Key Points
Question
How do associations between county-level age-adjusted breast cancer mortality and population demographic, environmental, lifestyle, and health care access characteristics vary geographically in the US?
Findings
This cross-sectional study of 2176 US counties found that the statistically significant positive association between obesity and breast cancer mortality was consistent across all counties in the US, but that access to factors in the built environment to support a healthy lifestyle had varying associations with mortality based on the county in which an individual lives.
Meaning
These results suggest that breast cancer mortality in the US can be affected by where individuals live, and that more comprehensive and geographically targeted interventions may lead to healthier communities.
This cross-sectional study examines US county-level data for breast cancer mortality to detail regional variations in the association between mortality and demographic, lifestyle, and clinical variables.
Abstract
Importance
Breast cancer mortality is complex and traditional approaches that seek to identify determinants of mortality assume that their effects on mortality are stationary across geographic space and scales.
Objective
To identify geographic variation in the associations of population demographics, environmental, lifestyle, and health care access with breast cancer mortality at the US county-level.
Design, Setting, and Participants
This geospatial cross-sectional study used data from the Surveillance, Epidemiology, and End Results (SEER) database on adult female patients with breast cancer. Statistical and spatial analysis was completed using adjusted mortality rates from 2015 to 2019 for 2176 counties in the US. Data were analyzed July 2022.
Exposures
County-level population demographics, environmental, lifestyle, and health care access variables were obtained from open data sources.
Main Outcomes and Measures
Model coefficients describing the association between 18 variables and age-adjusted breast cancer mortality rate. Compared with a multivariable linear regression (OLS), multiscale geographically weighted regression (MGWR) relaxed the assumption of spatial stationarity and allowed for the magnitude, direction, and significance of coefficients to change across geographic space.
Results
Both OLS and MGWR models agreed that county-level age-adjusted breast cancer mortality rates were significantly positively associated with obesity (OLS: β, 1.21; 95% CI, 0.88 to 1.54; mean [SD] MGWR: β, 0.72 [0.02]) and negatively associated with proportion of adults screened via mammograms (OLS: β, −1.27; 95% CI, −1.70 to −0.84; mean [SD] MGWR: β, −1.07 [0.16]). Furthermore, the MGWR model revealed that these 2 determinants were associated with a stationary effect on mortality across the US. However, the MGWR model provided important insights on other county-level factors differentially associated with breast cancer mortality across the US. Both models agreed that smoking (OLS: β, −0.65; 95% CI, −0.98 to −0.32; mean [SD] MGWR: β, −0.75 [0.92]), food environment index (OLS: β, −1.35; 95% CI, −1.72 to −0.98; mean [SD] MGWR: β, −1.69 [0.70]), exercise opportunities (OLS: β, −0.56; 95% CI, −0.91 to −0.21; mean [SD] MGWR: β, −0.59 [0.81]), racial segregation (OLS: β, −0.60; 95% CI, −0.89 to −0.31; mean [SD] MGWR: β, −0.47 [0.41]), mental health care physician ratio (OLS: β, −0.93; 95% CI, −1.44 to −0.42; mean [SD] MGWR: β, −0.48 [0.92]), and primary care physician ratio (OLS: β, −1.46; 95% CI, −2.13 to −0.79; mean [SD] MGWR: β, −1.06 [0.57]) were negatively associated with breast cancer mortality, and that light pollution was positively associated (OLS: β, 0.48; 95% CI, 0.24 to 0.72; mean [SD] MGWR: β, 0.27 [0.04]). But in the MGWR model, the magnitude of effect sizes and significance varied across geographical regions. Inversely, the OLS model found that disability was not a significant variable for breast cancer mortality, yet the MGWR model found that it was significantly positively associated in some geographical locations.
Conclusions and Relevance
This cross-sectional study found that not all social determinants associated with breast cancer mortality are spatially stationary and provides spatially explicit insights for public health practitioners to guide geographically targeted interventions.
Introduction
Breast cancer is the leading cause of cancer-related deaths among women in the US.1 Biological and behavioral determinants of breast cancer mortality are generally known and have guided successful interventions and prevention programs that target individuals at risk.2,3,4 However, due to the complex interrelation between individual and contextual determinants, geographic disparities in breast cancer mortality remain difficult to address.5,6
While traditional regression approaches, commonly used in urban health research, have been useful in identifying determinants of breast cancer mortality, they are limited in that they assume spatial stationarity, meaning that one measure is used to describe the association between the independent and response variable for the entire area under study. Toward addressing this assumption, spatial approaches such as geographically weighted regression (GWR)7 and geographical random forest (GRF)8 compute local associations or the relative importance of variables and breast cancer mortality for each geographic unit within the study area.9,10,11 However, these approaches disregard the possibility that variables affecting breast cancer likely manifest at different spatial scales. For example, on a smaller scale, neighborhoods may have varying degrees of access to exercise opportunities. On a larger scale, states may fund different programs that support remission care for uninsured individuals.
