Disparities and underutilization of MyChart among patients with pancreatic ductal adenocarcinoma

Background: Electronic health record portals such as MyChart support patient engagement, communication, and access to medical information. Although engagement may vary across demographic and socioeconomic groups, the relationship between portal utilization and clinical characteristics in oncology populations is poorly understood. In pancreatic ductal adenocarcinoma (PDAC), where timely diagnosis and treatment are critical, it is essential to understand how portal engagement relates to health outcomes. Objective: We aimed to determine predictors of MyChart utilization among patients with PDAC and associations between portal engagement and stage at diagnosis. We hypothesized that greater portal use is protective against advanced stage diagnosis. Methods: We conducted a retrospective cohort study of patients diagnosed with PDAC at the University of Chicago between 2012-2022. MyChart activity was abstracted and categorized into appointment/scheduling, health maintenance/test results, clinical messaging/document viewing, eCheck-in/eVisit, medication-related actions, and other categories. Demographic variables included age, sex, race, ethnicity, and Area Deprivation Index (ADI). Clinical variables included stage at diagnosis. We used multivariable logistic models for MyChart use and for pre-diagnosis use predicting stage, and multivariable linear regression for post-diagnosis engagement, modeling log-transformed session counts by stage. Results: Among 1,463 patients with PDAC, 478 used MyChart after diagnosis. Black patients had lower odds of MyChart use compared with white patients (OR 0.66, p = 0.014), and those in the highest-deprivation neighborhoods (ADI 75-100) had substantially lower engagement (OR 0.52, p = 0.006). Age, sex, and ethnicity were not significant predictors of utilization. Higher MyChart utilization after diagnosis was associated with lower odds of advanced-stage disease (OR = 0.82, p = 0.007), and each category of activity showed significant or near-significant inverse associations with advanced stage. High clinic-hours use showed a stronger inverse association with advanced stage (OR 0.78, p = 0.001) than after-hours use (OR 0.86, p = 0.033). Pre-diagnosis MyChart use (n = 101) was also associated with lower odds of advanced-stage presentation (OR 0.59, p = 0.026), though intensity of pre-diagnosis activity was not significantly associated with stage. Conclusions: Our findings highlight clear disparities in MyChart engagement among vulnerable populations. MyChart is underutilized in patients with advanced stage PDAC, both before and after diagnosis. These patterns suggest that patients who interact more frequently with digital health tools may enter the care pathway earlier or maintain stronger connectivity to healthcare services. Addressing disparities in digital engagement may support earlier diagnosis and streamlined cancer care.