U.S. National Trends

Breast Cancer Deaths Nationally by Number of Facilities Offering Mammography Services

We can visualize how the number of facilities offering mammographic services relates to breast cancer mortality nationally. For the most part, cities with more facilities per 100,000 people (indicated by larger dot size) have lower mortality rates (yellows and greens), though this is not consistent, as seen in Columbia, SC, for example. In the Shiny app, users may explore trends for specific cities of interest!

Mammography Use Nationally by Number of Facilities Offering Mammography Services

We can also visualize how the number of facilities offering mammographic services relates to mammography use among each city's population. It appears that cities with more mammographic facilities per 100,000 people (larger dot sizes) have higher mammography use (darker colors) nationally.

Breast cancer mortality trends across U.S cities were analyzed to better understand key drivers of disparities in
health outcomes.

Exploring 500 U.S. Cities

A multivariable linear regression model was fit to understand the association between breast cancer mortality and mammography rates, as well as other socioeconomic factors, across 500 U.S. cities. As expected, the fitted model shows an inverse relationship between mammography usage and breast cancer mortality. This can be seen in the maps above where cities in California, New England, and Southern Florida have some of the highest rates of mammography usage, and correspondingly, some of the lowest rates of breast cancer mortality nationwide.

Key Associations:

A 10% increase in mammography usage in a city would reduce breast cancer mortality by 1.8 deaths per 100,000 females (95% CI: 0.34-3.2), holding other factors constant

A city that is more segregated by 10 index units has on average 0.9 more deaths per 100,000 females (95% CI: 0.18 - 1.7), adjusting for all other factors

A city where the diversity index is 10 units higher (i.e. more representation of different races/ethnicities) has an average of 3.5 more breast cancer deaths per 100,000 females (95% CI: 1.48-5.56) than a city with a lower diversity index, after controlling for other factors.

LINEAR REGRESSION MODEL:

(R²_Adjusted= 0.112)

METRICS DEFINITIONS:

% mammography use - age-adjusted mammography use in the female population aged 50-74 years old
racial & ethnic diversity - this index quantifies diversity in a city by considering the racial and ethnic composition of an area. A 0 corresponds to a city where all residents belong to the same group and a score of 100 corresponds to a proportionate representation of each group
neighborhood segregation- this index measures how evenly distributed racial/ethnic groups are across a city’s census tracks, where 0 represents a perfectly even distribution and 100 represents complete segregation
high unemployment- indicates whether the annual unemployment rate in the city exceeds the national median rate of 5.9%

Understanding Racial Inequities

There are clear racial disparities in the rates of breast cancer deaths across the US cities, with Blacks experiencing much higher rates of breast cancer deaths than all other racial groups.

The average mortality rate in Blacks from breast cancer is 31.6 deaths per 100,000 females, which is 30% higher than the national average rate of 24.4 deaths per 100,000 females.

Nationwide, breast cancer mortality is lowest where segregation & diversity are lowest (bottom left quadrant highlighted in orange), and greatest in areas where segregation & diversity are highest (top right quadrant highlighted in purple).

Who can access health care?

The presence of additional mammography sites does not reduce mortality in highly segregated neighborhoods; in fact, mortality is highest in areas with some of the most number of facilities (7.5 - 10 sites per 100,000) if the location is also highly segregated as shown in the heatmap above. In these locations, it is most likely that communities of color do not have convenient access to the facilities, which are often located in racially and economically privileged neighborhoods. However, in areas where there are numerous sites available and the neighborhood segregation level is low (below national average index score of 10.9), breast cancer mortality rates tend to be lower. Notably, in neighborhoods with low levels of segregation, there is a drastic reduction in breast cancer mortality in Blacks as the number of screening facilities in the city increases.

Connections to Current Research

These national trends in racial inequities expand upon previous research conducted on smaller geographical scales. Kreiger et al (2020) found in a recent study that historical redlining from the 1930s has led to present day disparities in breast and lung cancer diagnoses across 28 Massachusetts municipalities, even for individuals who personally benefit from present-day economic and racial privilege. Additionally, in areas where the Home Owners’ Loan Corporation (HOLC) drew boundaries, notable differences in homeownership, credit scores, and segregation persist. Segregated communities often have imbalanced resource allocation, with many communities of color lacking access to health care services, public transportation, and other social determinants known to positively impact health (e.g. public green spaces). Historical redlining and present-day segregation, predatory lending, and racism undoubtedly perpetuate health inequities across U.S. cities (see additional references below).

Our research findings will help inform the development of policies and interventions that can reduce racial/ethnic and socioeconomic disparities in mammography use and breast cancer mortality by drawing attention to specific geographic areas that should be targeted for resource allocation.

ADDITIONAL RESOURCES

Mortgage Foreclosure and Health Disparities: Serial Displacement as Asset Extraction in African American Populations
Racism and Health: Evidence and Needed Research
Elucidating the Role of Place in Health Care Disparities: The Example of Racial/Ethnic Residential Segregation

Limitations in Analysis

The multiple linear regression model is fit using aggregated city level data, and does not provide insight into individual level factors that may contribute to breast cancer mortality. Because of the limited data source, there are possible unmeasured confounding effects at the individual level, such as ease of access to health services and quality of individual care received, which could further explain the association between breast cancer mortality and mammography rates.

Page updated

Google Sites

Report abuse