The spatial heterogeneity of breast cancer mortality across the US (Figure 1A) presents an opportunity to explore the contextual and environmental variables that might give rise to such spatial disparities and the potential for nonstationarity in these data across space and scales. One such approach, multiscale geographically weighted regression (MGWR), is an extension of GWR that allows for the association between determinants and breast cancer mortality to vary both across geographic space and at different scales.12 Therefore, the objective of this geospatial cross-sectional study is to identify county-level social determinants of health including population demographics, environment, lifestyle, health care access, and pollutant variables using MGWR to address both spatial heterogeneity and the effects of scale on breast cancer mortality. We focus primarily on age-adjusted female breast cancer mortality as our dependent variable, which normalizes county mortality rates based on age characteristics of the corresponding county using 2000 US census data.13,14 The goal of this study is to enable location-specific interventions that can be addressed at various levels of public health.
Figure 1. Female Breast Cancer Mortality Rates From 2015 to 2019 and Spatial Clusters and Outliers.
Mortality rate is defined as the number of deaths per 100 000 women per year. Cluster type refers to features of counties from Local Moran I statistics surrounded by counties with alike features, and outliers as counties surrounded by counties with different features—eg, high-high clusters indicate counties with high breast cancer mortality rates surrounded by counties that also had high rates, and high-low outliers indicate counties with high breast cancer mortality surrounded by counties with low rates.
Methods
Source of Data
Outcome
For each US county, excluding Alaska and Hawaii, age-adjusted female breast cancer mortality rates from 2015 to 2019 were retrieved from the Surveillance, Epidemiology, and End Results (SEER) database version 8.4.0.115 in June of 2022 (Figure 1A). Female breast cancer mortality rate is defined as the number of deaths per 100 000 women per year and age-adjusted rates are standardized to the 2000 US population.14 Approximately one-third of the 3108 counties that make up the contiguous US (932 counties for results on age-adjusted rates) had no reported data, which is standard practice for counties reporting less than 10 deaths from 2015 to 2019. Thus, these counties were excluded from the analysis. Since all data were publicly available and deidentified, neither informed consent nor institutional review board approval was required. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.
Independent Variables
County-level data were retrieved for 57 social determinants selected a priori for their known association with breast cancer incidence and mortality (eTable 1 in Supplement 1). Independent variables were collected from the Social Vulnerability Index,16 County Health Rankings & Roadmaps,17 OpenStreetMap,18 raw points of interest data from SafeGraph, NASA Black Marble,19 the US National Land Cover Data set,20,21 and ClinicalTrials.gov (eAppendix in Supplement 1).22 Variables were subclassified into 5 main categories: access to health care (7 variables), sociodemographics of the population (24 variables), lifestyle (5 variables), physical environment (15 variables), and pollutant (6 variables).
Of the 57 total variables, 24 variables were removed due to collinearity (r > 0.6 or variance inflation factor above 3.0) (eTable 1 and eFigure 1 in Supplement 1).23 The remaining 33 variables were evaluated using a leaps algorithm24 in R version 2.2.1 (R Project for Statistical Computing) to determine the best subset of variables. When using age-adjusted female breast cancer mortality, the final variable set contained 18 variables (Table 1 and Table 2). We note that none of the variables from the pollutant category were selected in the final model due to poor predictive capability. Some variables were log transformed to improve model convergence. All variables were scaled to have a mean of zero with an SD of 1.
Table 1. Multivariable Linear Regression Results Using Age-Adjusted Female Breast Cancer Mortality Rates as Dependent Variable.
| County variable (N = 2174)a | Standardized β | SE (95% CI) | P value |
|---|---|---|---|
| Intercept | 32.67 | 0.23 (32.23 to 33.11) | <.001 |
| Lifestyle | |||
| Smoking (% adults) | −0.65 | 0.17 (−0.98 to −0.32) | <.001 |
| Obesity (% adults) | 1.21 | 0.17 (0.88 to 1.54) | <.001 |
| Food environment index | −1.35 | 0.19 (−1.72 to −0.98) | <.001 |
| Long commute (% workers) | 0.08 | 0.13 (−0.17 to 0.33) | .56 |
| Exercise opportunities (% population) | −0.56 | 0.18 (−0.91 to −0.21) | .002 |
| Population demographics | |||
| Unemployment (% population aged ≥16 y) | −0.20 | 0.16 (−0.51 to 0.11) | .22 |
| Segregation (total population:White ratio) | −0.60 | 0.15 (−0.89 to −0.31) | <.001 |
| Disability (% population) | 0.18 | 0.17 (−0.15 to 0.51) | .28 |
| Income inequality (ratio 80th:20th percentile) | −0.15 | 0.15 (−0.44 to 0.14) | .31 |
| Access to health care | |||
| Uninsured (% population) | −0.32 | 0.15 (−0.61 to −0.03) | .03 |
| Mammograms (% adults screened) | −1.27 | 0.22 (−1.70 to −0.84) | <.001 |
| Mental health care physicians (ratio to total population) | −0.93 | 0.26 (−1.44 to −0.42) | <.001 |
| Primary care physicians (ratio to total population) | −1.46 | 0.34 (−2.13 to −0.79) | <.001 |
| Hospital access (No. per capita) | −0.26 | 0.15 (−0.55 to 0.03) | .09 |
| Environment | |||
| Mean radiance (Watts × cm−2 × sr−1) | 0.48 | 0.12 (0.24 to 0.72) | <.001 |
| Transit access (No. stops per capita) | −0.43 | 0.13 (−0.68 to −0.18) | .001 |
| Proportion of natural land per county | 0.02 | 0.14 (−0.25 to 0.29) | .91 |
| Grocery stores (No. per capita) | 0.52 | 0.19 (0.15 to 0.89) | .006 |
Basis for comparison between states included in parentheses.
