Improving the quality and guideline concordance of multidisciplinary gastric cancer care in a safety net hospital.

314 Background: Gastric cancer is a leading cause of cancer death worldwide and disproportionally affects people of East Asian and Latin American ethnicities. This study aimed to implement and evaluate an intervention to improve the quality and guideline concordance of multidisciplinary gastric cancer care at a public safety net hospital, Zuckerberg San Francisco General Hospital (ZSFG). Methods: A quality improvement (QI) intervention was developed and evaluated by pre-post comparison of clinical data. (1) An in-depth chart review of all patients with gastric adenocarcinoma in 2017-2019 (n=17) evaluated concordance with evidence-based guidelines and quality metrics. (2) Multidisciplinary meetings identified barriers to high quality care and established targets for improvement. (3) A clinical algorithm tailored for ZSFG was disseminated to standardize gastric cancer management. (4) A post-intervention chart review was conducted of all patients from 2021-2022 (n=22). (5) Descriptive statistics and pre-post comparison of QI metrics with Chi square and Kruskal Wallis tests were performed. Results: Pre-intervention clinical data and multidisciplinary input identified gaps in care: variation in provider knowledge and practices, rapidly evolving guidelines, limited access to advanced diagnostic modalities, and non-standardized surgical documentation. After implementation of a standardized approach to address these gaps, pre-post improvement was seen in rates of tumor board presentation (36.3% vs. 85.7%, P= 0.033), use of endoscopic ultrasound (9.09% vs. 64.2%, P= 0.017), pre-treatment nodal staging (27.2% vs. 78.5%, P= 0.030), receipt of neoadjuvant chemotherapy (20% vs. 80%, P=0.007 ), documentation of lymphadenectomy type (25% vs. 100%, P= 0.015), and a trend toward reduced time to PET/CT scan (median, 8.5 weeks pre vs 7.1 weeks post, P= 0.268). Conclusions: A QI intervention was successfully implemented to standardize gastric cancer care at ZSFG. Significant improvements were seen in all targeted metrics except time to outside PET scan. Potential confounders affecting pre-post care quality included personnel changes in a small workforce and the COVID-19 pandemic. Nevertheless, our intervention led to sustained improvements in practice patterns. We aim to replicate this success with other malignancies. Category Pre-Intervention Target Post-Intervention P value* Tumor Board% cM0 cases presented 36% (n=11) >90% 86% (n=14) 0.033 Endoscopic ultrasound (EUS)% cM0 cases referred for clinical staging 9% (n=11) >90% 64% (n=14) 0.017 Nodal staging (with EUS or PET) 27% (n=11) N/A 79% (n=14) 0.030 Receipt of neoadjuvant chemotherapy% eligible patients 20% (n=10) >80% 80% (n=10) 0.007 Surgery documentationLymphadenectomy type (i.e., D1 or D2) 25% (n=8) 100% 100% (n=7) 0.015 Time from diagnosis to PET/CT (median) 8.5 wks <5 wks 7.1 wks 0.268 *Chi square test, Kruskal-Wallis test.