5606 Background: The National Comprehensive Cancer Network (NCCN) recommends that patients with endometrial cancer under age 50 undergo germline genetic testing. Despite this recommendation, completion rates are suboptimal for patients with gynecologic malignancies. The reasons for failure to complete testing are unknown, but oncology providers may not refer patients to genetic counselors, who frequently order this testing. Best practice advisories (BPAs) in electronic medical records can alert providers of patients who meet guideline-based treatments and testing criteria. It is unknown how oncology providers interact with a BPA indicating patients with endometrial cancer meet NCCN germline testing criteria. The objective of this study is to evaluate genetic counseling referral patterns after implementing a BPA. Methods: Our team developed a BPA to prompt oncology providers to order genetic counseling referrals for patients diagnosed with endometrial or uterine cancer under age 50. We reviewed all provider-BPA interactions between go-live (January 2022) and January 2023. Patient demographic, histopathologic, and cancer treatment data were collected. Provider actions, including BPA cancellation and order placement, were collected. The Chi-Square goodness of fit test was used to identify differences in referral rates by patient or disease characteristics. Results: The BPA displayed 1121 times for 103 patients. Attending physicians received 535 alerts, fellow physicians – 269, and advanced practice providers (APPs) – 317. The median number of alerts per patient was 7 (range 1-87). Of the 103 patients, 23 were excluded from analysis due to no diagnosis of endometrial cancer, patient declined referral, previously completed, or a lack of provider visits within the study period. Of the 80 eligible patients, 36 (45%) received referrals; 18 were placed by attending physicians, 17 by fellows, and one by an APP. Patients under age 40 versus age 40-49 were more likely to be referred (65% vs. 35%; p = 0.011). Patients with active disease were more likely to be referred than those in surveillance (65% vs. 18%; p < 0.001). There were no differences in referral rates by primary language, race, body mass index ≥40, fertility-sparing versus definitive treatment, stage I versus stage II-IV disease, or grade 1-2 versus three tumors. Conclusions: Despite implementing a BPA, only 45% of eligible patients received indicated referrals. Attending and fellow physicians frequently manage patients with active disease and often order referrals; however, APPs frequently manage patients in surveillance and infrequently place referrals, highlighting an opportunity for education. In addition, patients aged 40-49 are less likely to be referred, suggesting a need for new strategies to improve referral rates. Further research is indicated to identify referral barriers for patients undergoing active treatment.