Table 2. Multiscale Geographically Weighted Regression Results Using Age-Adjusted Female Breast Cancer Mortality Rates as the Dependent Variable.
| County variable (N = 2174)a | Mean standardized (SD) β estimate | Range | Counties with significant result, %b | Bandwidth |
|---|---|---|---|---|
| Intercept | 22.76 (1.05) | 19.19 to 26.78 | 100 | 89 |
| Lifestyle | ||||
| Smoking (% adults) | −0.75 (0.92) | −2.99 to 1.50 | 16.3 | 259 |
| Obesity (% adults) | 0.72 (0.02) | 0.67 to 0.75 | 100 | 2179 |
| Food environment index | −1.69 (0.70) | −2.85 to 0.36 | 80.3 | 389 |
| Long commute (% workers) | 0.20 (0.04) | 0.06 to 0.23 | 0 | 2179 |
| Exercise opportunities (% population) | −0.59 (0.81) | −3.46 to 1.05 | 13.5 | 280 |
| Population demographics | ||||
| Unemployment (% population aged ≥16 y) | −0.13 (0.06) | −0.31 to −0.10 | 0 | 2179 |
| Segregation (total population:White ratio) | −0.47 (0.41) | −1.81 to 0.02 | 22.6 | 851 |
| Disability (% population) | 0.40 (0.17) | 0.08 to 0.61 | 45.0 | 1672 |
| Income inequality (80th:20th percentile ratio) | −0.30 (0.01) | −0.34 to −0.27 | 0 | 2179 |
| Access to health care | ||||
| Uninsured (% population) | −0.17 (0.02) | −0.23 to −0.12 | 0 | 2179 |
| Mammograms (% adults screened) | −1.07 (0.16) | −1.29 to −0.74 | 100 | 2059 |
| Mental health care physicians (ratio to population) | −0.48 (0.92) | −2.40 to 2.21 | 14.0 | 418 |
| Primary care physicians (ratio to population) | −1.06 (0.57) | −2.07 to −0.01 | 40.6 | 925 |
| Hospital access (No. per capita) | −0.14 (0.03) | −0.18 to −0.04 | 0 | 2179 |
| Environment | ||||
| Mean radiance (Watts × cm−2 × sr−1) | 0.27 (0.04) | 0.22 to 0.37 | 42.4 | 2179 |
| Transit access (No. stops per capita) | −0.31 (0.02) | −0.34 to −0.24 | 83.0 | 2179 |
| Proportion of natural land per county | 0.10 (0.01) | 0.08 to 0.11 | 0 | 2179 |
| Grocery store access (No. per capita) | 0.33 (0.05) | 0.26 to 0.46 | 0 | 2179 |
Basis for comparison between states included in parentheses.
Threshold for significance P < .05.
Statistical Analysis
To better visualize the spatial patterns of breast cancer mortality across the US, a cluster and outlier analysis of the age-adjusted breast cancer mortality rates were computed using a Local Moran I approach.25 Next, excluding counties with missing data, a linear regression model (OLS) was fit to the county-level data in which all variables were regressed against age-adjusted female breast cancer mortality. Linear models assume that a variable’s magnitude of effect is constant across the sample space.
To assess whether the effects of our independent variables vary geographically across the US, an MGWR model was also computed using the same variable sets. Unlike a linear model, MGWR allows the strength and direction of effect to vary across the sample space—potentially revealing county-specific variation in trends.12 Formally, MGWR computes a local regression model for every county (i) in the data set by borrowing data from other surrounding counties (j) that fall within county i’s neighborhood. The number of nearest neighbors from which data will be borrowed (that comprise j) is referred to as the bandwidth. MGWR recognizes that not all relationships occur at the same spatial scale. Thus, the bandwidth size varies for each variable, based on an optimization algorithm.
The MGWR model is expressed as:
 |
where βbwj is the estimation of the coefficient for county i and bwj is the optimal bandwidth size. The resulting bandwidths provide important information on the scale at which certain processes occur, thus indicating spatial nonstationarity. Smaller bandwidths indicate more local variation. Whereas larger values indicate a more global response similar to OLS. All statistical and spatial analysis were computed in ArcGIS Pro version 3.1.0 (Esri). Statistical significance was determined by 95% CIs. See eMethods in Supplement 1 for additional theoretical and technical details of the analyses.
