Association of historic redlining and present-day health in Baltimore · Huang SJ, Sehgal NJ

Future studies

Given these policy questions and some of the study limitations, there are multiple avenues for further study. A future study could use census tract-level mortality data and use census tracts or blocks instead of CSAs as the unit of analysis and could look nationwide beyond Baltimore. It could also use tract or block area data to look at demographic changes [50] to gauge whether or not a neighborhood’s community has been displaced.

Multiple authors [12,21,47] argue that continued diminished investment in neighborhoods due to redlining might be what leads to lower housing values and lower quality neighborhoods today. Gentrification then might be a countervailing trend that brings in resources that might ameliorate any potential neighborhood-level health impacts of redlining. Adding measures of gentrification over time might be a way to account for how neighborhood health status changes over time. However, gentrification also displaces populations from geographically delineated neighborhoods [50]. Redlining may also have current health and non-health impacts through intergenerational effects: housing and neighborhood quality impacts education and wealth accumulation [7,9] which in turn impacts the ability to provide for the next generation. In this case, while gentrification might change a neighborhood’s health status, it would not be due to improvements to infrastructure or wealth accumulation of the impacted community but rather by their replacement. Adding measures of community composition changes could be a way to account for these changes [50]. Additionally, adding analysis of current investment patterns [27] may help elucidate the ways that redlining interacted with population displacement.

Potential pathways from redlining through sustained disinvestment leading to exposure to hazards have been both theorized [9,23] and in some cases explored [27,28]. Future studies could look more closely at neighborhood factors like documented environmental hazards, exposure to policing, healthcare utilization, or mobility patterns to clarify potential causal pathways, including specifically in Baltimore.

Additional topics to explore on neighborhood quality could include the association of HOLC categories with hospital availability and quality, or the response times of EMS. Geographic data on police interactions could also be used to examine the relationship between HOLC categories and where communities are policed. Future studies could also more thoroughly examine childhood health outcomes and factors including asthma. Geocoded social data might also be an avenue for exploration: Chae et al. [52] found a strong geographical association between area racism as the proportion of Google searches containing the “N-word” in 196 designated market areas and Black mortality. Such an analysis could be extended to redlined neighborhoods.