US breast cancer mortality rates negatively linked to county-level mammogram access

Key takeaways:

• Obesity and mammogram screening access were linked to breast cancer mortality.
• Smoking, food environmental index, exercise, racial segregation and certain physician ratios were negatively linked to mortality.

County-level age-adjusted breast cancer mortality rates were higher among women with less access to mammogram screening and also among women with obesity, according to study results published in JAMA Network Open.

“The spatial heterogeneity of breast cancer mortality across the U.S. presents an opportunity to explore the contextual and environmental variables that might give rise to such spatial disparities and the potential for non-stationarity in these data across space and scales,” Taylor Anderson, PhD, assistant professor in the department of geography and geoinformation science at George Mason University, and colleagues wrote. “One such approach, multiscale geographically weighted regression, is an extension of geographically weighted regression that allows for the association between determinants and breast cancer mortality to vary both across geographic space and at different scales.”

Anderson and colleagues conducted a geospatial cross-sectional study utilizing data from the Surveillance, Epidemiology and End Results database of adult women with breast cancer. Researchers identified county-level geographic variations in associations of population demographics, environmental, lifestyle and health care access with breast cancer mortality in 2,176 counties in the U.S. Researchers also used multivariable linear regression and multiscale geographically weighted regression to understand the impact and significance of variables across geographic U.S. regions.

Both multivariable linear regression and multiscale geographically weighted regression models showed that county-level age-adjusted breast cancer mortality rates were significantly positively associated with obesity (beta = 1.21 and 0.72, respectively) and negatively associated with the proportion of adults screened through mammograms (beta = –1.27 and –1.07, respectively), the researchers wrote.

The multiscale geographically weighted regression model demonstrated that obesity and mammogram screening access were associated with a stationary effect on breast cancer mortality across the U.S. This model also provided researchers with insights on other county-level factors differentially linked to breast cancer mortality in the U.S.

Both the multivariable linear regression (OLS) and multiscale geographically weighted regression (WGWR) models showed that the following were all negatively associated with breast cancer mortality:

• smoking (OLS: beta, –0.65; WGMR: beta, –0.75);
• food environmental index (OLS: beta, –1.35; WGMR: beta, –1.69);
• exercise opportunities (OLS: beta, –0.56; WGMR: beta, –0.59);
• racial segregation (OLS: beta, –0.6; WGMR: beta, –0.47);
• mental health care physician ratio (OLS: beta, –0.93; WGMR: beta, –0.48); and
• primary care physician ratio (OLS: beta, –1.46; WGMR: beta, –1.06).

In addition, light pollution was positively associated with breast cancer mortality for both the multivariable linear regression and multiscale geographically weighted regression models (beta = 0.48 and 0.27, respectively).

The multiscale geographically weighted regression model showed significant variation in the magnitude of effect sizes across geographical regions in the U.S. The multivariable linear regression model showed disability to not be a significant breast cancer mortality variable. However, the multiscale geographically weighted regression model found a significant positive association with disability in some U.S. regions.

“As suggested by our analysis, this approach may have an unparalleled ability to identify vulnerable populations and geographic areas where targeted interventions may lead to healthier communities,” the researchers wrote.