Research has shown that neighborhoods, including the built, social, and natural environments they contain, are influential determinants of health and health inequity. Numerous neighborhood factors have been found to be associated with cancer disparities, including neighborhood socioeconomic status (SES) and walkability. Among these, residential racial segregation has been recognized as a fundamental neighborhood determinant. Racial segregation results in fundamentally different exposures and experiences based on race, impacting health disparities. Further, evidence indicates that there are wide gaps in home ownership and home equity derived wealth by race and ethnicity in the United States.

An important driver of residential segregation, and the economic health of a neighborhood or an individual, is mortgage lending. Mortgage lending bias – the systematic denial of mortgage financing to specific neighborhoods or applicants – can thus have important implications for housing access, wealth accumulation, economic development, and segregation. By promoting neighborhood economic investment differentials, mortgage lending bias is a key upstream driver of housing access as well as other neighborhood health determinants including access to resources, SES, and built environment features such as parks and tree canopy. By promoting differentials in applicant access to mortgage funding based on the race or ethnicity of the applicant, mortgage lending bias has implications for individual wealth, housing tenure, and socioeconomic disparities among racial and ethnic groups.

We developed three measures of mortgage lending bias using a combination of the disease mapping method adaptive spatial filtering (ASF) and logistic regression models predicting application denial:

1. Location bias or redlining: the odds ratio of denial of a mortgage application for a property in a local area, compared to properties across the metropolitan statistical area (MSA)
2. Racial bias: the odds ratio of denial of a mortgage application for a Black/African American applicant, compared to a non-Hispanic white applicant, for a property in the local area.
3. Ethnic bias: the odds ratio of denial of a mortgage application for a Hispanic applicant, compared to a non-Hispanic white applicant, for a property in the local area.

We then examined the relationship between structural racism in housing metrics and breast cancer survival.

Finally, we conducted over 100 interviews with breast cancer survivors in the Milwaukee metropolitan area to understand more about their survivorship experiences, including the impact of discrimination and neighborhood characteristics on their health.

Research Outcomes

• March 2022 (article): Patterns of historical redlining in the 1930’s closely correspond to current patterns of redlining and racial settlement.
  
  
• February 2022 (article): No association between racial bias and incidence of triple negative and luminal A breast cancer in California
  
  
• November 2021 (podcast): JCO After Hours podcast on mortgage lending bias, segregation and cancer disparities
  
  
• September 2021 (article): Older women diagnosed with breast cancer do not move frequently, having implications for the importance of residential history collection in this group
  
  
• June 2021 (article): Older women living in currently redlined areas die faster after breast cancer diagnosis.
  
  
• January 2021 (article): Neighborhood redlining is associated with breast cancer mortality in Atlanta, GA
  
  
• July 2019: Metropolitan area racial bias in mortgage lending is a stronger predictor of racial cancer mortality disparities than racial segregation
  
  
• July 2016: New measures of redlining and racial bias in mortgage lending propose