discharge. Primary outcomes were unplanned readmission within 30 days of discharge and time from discharge to post-discharge visit.

Results: A total of 469 patients who received an automated call (1.4 ± 1.0 days) after discharge were analyzed, 271 in the control group and 198 in the PI group. Mean age was 64 ± 16, and 65% were male. No significant change in overall 30-day readmission rates was observed (11.8% control vs. 10.6% PI; p=0.7). Request for escalation to the clinical nurse was significantly less frequent after intervention (28.0% control vs. 16.7% PI; p=0.004). Escalation was associated with increased 30-day readmission (8.6% vs 20.1%; p=0.001) for all patients. Direct calls were made for patients considered high risk for readmission (2% control vs 43% PI; p< 0.001) regardless of escalation status. On multivariate analysis (TABLE) escalation and direct calls were each independently associated with readmission. Two-thirds of readmissions occurred before scheduled post-discharge visit, with median interval from discharge to readmission of 8 days (IQR 5-13 days). The median interval from discharge to post-discharge visit was markedly improved with PI (24 days [IQR 14-33] compared to 16 days [IQR 11-32]; p=0.007). Compared with control, median days from discharge to visit with escalation was not significantly improved (control: 18 [IQR 14-28] vs. PI: 15 [IQR 9-36]; p=0.3).

Conclusions: Restructuring the discharge process decreased escalation to clinical staff from the automated post-discharge phone call and improved early post-discharge visits. Thirty-day readmission rates remained unchanged despite these efforts, and was increased for patient requesting escalation. These findings support the benefits of a structured discharge approach for post-discharge transition of care and suggest that other actionable factors are required to decrease readmission after vascular procedures.