J. Mora1, L. L. Thompson1, M. Hurtado Jr.2, E. Kim3, N. E. Martin3, K. S. Oh4, and R. B. Jimenez5; 1Harvard Radiation Oncology Program, Boston, MA, 2Northwestern University, Chicago, IL, 3Department of Radiation Oncology, Brigham and Women’s Hospital/Dana-Farber Cancer Institute, Boston, MA, 4Department of Radiation Oncology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, 5Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA
Purpose/Objective(s): Cancer care often requires complex treatments and care coordination, which electronic health record (EHR) portals may support timely communication and access to medical information for patients. Whether these resources are used differently by patients remains unclear. In a post-COVID pandemic setting, we sought to assess relationships between English and non-English language preference (ELP; NELP), patient characteristics, and patterns of portal utilization among those receiving radiotherapy. Materials/
Methods: A retrospective review of a prospectively maintained EHR database was queried for patients at least 18 years of age seen in a multisite radiation oncology (RO) enterprise department. Patient-reported sociodemographics including age, sex, marital status, Hispanic ethnicity, race, preferred language, need for interpreter, and education level were abstracted. Portal utilization was defined by EHR portal activation status and number of portal messages sent from/to patients. We assessed relationships between language preference and 1) portal activation; 2) number of patient-provider messages, using logistic and linear regression models, respectively. Models were adjusted for age, sex, and covariates with 2-sided p-value <0.10 in univariate analysis. To address the issue of multiple testing, we used the false discovery (FDR) control method with FDR of 0.15. Results: From 1/1/2022 to 12/31/2023, a total of 20,552 patients had least one radiation treatment or on-treatment visit encounter. Median age was 66 years, 52.7% female. 85.3% white, 4.4% Black, 3.5% Asian, 3.2% other, 3.0% unknown or declined, 0.5% two races or more, 0.1% Native American. 4.3% identified as Hispanic. 93.1% had ELP, 5.9% had NELP. In NELP group, 87.2% were identified needing an interpreter, the most common NELP was Spanish (35.7%). Overall, 84.7% of all patients had portal activation, 79.7% sent messages to providers. Comparing ELP and NELP groups, in a multivariate logistic regression model accounting for age, sex, race, marital status, and education level, NELP group had significantly reduced odds of portal access (OR=0.73, 95% CI: 0.60-0.89, p<0.001). In a linear regression model adjusted for the same set of covariates assessing associations between ELP/NELP and patient-provider messages, NELP group exchanged on average fewer messages with providers than ELP group (Coef. -62.57, 95% CI: -77.10—41.51, p<0.001). Conclusion: In those receiving radiotherapy, NELP group was significant less likely to have portal access or send messages to their RO providers, independent of factors like sex, marital status, race, or education level. Robust interventions to improve language tools in patient portals are needed to bridge this disparity in health information access.