Results
The Local Moran I analysis identified spatial clusters and outliers of counties based on their age-adjusted breast cancer mortality rates (Figure 1B). A belt of counties with high breast cancer mortality rates (high-high cluster) was found to stretch from Kansas through Oklahoma east to Arkansas, Louisiana, Mississippi, Alabama, and Georgia and then up through South and North Carolina to Virginia. Another high-high cluster was observed along the borders of Kentucky, West Virginia, and Ohio. In contrast, clusters of counties with low breast cancer mortality rates (low-low cluster) were observed in California, Arizona, much of the Northeast, and parts of the Midwest. The map also highlights counties that have statistically high or low breast cancer mortality rates relative to their spatial neighbors (low-high outlier, high-low outlier). For example, Buffalo County, New York, has a much higher breast cancer mortality rate than the surrounding counties. In another example, Madison County, Tennessee, has a much lower breast cancer mortality rate than the surrounding counties.
We attempt to explain these spatial patterns of breast cancer mortality by comparing the coefficient of determination between risk factors and mortality rates using a conventional linear regression and the MGWR model. The MGWR was better at explaining the association between independent variables and breast cancer mortality rates across the US for adjusted mortality rates. For example, the linear model for age-adjusted female breast cancer mortality rates yielded an adjusted R2 = 0.17 compared with an adjusted R2 = 0.28 for the MGWR model (Tables 1 and 2, respectively).
A positive, statistically significant association between obesity and breast cancer mortality was observed in both the OLS (β, 1.21; 95% CI, 0.88 to 1.54; P < .001) and the MGWR (mean [SD] β, 0.72 [0.02]). Similarly, a negative and statistically significant association between the proportion of adults screened with mammograms and breast cancer mortality was observed in the OLS (β, −1.27; 95% CI, −1.70 to −0.84; P < .001) and the MGWR (mean [SD] β, −1.07 [0.16]). Furthermore, given that there are only small changes in the coefficients for obesity (Figure 2A) and proportion of adults screened for mammograms (Figure 2B), the MGWR results indicate that that the effects of these variables on mortality are spatially stationary.
Figure 2. Multiscale Geographically Weighted Regression for the Association Between Age-Adjusted Female Breast Cancer Mortality and Spatially Stationary Variables.
Panel A, obesity and breast cancer mortality are positively associated; the association is spatially stationary across the US, although the effect size of the association is greater in the South. Panel B, mammogram testing and breast cancer mortality are negatively associated; the association is spatially stationary across the US, although the effect size is observed in the East.
The OLS and MGWR model agreed that in general breast cancer mortality was significantly negatively associated with smoking (OLS: β, −0.65; 95% CI, −0.98 to −0.32; P < .001; mean [SD] MGWR β, −0.75 [0.92]), food environment index (OLS: β, −1.35; 95% CI, −1.72 to −0.98]; P < .001; mean [SD] MGWR: β, −1.69 [0.70]), exercise opportunities (OLS: β, −0.56; 95% CI, −0.91 to −0.21; P = .002; mean [SD] MGWR: β, −0.59 [0.81]), segregation (OLS: β, −0.60; 95% CI, −0.89 to −0.31; P < .001; mean [SD] MGWR: β, −0.47 [0.41]), mental health care physician ratio (OLS: β, −0.93; 95% CI, −1.44 to −0.42; P < .001; mean [SD] MGWR: β, −0.48 [0.92]), and primary care physician ratio (OLS: β, −1.46; 95% CI, −2.13 to −0.79; P < .001; mean [SD] MGWR: β, −1.06 [0.57]), while positively associated with light pollution (mean radiance) (OLS: β, 0.48; 95% CI, 0.24 to 0.72; P < .001; mean [SD] MGWR: β, 0.27 [0.04]) (Tables 1 and 2).
However, while the OLS found that these variables are significant factors associated with breast cancer mortality overall, MGWR showed that they are only significant in some geographical locations. For example, where obesity and mammogram testing have a significant association with mortality in 100% of US counties, smoking had a significant effect in only 16.3%, food environment index in 80.3%, segregation in 22.6%, mental health care physician ratio in 14.0%, primary care physician ratio in 40.6%, and light pollution in 42.4%. Furthermore, the MGWR revealed that the magnitude of effect size of these variables varied from county to county, as demonstrated by the larger standard deviation of the beta coefficients and the smaller bandwidth sizes for these variables (Table 2). Thus, the association between these variables and breast cancer mortality can be considered spatially nonstationary with effects that vary regionally in scale. For example, the food environment index was not significantly associated with breast cancer mortality in the western US (Figure 3A). Yet, in most of the southern and eastern US, the food environment index was positively associated with breast cancer mortality with coefficients ranging from −1.55 to −2.85. This association had the largest effect sizes (ranging from β = −2.36 to β = −2.85) in Louisiana, Mississippi, Arkansas, and Alabama as well as North Carolina and parts of South Carolina and Virginia. Additionally, where access to exercise opportunities and breast cancer mortality was not significant for most of the US, a positive association with coefficients ranging from −1.30 to −3.46 was found in central US and Florida (Figure 3B).
Figure 3. Multiscale Geographically Weighted Regression for the Association Between Age-Adjusted Female Breast Cancer Mortality and Spatially Nonstationary Variables.
