Leveraging EMR prompts to promote equitable clinical trial participation in gynecologic oncology: RE-AIM implementation results.

Abstract

e23294 Background: Lack of diversity and low enrollment limit clinical trial success in gynecologic oncology. Providers represent an underutilized resource for trial recruitment. Methods: At a single urban academic institution, gyn onc clinic note templates for new and return visits were modified to include a clinical trial screening prompt (“This patient was considered for clinical trial:” yes/no). Implementation was guided by the RE-AIM framework. A retrospective pre- and post-implementation review evaluated provider screening and trial enrollment. Patients were identified using clinic location and ICD-10 codes for cervical, uterine, and ovarian cancers; all confirmed diagnoses were included. Collected variables included cancer status, age, sex, race/ethnicity, insurance, and provider type. Descriptive statistics summarized patient characteristics. Associations were assessed using Firth penalized logistic regression. Results: Screening increased significantly post-implementation (OR 3.63, 95% CI 1.92–7.12; p < 0.001). African American patients had higher odds than White patients (OR 3.88, 95% CI 1.28–19.1; p = 0.013), indicating improved equity. Screening was higher among recurrence visits versus at the time of new diagnoses (OR 7.22, 95% CI 3.74–14.0; p < 0.001) and lower among surveillance patients (OR 0.09, 95% CI 0.01–0.37; p < 0.001). Compared with ovarian cancer, cervical cancer had lower odds, while corpus uteri cancer showed no difference. Post-implementation, use of the tool was higher for new diagnoses (OR 1.71, 95% CI 1.08–2.71; p = 0.022) and recurrence (OR 4.53, 95% CI 2.27–9.34; p < 0.001) versus surveillance. Conclusions: EMR-based clinical trial prompts significantly increased screening in Gyn Onc clinics. Higher screening among African American patients suggests this low-cost intervention may reduce inequities. Variability by disease status and cancer type identifies targets for ongoing improvement. EMR-integrated prompts offer a scalable strategy to promote equitable trial access. Characteristic N Event N Event % OR 95% CI p-value Race/Ethnicity White 203 2 1.0% — — African American 706 32 4.5% 3.88 1.28, 19.11 0.013 Other 57 3 5.3% 5.18 0.98, 31.75 0.052 Hispanic 41 2 4.9% 5.10 0.77, 33.94 0.087 Unknown 20 2 10% 10.89 1.60, 74.47 0.018 Asian 16 0 0% 2.44 0.02, 31.70 0.606 Diagnosis Code C56.X: Malignant neoplasm of ovary 284 11 3.9% — — C54.X: Malignant neoplasm of corpus uteri 608 30 4.9% 1.25 0.64, 2.61 0.516 C53.X: Malignant neoplasm of cervix uteri 151 0 0% 0.08 0.00, 0.61 0.009 Cancer Status New diagnosis 526 19 3.6% — — Recurrence 98 21 21% 7.22 3.74, 14.04 <0.001 Surveillance 417 1 0.2% 0.09 0.01, 0.37 <0.001 Type of Insurance Private 430 18 4.2% — — Medicaid 156 5 3.2% 0.81 0.28, 2.01 0.664 Medicare 444 18 4.1% 0.97 0.50, 1.88 0.921 Self-pay 11 0 0% 0.97 0.01, 7.97 0.983 Other 2 0 0% 4.46 0.03, 57.45 0.418