Research Square · Oct 13, 2026
Equity-aware stochastic optimization of mammography screening facility location under demand and capacity uncertainty
By Madadi M, Wilson C, Bauer C et al.
Uneven geographic access to mammography can limit participation in breast cancer screening, particularly in rural and underserved communities. We developed a statewide planning framework that combines geospatial accessibility analysis with stochastic facility-location optimization to identify locations for new mammography facilities under uncertainty in screening demand and existing-facility capacity. The analysis included 5.58 million Texas women aged 40–74 years, 615 existing facilities, and 222 candidate locations. The model first maximized expected screening coverage and then reduced population-weighted differences in potential access across eight race and ethnicity groups without sacrificing coverage. Opening 62 facilities across 58 counties increased expected coverage from 92.47% to 98.09% and reduced expected unmet demand from 420,089 to 106,822 women annually. Between-group variance (BGV) in access decreased by 32%, while access for the worst-served group increased from 0.858 to 0.953 effective screening slots per eligible woman. All selected locations were in baseline access-gap areas; 44% were in nonmetropolitan counties and 39% established the first mammography facility in their county. Explicitly modeling uncertainty reduced expected unmet demand by 29,643 women compared with deterministic planning. These results show that a modest, equity-aware expansion of fixed mammography capacity can substantially widen screening access and narrow both racial and geographic disparities—by placing facilities where under-served populations live rather than where volume is highest—while the most remote communities are better served by mobile screening than by additional fixed sites.
doi.org/10.21203/rs.3.rs-10844614/v1Open accessInterventions