Panel A, the association between food environment index and breast cancer mortality was spatially nonstationary, with the largest negative effect sizes in Louisiana, Arkansas, Alabama, North and South Carolina, and Virginia. B, the association between food environment index and breast cancer mortality is spatially nonstationary, with the largest negative effect sizes in central US and Florida.
Finally, where OLS estimated that disability was not significant, the MGWR estimated that it was significant in 45% of counties and that on average it was positively associated with breast cancer mortality (mean [SD] MGWR β, 0.4 [0.17]). In contrast, where OLS found a negative association between the uninsured and breast cancer mortality (β, −0.32; 95% CI, −0.61 to −0.03; P = .03), the MGWR found that the coefficients for this variable were not statistically significant for any county in the US. The 2 models agreed that unemployment, long commute, income inequality, number of hospitals, and proportion of natural land were not significantly associated with breast cancer mortality at the county level, with MGWR results not significant for 100% of counties. The methodology was also applied using unadjusted breast cancer mortality rates (2015-2019) as an outcome for comparison. The findings are consistent across both adjusted and unadjusted breast cancer mortality rates (eMethods in Supplement 1).
Discussion
To our knowledge, this is the first study applying an MGWR model to assess how associations between breast cancer mortality and county-level social determinants vary across space and scale in the US. Based on the SEER age-adjusted rates collected between 2015 and 2019, breast cancer–associated mortality rates differed considerably across the US (Figure 1A and B). Alabama is a clear example of the diverse outcomes experienced by breast cancer patients based on their geographic location even under unified state programs. While the northern part of the state showed significant variation in age-adjusted mortality rates between counties, the southern part of the state displayed more homogeneous rates.
While the MGWR was better at explaining age-adjusted breast cancer mortality in general, both models showed a significant negative and spatially stationary association between breast cancer mortality and access to mammogram screening. Similarly, county-level obesity emerged as a variable with a positive association with breast cancer mortality that had a stationary effect across the US, but that the association had slightly higher effect sizes in the Southern states. Association between obesity and breast cancer incidence and mortality have been thoroughly examined in epidemiological, clinical, and preclinical studies.26,27,28,29
Of interest, lifestyle factors that affect obesity, like the food environment index and exercise opportunities were also negatively associated with breast cancer mortality in the OLS and MGWR models. However, their effects were spatially nonstationary with regional-scale variation (Figure 3). For example, food environment index, a variable that combines both physical and financial access to healthy foods, effect sizes for the association with reduced mortality were especially pronounced in areas that have previously been reported as cancer hot spots for non-Hispanic Black women,30 such as areas along the Mississippi river, rural southern Virginia, and North Carolina (Figure 1B). Thus, our results indicate that more comprehensive and geographically targeted public health programs with a combined approach that seeks to both increase access to healthy and nutritional foods in underserved areas31 and modify eating habits32,33,34 could support filling the cancer disparity gap in this region. This highlights the importance of considering spatial nonstationarity of cancer mortality rates.
Access to physical exercise opportunities also emerged as a nonstationary risk factor associated with breast cancer mortality (Figure 3). The beneficial effect of exercise and physical activity have been thoroughly described in the context of breast cancer incidence and mortality, including in individuals harboring genomic alterations of the BRCA1 and BRCA2 genes.35,36,37,38,39,40 Meta-analyses have provided suggestive evidence that links availability of and engagement in physical activity with improved outcome for breast cancer patients.41,42,43 Our MGWR model results indicated that access to exercise opportunities has a positive impact on breast cancer survivorship in areas highly populated by Latino and indigenous Native American communities, like New Mexico, Texas, and Florida, and at the 4 corners between New Mexico, Colorado, and Arizona. Understanding the effects of physical activity on breast cancer mortality in women of different ethnic background may open new opportunities for developing culturally specific educational programs.44,45,46,47
Strengths and Limitations
While numerous studies have assessed social determinants of breast cancer mortality, most previous analyses were either limited to specific geographic areas or were conducted under the assumption that mortality determinants are spatially stationary. Our analysis provides unique insights on the spatial and scale-dependent relationship between health determinants and breast cancer mortality.
Because breast cancer death rates are relatively rare events in the general population, a few limitations of this study need to be addressed. While the SEER database remains the most reliable and comprehensive source of cancer-related mortality data across the US, to protect patients’ confidentiality, mortality rates are not reported for less populated areas where death totals do not reach the minimum reporting threshold. While we tested several approaches for imputing missing data, we found that imputation risked inflating mortality rates in counties with small populations or decreased the spatial variance that is observed in the nonimputed data. Thus, our analysis is biased toward counties that have 10 or more deaths in 5 years and may affect our findings.
In addition, most variables included in our analysis were measured at the county level, not specifically in women at risk for or affected by breast cancer, which may have affected our estimates. We also note that our final MGWR produces a moderate coefficient of determination, especially using the age-adjusted mortality rates as a dependent variable. This is likely due to the complexity of breast cancer mortality and determinants, producing variation that is difficult to capture in models. This is reflected in similar studies that use county-level data that also report moderate model performance,10,11 but in general, especially when using individual level data, studies often choose not to report it at all.
