Implementing a discharge intervention to reduce 30-day readmission and mortality in patients with high-risk cancer: A real-world prospective quality improvement study

• Published in: Journal of clinical oncology Vol. 40; no. 28_suppl; p. 329
• Main Authors: Chin, Michelle M., Jones, Ashlie R, Steyl, Allison, Geddie, Patricia, Cruz, Zilipah, Bean, Joanne A, Cooper, Wendy S, Vu, Tek, Serafimovska, Daniela, Warner, Terra, Nanda, Akash, Avgeropoulos, Nicholas George, Stallard, Susan, Saif, Wasif M., Shoup, Margo, Laughlin, Amy Iarrobino
• Format: Journal Article
• Language: English
• Published: 01.10.2022
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/JCO.2022.40.28_suppl.329

329 Background: One in four patients admitted with metastatic cancer are readmitted within 30 days. We created a bundled multipronged intervention to reduce 30-day readmission, unplanned (ED) visits, and mortality among high-risk stage IV cancer patients. Methods: We conducted a prospective quality improvement study to reduce readmissions and 30-day mortality among high-risk oncology patients. Baseline patients were identified 1/2022-3/2022. The electronic medical record (Epic) was leveraged to automatically identify patients admitted with cancer diagnosis and set risk factors (readmission, low albumin, fluids/antibiotics, head imaging, low oxygen saturation). Dashboard was reviewed three times per week and stage IV solid malignancy confirmed. Starting 3/2022, intervention included three components: coordination of care (CC), patient education and symptom management (PESM), and post-discharge follow-up (PDF). In CC, a dedicated scheduler coordinated visits with the primary oncology team within 7 business days, with options for palliative care, nutrition, and integrative medicine. In PESM, the attending physician was notified, referred to oncology nurse navigation and encouraged to use the oncology urgent care. In PDF, patients were contacted after discharge to collect hospital readmission, unplanned ED visit, mortality outcome and satisfaction. Patients discharged with hospice were excluded. Control charts were used to assess special cause and chi-square to compare pre and post intervention cohorts. Results: As of 6/2022, 36 baseline and 125 patients have been identified. Of 117 discharged patients, 23 were discharged on hospice, of the remaining 94, 98% were contacted by the concierge scheduler and 53% received an appointment within 7 days. Among the intervention patients, the most common tumor type was lung (n = 16), prostate (n = 14), and colon (n = 13). The most common reason for admission was pain (n = 30), and respiratory/dyspnea (n = 21). During the baseline period, mean 30-day readmission rate was 25.5% and 30-day mortality 19.6%. Preliminary data reveal early improvement in readmission and 30-day mortality (Table). Conclusions: We have demonstrated that identifying high-risk cancer patients using the electronic medical record is feasible and identifies a group at high risk for readmission and mortality. By prospectively identifying patients, we were able to provide a multi-pronged intervention at a tertiary hospital.[Table: see text]