Reading layer `igwz-8jzy' from data source 
  `https://data.cityofchicago.org/resource/igwz-8jzy.geojson' 
  using driver `GeoJSON'
Simple feature collection with 77 features and 5 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -87.94011 ymin: 41.64454 xmax: -87.52414 ymax: 42.02304
Geodetic CRS:  WGS 84
Reading layer `qqq8-j68g' from data source 
  `https://data.cityofchicago.org/resource/qqq8-j68g.geojson' 
  using driver `GeoJSON'
Simple feature collection with 1 feature and 4 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -87.94011 ymin: 41.64454 xmax: -87.52414 ymax: 42.02304
Geodetic CRS:  WGS 84

This project is an R coding and data analysis assignment for my graduate level Web GIS & Spatial Data Visualization course (GEO 447). It is intended to practice skills and show proficiency in data collection and visualization, statistical analysis, and analytical writing using R. All data was gathered from the Chicago Community Health Atlas (CCHA).

There are some key limitations on this analysis. This analysis in no way attempts to identify a causal link between any of the social indicators of health or the mortality rates used in the exercise. Rather, this exercise hopes to map the geographic relationships between poverty and drug related and diet related mortality. Areas on the south and west sides of the city have been historically under served. I believe there will be a strong clustering of outcomes in these areas, illustrating a spatial relationship between disadvantage and advantage within the city. I’ve chosen to study households with SNAP and those which qualify for SNAP but don’t receive it, as I’m interested in whether there is a disparity in health outcomes between the two cohorts.

The mortality indicators chosen for this exercise include Drug Induced Overdose Mortality Rate and Diet-Related Mortality Rate. I chose these indicators because I believe both food insecurity and problems with illicit drug mortality are significant issues for Chicago residents living under the poverty line. This exercise does not look to identify a causal link between the mortality indicators and social determinants measured in it. Rather, I have only set out to measure a relationship between them.

This data shows a slow but steady increase in both drug induced mortality per 100,000 people and diet related mortality per 100,000 people from 2010 through 2023. It also shows several clusters of the city where these problems are most pronounced - specifically on the south and west sides. These are areas of the city which have historically suffered from poverty and lack of quality food access.

Drug-Induced Overdose Mortality Rate, (per 100,000), 2019-2023
Diet-Related Mortality Rate (per 100,000)

The two social indicators I’ve chosen are Household poverty with and without SNAP coverage. These are important metrics to measure in isolation from each other, as one shows areas where government assistance is provided, and the other shows areas where many people qualify for government assistance but do not have it, for any number of reasons.

This map shows that Chicago city neighborhoods have moderate access to food stamps, but identifies several areas on the north side which have large clusters of qualifying households without coverage.

Household Poverty Rates, No SNAP Coverage
Household Poverty Without SNAP
Histogram
Histogram

The spatial distribution of mortality rates and social determinants of health in Chicago reveals strong geographic disparities. High rates of diet-related mortality and drug-induced mortality appear in neighborhoods on the South and West sides, aligning with historic disadvantages for neighborhoods in these areas of Chicago. North side neighborhoods do not experience a similar impact from these mortality indicators.

While social indicator maps indicate some neighborhoods on the north side may qualify for more SNAP assistance than they’ve been receiving, the Local Moran’s analysis clearly identifies areas of greatest need on the south and west sides, when taking into account the mortality indicators used in this study. This highlights the geographic inequalities experienced across Chicago, and brings attention to the complex dimensions of those inequalities.

The Bivariate Correlations and LISA analyses suggest there is a correlation in the geographic clustering between these social determinants and mortality outcomes. A strong Bivariate Correlation is usually taken to be positive or negative 0.70 or more. As such, it is clear that a spatial relationship between poverty, food insecurity, and health outcomes exists in these neighborhoods. These patterns show that neighboring communities share similar socioeconomic and health conditions.

Overall, these findings show that geography reinforces social and health disparities across Chicago. The clustering of poverty and poor health outcomes highlights the need for targeted, place-based strategies to reduce inequities across Chicago, especially on the South and West sides.

Local Moran’s I (LISA) Households in Poverty without SNAP
Local Moran’s I (LISA) Households in Poverty with SNAP
Local Moran’s I (LISA) Diet-Induced Mortality
Local Moran’s I (LISA) Drug-Related Mortality