Even with these limitations, the MGWR model demonstrated that factors known to be associated with breast cancer have heterogenous effects across geographic regions. By accounting for the inherent spatial distribution of risk factors, population diversity, and their effect on mortality, the MGWR model provides unique opportunities for identifying trends and conceiving policies and health interventions that target specific population characteristics.
Conclusions
In this cross-sectional study, we found county-level age-adjusted breast cancer mortality rates were significantly positively associated with obesity and negatively associated with proportion of adults screened via mammograms, and that this association was spatially stationary. Smoking, food environment index, exercise opportunities, segregation, mental health care physician ratio, and primary care physician ratio were negatively associated with breast cancer mortality, and light pollution was positively associated. However, the MGWR revealed that the magnitude of effect and significance of these variables varied across geographical regions.
Devising new approaches to address health disparities is a growing priority in cancer research. It is well known that health disparities are driven by complex and often interrelated factors. Untangling these complex relationships requires innovative and multidisciplinary approaches able to tie place-specific factors with disease-related outcomes. The MGWR approach proposed brought a novel perspective for capturing the spatial interrelations between individuals and contextual factors on a large geographic scale. 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.
eAppendix. Data Processing and Aggregation
eTable 1. Variables, Description, Source, and Year
eFigure 1. Variable Selection Process
eMethods.
eTable 2. OLS Results Using Crude Female Breast Cancer Mortality as the Dependent Variable
eTable 3. MGWR Results Using Crude Female Breast Cancer Mortality as the Dependent Variable
eFigure 2. MGWR Standardized Beta Coefficients Describing the Association Between Crude Breast Cancer Mortality (Person Years at Risk) and Obesity, Food Index, Mammogram Testing, and Female Population Over Age 65
eFigure 3. MGWR Standardized Beta Coefficients Describing the Association Between Crude Breast Cancer Mortality (Person Years at Risk) and Exercise, Primary Care Physicians’ Ratio, Unemployment, and Income Inequality
eReferences.
Data Sharing Statement
References
- 1.Akram M, Iqbal M, Daniyal M, Khan AU. Awareness and current knowledge of breast cancer. Biol Res. 2017;50(1):33. doi: 10.1186/s40659-017-0140-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gerend MA, Pai M. Social determinants of Black-White disparities in breast cancer mortality: a review. Cancer Epidemiol Biomarkers Prev. 2008;17(11):2913-2923. doi: 10.1158/1055-9965.EPI-07-0633 [DOI] [PubMed] [Google Scholar]
- 3.Musgrove EA, Sutherland RL. Biological determinants of endocrine resistance in breast cancer. Nat Rev Cancer. 2009;9(9):631-643. doi: 10.1038/nrc2713 [DOI] [PubMed] [Google Scholar]
- 4.Coughlin SS. Social determinants of breast cancer risk, stage, and survival. Breast Cancer Res Treat. 2019;177(3):537-548. doi: 10.1007/s10549-019-05340-7 [DOI] [PubMed] [Google Scholar]
- 5.Garland FC, Garland CF, Gorham ED, Young JF. Geographic variation in breast cancer mortality in the United States: a hypothesis involving exposure to solar radiation. Prev Med. 1990;19(6):614-622. doi: 10.1016/0091-7435(90)90058-R [DOI] [PubMed] [Google Scholar]
- 6.Chien LC, Yu HL, Schootman M. Efficient mapping and geographic disparities in breast cancer mortality at the county-level by race and age in the U.S. Spat Spatiotemporal Epidemiol. 2013;5:27-37. doi: 10.1016/j.sste.2013.03.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brunsdon C, Fotheringham S, Charlton M. Geographically weighted regression. J R Stat Soc Ser Stat. 1998;47(3):431-443. doi: 10.1111/1467-9884.00145 [DOI] [PubMed] [Google Scholar]
- 8.Georganos S, Grippa T, Niang Gadiaga A, et al. Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling. Geocarto Int. 2021;36(2):121-136. doi: 10.1080/10106049.2019.1595177 [DOI] [Google Scholar]
- 9.Tian N, Wilson JG, Zhan FB. Spatial association of racial/ethnic disparities between late-stage diagnosis and mortality for female breast cancer: where to intervene? Int J Health Geogr. 2011;10(1):24. doi: 10.1186/1476-072X-10-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Banerjee S, Jones M. Using geographically weighted linear regression for county-level breast cancer modeling in the United States. medRxiv. Posted online April 4, 2022. doi: 10.1101/2022.03.28.22272969 [DOI]
