JMIR Publications

ABSTRACT

Background:

Uninterrupted delivery of radiation therapy is critical to cancer treatment quality. Unplanned radiation treatment interruptions (RTI) compromise tumor control and survival. Although previous studies have explored clinical and sociodemographic factors influencing treatment adherence, the role of psychosocial distress and related symptoms remains unclear

Objective:

To identify patient-reported psychosocial predictors of RTI from patient-reported distress screening data integrated with clinical and socioeconomic correlates via traditional and machine learning-enhanced modeling

Methods:

This institutional cohort study included 1,953 patients who underwent external beam radiation therapy between August 2020 and May 2024. Patient demographics, clinical characteristics, treatment details, and baseline distress levels were extracted from a hospital-level electronic health record (EHR) data repository. Distress was assessed using the Distress Thermometer developed by the National Comprehensive Cancer Network (NCCN). These data were linked with census tract–level socioeconomic indicators. Predictors of treatment interruption (measured as missed treatment days) were evaluated using negative binomial regression and an Extreme Gradient Boosting (XGBoost) model with a Poisson objective function

Results:

The study cohort missed a mean of 1.19 ± 2.18 treatment days (range: 0–15). Higher distress score, presence of learning or communication barriers, and specific cancer sites—particularly head and neck and genitourinary malignancies—were significantly associated with increased interruption risk in the negative binomial regression model. The XGBoost model confirmed these features as top contributors to prediction performance, while yielding an incrementally lower root mean squared error of 2.20 (versus 2.24 for the regression model)

Conclusions:

Patient-reported distress and health educational barriers strongly associate with treatment interruptions, underscoring the importance of incorporating psychosocial assessments into radiation oncology workflows. Both parametric and machine learning models provided stable and closely aligned predictions of missed radiation treatment days. Combining traditional regression and machine learning approaches improved interpretability and predictive accuracy, supporting continued integration of complementary modeling into the development of early intervention strategies to enhance treatment adherence