A Novel Evaluation of Patient Socioeconomic Characteristics That Predict Clinical Trial Enrollment.
AI interpretation is pending for this paper.
Open original publication →What the AI sees
Not AI summarized yet.
Research significance
Pending deeper interpretation.
Source abstract
BACKGROUND: Clinical trial participation offers patients with cancer access to cutting-edge therapies and generates critical evidence that informs and improves future cancer treatment guidelines. However, disparities in trial enrollment hinder inclusivity and limit the generalizability of trial findings. This study investigates patient-specific factors influencing therapeutic trial enrollment to identify and address barriers to equitable participation. METHODS: This retrospective cohort study included patients with cancer aged 18 to 75 years diagnosed between 2018 and 2021 at a regional hospital system in Northeast Ohio. Patients' electronic medical records were linked to the individual-level LexisNexis Social Determinants of Health data. Propensity score matching was performed to balance trial and nontrial patients with respect to age, biological sex, and cancer type. Key predictors of trial enrollment were identified using a systematic variable selection process, the Boruta machine learning algorithm, and stepwise logistic regression. Geographic variation in enrollment were examined at the ZIP code level. RESULTS: Among 12,630 patients with cancer, 649 (5.1%) participated in treatment-related trials. After 1:7 propensity score matching, the Boruta algorithm identified income, property ownership, and household factors as the strongest predictors of enrollment, whereas race/ethnicity, college attendance, and distance to closest relatives were less influential. Logistic regression revealed that non-Hispanic Black individuals (odds ratio [OR], 0.70; 95% CI, 0.54-0.89) and those in other minority groups (OR, 0.57; 95% CI, 0.36-0.88) were less likely to enroll compared with non-Hispanic White individuals. After adjusting for income, non-Hispanic Black race was no longer significant (OR, 0.85; 95% CI, 0.64-1.11). Higher estimated income was associated with increased enrollment (up to 67% higher odds), whereas patients insured through Medicaid were 29% less likely to enroll than those with private insurance (OR, 0.71; 95% CI, 0.53-0.93). CONCLUSIONS: Findings from this study suggest that financial resources may play a greater role in clinical trial enrollment than traditional demographic characteristics. Improving equity in trial enrollment and enhancing the generalizability of trial findings will likely require increased financial and structural support for diverse and vulnerable patient populations.