Baltimore’s unprecedented overdose crisis has not been suffered equally.

The neighborhoods with the highest overdose rates were often the same ones with the highest rates of poverty, according to a Banner analysis of Maryland autopsy data. The poorest neighborhoods had overdose rates more than 10 times higher than those in the city’s wealthiest.

This analysis looks at where people fatally overdosed, not where they lived. However, the results of this analysis are the same even if you look at where the person lived, meaning people who live in poorer census tracts also overdose at higher rates.

The data, obtained by The Banner through a lawsuit against the Office of the Chief Medical Examiner, helped to show that Baltimore suffered the worst overdose epidemic ever seen in a major American city and took a disproportionate toll on one generation of older Black men in the city.

Each point in the chart above is a census tract in Baltimore, divisions of the city made by the U.S. Census Bureau to display population data. They can generally be referred to as neighborhoods as defined by the U.S. Census Bureau, though they are different from the neighborhoods defined by the city.

The disparity in overdose rates is clearest between neighborhoods with next to no poverty and those where 20% of residents live in poverty. Census Bureau deems a neighborhood to have high poverty if 20% or more of its residents live in poverty.

In neighborhoods with poverty rates higher than 20%, the effect levels off. The pink line is a simple statistical model that shows the effect of poverty on the overdose rate leveling off after 20%. The difference in overdose rates between a neighborhood with no poverty and one with 20% poverty is larger than the difference between a neighborhood with 20% poverty and one with 40%.

Prior reporting showed that Baltimore’s Black community was hit hardest by overdoses.

That difference is reflected in the boxplot chart above, where the black horizontal line represents the median overdose rate for majority Black and non-majority-Black census tracts in the respective boxes.

The boxes surrounding those horizontal lines represent the range from the 25th percentile of overdose rate to the 75th percentile. The left box is taller than the right box, which tells us that majority Black census tracts have more varied overdose rates than non-majority Black tracts, which are more concentrated near zero.

However, the differences between majority Black and non-majority Black tracts shrink substantially when we also account for poverty.

In the boxplot below, affluent Black and white neighborhoods have very similar overdose rate patterns. Both have seen generally lower rates of overdose than their higher-poverty counterparts with the exception of some outliers.

High and extremely high poverty areas have only small differences in median overdose rates between white and Black neighborhoods, but, again, majority Black neighborhoods have a much larger spread. Part of this is because there are many, many more high- and extremely high-poverty census tracts that are majority Black than are majority white.

Poverty is far from the entire story. While the hardest-hit locations in the city are high-poverty areas, four of the top 10 are not. And other socially influenced health outcomes, like homicide and heart disease, are more closely linked to poverty in Baltimore than overdose is.