Reducing Hospitalization in Mild Acute Pancreatitis: Results of Long-term Follow-up.

Abstract

Acute pancreatitis (AP) is a common emergency department (ED) diagnosis, amounting to enormous costs. Our previous pilot study demonstrated the feasibility of reducing hospitalization using an ED-based observation pathway. In this follow-up study, we hypothesize that the pathway is durable in clinical practice, outside of research supervision, and patients can safely be managed without hospitalization. We reviewed patients prospectively enrolled in the observation pathway after the end of the pilot study. We compared outcomes to patients enrolled in our pilot study and with a historic cohort of patients admitted with mild AP. Our primary outcome was hospitalization rate during the enrollment period and secondary outcomes included length of stay, 30-day readmissions, mortality, and health care utilization. Over a 2-year period 165 patients met criteria for AP with 118 (71.5%) having mild AP. Fifty-four of 118 patients (45.8%) were enrolled in the observation pathway and of these, 45 patients were discharged from the ED, reducing hospitalization by 31.2%, compared with pilot study (22.2%) and historic cohort (0%) (P<0.05). Median length of stay was shorter [19.9 (observation) vs. 72.0 h (historic cohort), P<0.01]. There were fewer radiographic examinations in the observation cohorts (pilot and current study) than in the historic cohort (P<0.05), with similar 30-day readmissions, and no reported deaths. This follow-up study demonstrates the durability of an observation-based pathway to manage mild AP outside of a research protocol and maintain its ability to reduce hospitalizations without affecting readmission rates or mortality.