Shedding New Light on Health Disparities, by Neighborhood

Shedding New Light on Health Disparities, by Neighborhood

January 2022

Districts rich and poor, some studded with amenities and some seemingly forgotten, make up the historic city of Philadelphia. Known as the city of neighborhoods, it provides the perfect backdrop for a growing body of health research that seeks to disentwine neighborhood characteristics, elucidating each one’s real effects and suggesting targeted approaches for big societal issues. And it’s here that Mary Regina Boland, MA, MPhil, PhD, FAMIA,  and colleagues have shown that viewing neighborhoods using a combination of data science and informatics approaches, in particular, can shed new light on health disparities and what to do about them.

Her team’s neighborhood studies have gleaned striking findings about a number of pressing issues. “We know that living in neighborhoods that are deprived and that lack resources increases one's risk of many diseases later in life — including Alzheimer’s Disease,” Dr.  Boland says. “And our team has found that COVID-19 also demonstrates neighborhood clustering effects that correlate with deprivation.” That paper, published in May 2021, used neighborhood factors to highlight social determinants that may have underlain SARS-COV-2 positivity rates. While many researchers had already noted that the pandemic more dramatically affected non-white people and people with lower incomes, Dr. Boland and her colleagues wanted to find out precisely which social factors held the most influence and should secure top priority for policy interventions. They developed a way to integrate SARS-CoV-2 data with multiple neighborhood-level factors from the American Community Survey and, and then employed spatial association techniques to tie neighborhood-level factors with test-positivity rates. (Learn more on YouTube about the methods for this study.)

Some of their results confirmed already-familiar trends, such as higher positivity rates among neighborhoods with higher proportions of people who identified as Hispanic/Latinx. But their analysis also yielded a surprising and promising finding: Patients from neighborhoods where higher proportions of individuals (women ages 15-50) receive public assistance were less likely to test positive for SARS-CoV-2. “This association held even when adjusting for other correlated variables (e.g., poverty, income, education, and various race/ethnicity factors),” the authors wrote. The finding showed some of the impact we could achieve by increasing public assistance to at-risk communities, they added: People who receive public assistance are impoverished, but government assistance might enable them to avoid environments that put them at higher risk of infection. 

They used similar techniques in a pair of studies that revealed new insights about maternal outcomes among Black women. In April 2021, Dr. Boland and lead author Jessica Meeker, then an epidemiology PhD candidate, published a study of how neighborhood factors affected severe maternal morbidity — unexpected outcomes of labor and delivery that significantly affect women’s health — among women who gave birth in the University of Pennsylvania Health System over a seven-year period. Using a multivariate model, they were able to examine multiple characteristics of the neighborhoods (in this case, US Census tracts) where those women lived, and they could isolate the health effects of various ones. They found that for every 10 percent increase in people who identified as Black/African American, these negative outcomes increased by 2.4 percent — a conclusion they reached by holding constant the number of crimes and percentage of people in a tract who identified as white. “In effect, we looked only at the ‘true’ relationship between Black neighborhoods and these negative outcomes,” commented Dr. Meeker, who now works for the Epidemic Intelligence Service of the CDC. 

The team’s most recent work added evidence that disadvantaged neighborhoods are linked to a specific, generally undesirable delivery outcome: It found that people living in the most deprived areas were 29 percent more likely to end up with cesarean delivery after having their labor induced, compared to those who lived in areas least deprived. More than 20 percent of the 3.7 million births in the US each year involve induced labor, and a vaginal delivery is often preferred. Drawing again from 2010 to 2017 Penn Medicine data, the researchers examined almost 9000 labor inductions in a cohort from diverse neighborhoods, from deprived to privileged. Finding that race/ethnicity did change the most significant effect size by greater than 10 percent — even when controlling for sociodemographic factors and medical comorbidities — the team opted to include race as a factor at the individual level, in their fully adjusted model. 

While these studies highlighted the negative health effects of racism and the hopeful effects of public assistance, researchers potentially could use similar techniques in future to continue to reveal other systemic inequities and explore additional factors that play a role in disparities, Dr. Boland adds. 

“By studying exposures at the neighborhood level, we can pinpoint the effects of environmental pollutants such as air pollution or car exhaust, and also of social factors like education level or income — and how they relate to disease,” Dr. Boland says.  “Many of these solutions will require the communities themselves to be partners, and some will require funding from governmental agencies, both for the research itself and for related community-based solutions.” 

Working with communities to address issues of inequity is an ever-important goal that the Boland lab hopes to work toward in future as the models her lab develops uncover more information. Models help to pinpoint areas in the city that require more attention and can help identify potential solutions ( such as social support), but working with the communities is of paramount importance, she says.


Mary Regina Boland, Jessica R. Meeker and various colleagues led the three studies cited; see each link for the full list.

Read the study “Association of Neighborhood-Level Factors and COVID-19 Infection Patterns in Philadelphia Using Spatial Regression” in AMIA Summits on Translational Science Proceedings

Read the study “Individual-Level and Neighborhood-Level Risk Factors for Severe Maternal Morbidity” in Obstetrics & Gynocology

Read the study “Neighborhood Deprivation Increases the Risk of Post-Induction Cesarean Delivery” in the Journal of the American Medical Informatics Association.

About Us

To understand health and disease today, we need new thinking and novel science —the kind  we create when multiple disciplines work together from the ground up. That is why this department has put forward a bold vision in population-health science: a single academic home for biostatistics, epidemiology and informatics. 

© 2023 Trustees of the University of Pennsylvania. All rights reserved.. | Disclaimer

Follow Us