TY - JOUR
T1 - The joint effects of census tract poverty and geographic access on late-stage breast cancer diagnosis in 10 US States
AU - Henry, Kevin A.
AU - Sherman, Recinda
AU - Farber, Steve
AU - Cockburn, Myles
AU - Goldberg, Daniel W.
AU - Stroup, Antoinette M.
N1 - Copyright © 2013 Elsevier Ltd. All rights reserved.
PY - 2013/5
Y1 - 2013/5
N2 - This study evaluated independent and joint effects of census tract (CT) poverty and geographic access to mammography on stage at diagnosis for breast cancer. The study included 161,619 women 40+ years old diagnosed with breast cancer between 2004 -2006 in ten participating US states. Multilevel logistic regression was used to estimate the odds of late-stage breast cancer diagnosis for the entire study population and by state. Poverty was independently associated with late-stage in the overall population (poverty rates >20% OR=1.30, 95% CI=1.26- 1.35) and for 9 of the 10 states. Geographic access was not associated with late-stage diagnosis after adjusting for CT poverty. State-specific analysis provided little evidence that geographic access was associated with breast cancer stage at diagnosis, and after adjusting for poverty, geographic access mattered in only 1 state. Overall, compared to women with private insurance, the adjusted odds ratios for late stage at diagnosis among women with either no insurance, Medicaid, or Medicare were 1.80 (95% CI = 1.65, 1.96), 1.75 (95% CI = 1.68, 1.84), and 1.05 (95% CI 1.01, 1.08), respectively. Although geographic access to mammography was not a significant predictor of late-stage breast cancer diagnosis, women in high poverty areas or uninsured are at greatest risk of being diagnosed with late-stage breast cancer regardless of geographic location and may benefit from targeted interventions.
AB - This study evaluated independent and joint effects of census tract (CT) poverty and geographic access to mammography on stage at diagnosis for breast cancer. The study included 161,619 women 40+ years old diagnosed with breast cancer between 2004 -2006 in ten participating US states. Multilevel logistic regression was used to estimate the odds of late-stage breast cancer diagnosis for the entire study population and by state. Poverty was independently associated with late-stage in the overall population (poverty rates >20% OR=1.30, 95% CI=1.26- 1.35) and for 9 of the 10 states. Geographic access was not associated with late-stage diagnosis after adjusting for CT poverty. State-specific analysis provided little evidence that geographic access was associated with breast cancer stage at diagnosis, and after adjusting for poverty, geographic access mattered in only 1 state. Overall, compared to women with private insurance, the adjusted odds ratios for late stage at diagnosis among women with either no insurance, Medicaid, or Medicare were 1.80 (95% CI = 1.65, 1.96), 1.75 (95% CI = 1.68, 1.84), and 1.05 (95% CI 1.01, 1.08), respectively. Although geographic access to mammography was not a significant predictor of late-stage breast cancer diagnosis, women in high poverty areas or uninsured are at greatest risk of being diagnosed with late-stage breast cancer regardless of geographic location and may benefit from targeted interventions.
KW - Access to care
KW - Breast cancer
KW - Cancer stage at diagnosis
KW - Cancer survival
KW - Urban/rural
UR - http://www.scopus.com/inward/record.url?scp=84874574121&partnerID=8YFLogxK
U2 - 10.1016/j.healthplace.2013.01.007
DO - 10.1016/j.healthplace.2013.01.007
M3 - Article
C2 - 23454732
SN - 1353-8292
VL - 21
SP - 110
EP - 121
JO - Health and Place
JF - Health and Place
ER -