Abstract
BACKGROUND: County-level poverty is associated with higher rates of late-stage breast cancer (LSBC). The role of persistent poverty (> 20% residents in poverty for 30 + years) in breast cancer stage at diagnosis is unclear and may vary by rural/urban status, race, and over time. METHODS: We analyzed county-level data from the Surveillance, Epidemiology, and End Results Program for females (ages 20-74 years) with late-stage (regional and distant) breast cancer from 2017 to 2021 (excluding 2020). We estimated mean rate differences (RDs) in LSBC between persistently poor (n = 156) and non-persistently poor (n = 923) counties using multivariable linear regression models stratified by rural/urban status and race. We used Joinpoint regression analysis to estimate overall annual percent changes (APC) in breast cancer rates by stage from 2004 to 2019, as well as by persistent poverty and rural/urban status. RESULTS: Age-adjusted LSBC rates were higher in persistently poor counties, but differences were diminished in multivariable models (RD = 2.74, 95% CI - 1.6, 7.1). Adjusted results did not differ based on rural/urban status (p(int) = 0.94) or race (p(int) = 0.08). LSBC rates declined from 2004 to 2017 but increased from 2017 to 2019, particularly in persistently poor counties (APC = 5.54, 95% CI 0.06, 8.69). LSBC rates declined from 2004 to 2019 in urban (APC = - 0.87, 95% CI - 1.63, - 0.65) but not rural (APC = - 0.01, 95% CI - 0.34, 0.32) counties. CONCLUSION: Elevated LSBC rates in persistent poverty counties are largely explained by recent poverty and race/ethnicity. Given the rising rates of LSBC in persistent poverty counties, our findings emphasize the importance of addressing breast cancer screening barriers among disadvantaged populations.