Individual and institutional factors influencing dentists' practice in underserved areas.

Access to dental care is a key determinant of oral health, yet disparities in provider distribution across the United States contribute to inequitable access, particularly in underserved areas. To predict dentists’ likelihood of practicing in underserved settings and identify factors influencing practice location decisions, we developed explainable machine learning models using national data from 56,175 dentists who graduated between 2000 and 2022. We examined 76 predictors including individual- and dental school-level characteristics and defined outcomes as practicing in Federally Qualified Health Centers, dental shortage areas, or rural dental shortage areas. Our models demonstrated strong predictive performance, and the results revealed that general practice specialty, fewer years of experience, non-owner practice status, and demographic factors such as gender and race were strongly associated with practicing in underserved areas. Institutional characteristics, including dental school location and diversity index, also played a significant role. The relationship between educational debt, experience, and practice outcomes varied by practice type and race/ethnicity. These findings highlight the value of explainable machine learning in informing targeted workforce policies that address individual, institutional, and demographic drivers of dentist distribution to improve access to dental care in underserved communities.