Preventing 30-day readmissions for patients with cancer: A root-cause analysis.

Abstract

226 Background: The high rate of unplanned 30-day readmissions for patients with cancer is a significant driver of costs and a marker of poor quality. In this study, we analyzed 30-day readmissions at our cancer center to determine causality and propose key drivers to prevent them. Methods: Patients with known cancer who were readmitted to our academic medical center within 30 days of a previous hospitalization were identified in our electronic health record by a third-party algorithm. Among patients with hematologic malignancies, only those undergoing bone marrow transplant care were included. Surveys querying causality and preventability of the readmissions were sent to the patients’ attending oncologists. Electronic chart documentation of readmissions were reviewed by two investigators to assess causality and preventability of each readmission. Results were discussed in focus groups to determine key drivers to prevent 30-day readmissions. Results: 437 readmissions were identified between 9/1/19 and 8/31/20, and 182 readmissions with corresponding surveys completed by their oncologists were identified (Table). Based on survey responses, 30 (16%) of the 182 readmissions were preventable, whereas based on our review, 56 (31%) were preventable. The top three causes of the 56 preventable readmissions were: underutilized ambulatory care (43%), premature discharge (23%), and goals of care discordance (16%). For underutilized ambulatory care, the primary treatments provided during those readmissions were: procedures such as thoracentesis and paracentesis (42%), medication administration for pain or nausea (33%), and infusions or transfusions (25%). Notably, most of these patients either did not attempt to seek outpatient care (42%) or were not able to secure an ambulatory appointment (29%) prior to their readmission. Through focus group discussions, we found that the key drivers to reduce preventable 30-day readmissions at our institution are (1) timely access to outpatient pleural effusion and ascites management, (2) timely access to ambulatory management of cancer-related symptoms (e.g., pain, nausea, weakness), (3) increased systems-wide awareness and utility of avenues of urgent care, and (4) increased palliative care efforts in patients with readmissions. Conclusions: Systematic review of 30-day readmissions revealed a greater than anticipated portion of preventable readmissions. Root-cause analysis yielded key drivers to reduce 30-day readmissions at our cancer center.[Table: see text]