- 11.Dong W, Bensken WP, Kim U, et al. Variation in and factors associated with US county-level cancer mortality, 2008-2019. JAMA Netw Open. 2022;5(9):e2230925. doi: 10.1001/jamanetworkopen.2022.30925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fotheringham AS, Yang W, Kang W. Multiscale geographically weighted regression (MGWR). Ann Am Assoc Geogr. 2017;107(6):1247-1265. doi: 10.1080/24694452.2017.1352480 [DOI] [Google Scholar]
- 13.Anderson RN, Rosenberg HM. Report of the second workshop on age adjustment. Vital Health Stat Ser 4. 1998;30:I-VI, 1-37. [PubMed] [Google Scholar]
- 14.Anderson RN, Rosenberg HM. Age standardization of death rates: implementation of the year 2000 standard. Natl Vital Stat Rep. 1998;47(3):1-16, 20. [PubMed] [Google Scholar]
- 15.Surveillance, Epidemiology, and End Results Program . US Mortality Data, 1969-2020. SEER US data sets. Published online 2019. Accessed June 2022. https://seer.cancer.gov/mortality/
- 16.Agency for Toxic Substances and Disease Registry . CDC/ATSDR Social Vulnerability Index. ATSDR website. Last reviewed July 12, 2023. Accessed July 1, 2022. https://www.atsdr.cdc.gov/placeandhealth/svi/index.html
- 17.University of Wisconsin Population Health Institute . County health rankings & roadmaps. Accessed July 1, 2022. https://www.countyhealthrankings.org/
- 18.OpenStreetMap contributors. OpenStreetMap United States of America. Published online 2022. Accessed July 1, 2022. https://download.geofabrik.de/north-america/us.html
- 19.National Aeronautics and Space Administration . NASA Black Marble. Accessed July 1, 2022. https://blackmarble.gsfc.nasa.gov/
- 20.Multi-Resolution Land Characteristics Consortium . National Land Cover Database. 2021. Accessed July 1, 2022. https://www.mrlc.gov/data
- 21.Dewitz J. National Land Cover Database (NLCD) 2019 Products: US Geological Survey data release. June 2021. Accessed July 1, 2022. doi: 10.5066/P9KZCM54 [DOI]
- 22.US National Library of Medicine . ClinicalTrials.gov. 2019. Accessed July 1, 2022. https://clinicaltrials.gov/
- 23.Yoo W, Mayberry R, Bae S, Singh K, Peter He Q, Lillard JW Jr. A study of effects of multicollinearity in the multivariable analysis. Int J Appl Sci Technol. 2014;4(5):9-19. [PMC free article] [PubMed] [Google Scholar]
- 24.Lumley T. Package ’leaps’—R. October 13, 2022. Accessed February 1, 2023. https://cran.r-project.org/web/packages/leaps/leaps.pdf
- 25.Anselin L. Local indicators of spatial association—LISA. Geogr Anal. 1995;27(2):93-115. doi: 10.1111/j.1538-4632.1995.tb00338.x [DOI] [Google Scholar]
- 26.Chan DSM, Vieira R, Abar L, et al. Postdiagnosis body fatness, weight change and breast cancer prognosis: Global Cancer Update Program (CUP global) systematic literature review and meta-analysis. Int J Cancer. 2023;152(4):572-599. doi: 10.1002/ijc.34322 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lee E, Kady V, Han E, Montan K, Normuminova M, Rovito MJ. Healthy eating and mortality among breast cancer survivors: a systematic review and meta-analysis of cohort studies. Int J Environ Res Public Health. 2022;19(13):7579. doi: 10.3390/ijerph19137579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xie Z, Xie W, Liang Y, et al. Associations of obesity, physical activity, and screening with state-level trends and racial and ethnic disparities of breast cancer incidence and mortality in the US. JAMA Netw Open. 2022;5(6):e2216958. doi: 10.1001/jamanetworkopen.2022.16958 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Luís C, Dias J, Firmino-Machado J, et al. A retrospective study in tumour characteristics and clinical outcomes of overweight and obese women with breast cancer. Breast Cancer Res Treat. 2023;198(1):89-101. doi: 10.1007/s10549-022-06836-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Moore JX, Royston KJ, Langston ME, et al. Mapping hot spots of breast cancer mortality in the United States: place matters for Blacks and Hispanics. Cancer Causes Control. 2018;29(8):737-750. doi: 10.1007/s10552-018-1051-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Fong AJ, Lafaro K, Ituarte PHG, Fong Y. Association of living in urban food deserts with mortality from breast and colorectal cancer. Ann Surg Oncol. 2021;28(3):1311-1319. doi: 10.1245/s10434-020-09049-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Satia JA, Galanko JA, Neuhouser ML. Food nutrition label use is associated with demographic, behavioral, and psychosocial factors and dietary intake among African Americans in North Carolina. J Am Diet Assoc. 2005;105(3):392-402. doi: 10.1016/j.jada.2004.12.006 [DOI] [PubMed] [Google Scholar]
