JMIR Publications

ABSTRACT

Background:

Breast cancer screening plays a pivotal role in early detection and subsequent effective management of the disease, significantly impacting patient outcomes and survival rates.

Objective:

This study aims to assess breast cancer screening rates nationwide in the United States (U.S.) and investigate the impact of social determinants of health on these screening rates

Methods:

Data on mammography screening for the years 2018 and 2020 at the census tract level were collected from the Behavioral Risk Factor Surveillance System (BRFSS). We developed a large-scale dataset of social determinants of health with 13 variables for 72,337 census tracts. Spatial analysis, employing Getis-Ord Gi statistics, was used to identify clusters of high and low breast cancer screening rates. Next, we implemented a random forest (RF) model and compared its performance with linear regression (LR) and support vector machine (SVM). Shapley Additive exPlanations (SHAP) values were then used to assess variable importance and directions of influence. Geospatial analysis revealed elevated screening rates in the eastern and northern U.S., while central and Midwestern areas exhibited lower rates.

Results:

The RF model demonstrated superior performance compared to LR and SVM. SHAP values indicated that the percentage of the Black population, the number of mammography facilities within a 10-mile radius, and the percentage of the population with at least a bachelor's degree were the most influential variables, all positively associated with mammography screening rates.

Conclusions:

These findings underscore the significance of social factors and the accessibility of mammography services in explaining the variability of breast cancer screening in the U.S. emphasizing the need for targeted policy interventions in areas with relatively lower screening rates.