- 33.Bovell-Benjamin AC, Dawkin N, Pace RD, Shikany JM. Use of focus groups to understand African-Americans’ dietary practices: implications for modifying a food frequency questionnaire. Prev Med. 2009;48(6):549-554. doi: 10.1016/j.ypmed.2009.03.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Savoca MR, Arcury TA, Leng X, et al. The diet quality of rural older adults in the South as measured by healthy eating index-2005 varies by ethnicity. J Am Diet Assoc. 2009;109(12):2063-2067. doi: 10.1016/j.jada.2009.09.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bandera EV, Alfano CM, Qin B, Kang DW, Friel CP, Dieli-Conwright CM. Harnessing nutrition and physical activity for breast cancer prevention and control to reduce racial/ethnic cancer health disparities. Am Soc Clin Oncol Educ Book. 2021;41:1-17. doi: 10.1200/EDBK_321315 [DOI] [PubMed] [Google Scholar]
- 36.Chan DSM, Abar L, Cariolou M, et al. World Cancer Research Fund International: Continuous Update Project—systematic literature review and meta-analysis of observational cohort studies on physical activity, sedentary behavior, adiposity, and weight change and breast cancer risk. Cancer Causes Control. 2019;30(11):1183-1200. doi: 10.1007/s10552-019-01223-w [DOI] [PubMed] [Google Scholar]
- 37.Larson EA, Dalamaga M, Magkos F. The role of exercise in obesity-related cancers: current evidence and biological mechanisms. Semin Cancer Biol. 2023;91:16-26. doi: 10.1016/j.semcancer.2023.02.008 [DOI] [PubMed] [Google Scholar]
- 38.Chen X, Wang Q, Zhang Y, Xie Q, Tan X. Physical activity and risk of breast cancer: a meta-analysis of 38 cohort studies in 45 study reports. Value Health. 2019;22(1):104-128. doi: 10.1016/j.jval.2018.06.020 [DOI] [PubMed] [Google Scholar]
- 39.Zagalaz-Anula N, Mora-Rubio MJ, Obrero-Gaitán E, Del-Pino-Casado R. Recreational physical activity reduces breast cancer recurrence in female survivors of breast cancer: a meta-analysis. Eur J Oncol Nurs. 2022;59:102162. doi: 10.1016/j.ejon.2022.102162 [DOI] [PubMed] [Google Scholar]
- 40.Bucy AM, Valencia CI, Howe CL, et al. Physical activity in young BRCA carriers and reduced risk of breast cancer. Am J Prev Med. 2022;63(5):837-845. doi: 10.1016/j.amepre.2022.04.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Cariolou M, Abar L, Aune D, et al. Postdiagnosis recreational physical activity and breast cancer prognosis: Global Cancer Update Programme (CUP Global) systematic literature review and meta-analysis. Int J Cancer. 2023;152(4):600-615. doi: 10.1002/ijc.34324 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Friedenreich CM, Stone CR, Cheung WY, Hayes SC. Physical activity and mortality in cancer survivors: a systematic review and meta-analysis. J Natl Cancer Inst Cancer Spectr. 2019;4(1):pkz080. doi: 10.1093/jncics/pkz080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Lahart IM, Metsios GS, Nevill AM, Carmichael AR. Physical activity, risk of death and recurrence in breast cancer survivors: a systematic review and meta-analysis of epidemiological studies. Acta Oncol. 2015;54(5):635-654. doi: 10.3109/0284186X.2014.998275 [DOI] [PubMed] [Google Scholar]
- 44.Gonzalo-Encabo P, Wilson RL, Kang DW, et al. Reducing Metabolic Dysregulation in Obese Latina and/or Hispanic Breast Cancer Survivors Using Physical Activity (ROSA) trial: a study protocol. Front Oncol. 2022;12:864844. doi: 10.3389/fonc.2022.864844 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Pinkston CM, Baumgartner RN, Connor AE, Boone SD, Baumgartner KB. Physical activity and survival among Hispanic and non-Hispanic white long-term breast cancer survivors and population-based controls. J Cancer Surviv. 2015;9(4):650-659. doi: 10.1007/s11764-015-0441-3 [DOI] [PubMed] [Google Scholar]
- 46.Cao Y, Baumgartner KB, Visvanathan K, Boone SD, Baumgartner RN, Connor AE. Ethnic and biological differences in the association between physical activity and survival after breast cancer. NPJ Breast Cancer. 2020;6:51. doi: 10.1038/s41523-020-00194-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Nechuta SJ, Lipworth L, Chen WY, Shu XO, Zheng W, Blot WJ. Physical activity in association with mortality among Black women diagnosed with breast cancer in the Southern Community Cohort Study. Cancer Causes Control. 2023;34(3):277-286. doi: 10.1007/s10552-022-01663-x [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eAppendix. Data Processing and Aggregation
eTable 1. Variables, Description, Source, and Year
eFigure 1. Variable Selection Process
eMethods.
eTable 2. OLS Results Using Crude Female Breast Cancer Mortality as the Dependent Variable
eTable 3. MGWR Results Using Crude Female Breast Cancer Mortality as the Dependent Variable
eFigure 2. MGWR Standardized Beta Coefficients Describing the Association Between Crude Breast Cancer Mortality (Person Years at Risk) and Obesity, Food Index, Mammogram Testing, and Female Population Over Age 65
eFigure 3. MGWR Standardized Beta Coefficients Describing the Association Between Crude Breast Cancer Mortality (Person Years at Risk) and Exercise, Primary Care Physicians’ Ratio, Unemployment, and Income Inequality
eReferences.
Data Sharing Statement