Exploring the Association Between Structural Racism and Mental Health: Geospatial and Machine Learning Analysis.

BACKGROUND: Structural racism produces mental health disparities. While studies have examined the impact of individual factors such as poverty and education, the collective contribution of these elements, as manifestations of structural racism, has been less explored. Milwaukee County, Wisconsin, with its racial and socioeconomic diversity, provides a unique context for this multifactorial investigation. OBJECTIVE: This research aimed to delineate the association between structural racism and mental health disparities in Milwaukee County, using a combination of geospatial and deep learning techniques. We used secondary data sets where all data were aggregated and anonymized before being released by federal agencies. METHODS: We compiled 217 georeferenced explanatory variables across domains, initially deliberately excluding race-based factors to focus on nonracial determinants. This approach was designed to reveal the underlying patterns of risk factors contributing to poor mental health, subsequently reintegrating race to assess the effects of racism quantitatively. The variable selection combined tree-based methods (random forest) and conventional techniques, supported by variance inflation factor and Pearson correlation analysis for multicollinearity mitigation. The geographically weighted random forest model was used to investigate spatial heterogeneity and dependence. Self-organizing maps, combined with K-means clustering, were used to analyze data from Milwaukee communities, focusing on quantifying the impact of structural racism on the prevalence of poor mental health. RESULTS: While 12 influential factors collectively accounted for 95.11% of the variability in mental health across communities, the top 6 factors-smoking, poverty, insufficient sleep, lack of health insurance, employment, and age-were particularly impactful. Predominantly, African American neighborhoods were disproportionately affected, which is 2.23 times more likely to encounter high-risk clusters for poor mental health. CONCLUSIONS: The findings demonstrate that structural racism shapes mental health disparities, with Black community members disproportionately impacted. The multifaceted methodological approach underscores the value of integrating geospatial analysis and deep learning to understand complex social determinants of mental health. These insights highlight the need for targeted interventions, addressing both individual and systemic factors to mitigate mental health disparities rooted in structural racism.

The journey to overdose: Using spatial social network analysis as a novel framework to study geographic discordance in overdose deaths.

INTRODUCTION: Drug overdose deaths are often geographically discordant (the community in which the overdose death occurs is different from the community of residence). Thus, in many cases there is a journey to overdose. METHODS: We applied geospatial analysis to examine characteristics that define journeys to overdoses using Milwaukee, Wisconsin, a diverse and segregated metropolitan area in which 26.72 % of overdose deaths are geographically discordant, as a case study. First, we deployed spatial social network analysis to identify hubs (census tracts that are focal points of geographically discordant overdoses) and authorities (the communities of residence from which journeys to overdose commonly begin) for overdose deaths and characterized them according to key demographics. Second, we used temporal trend analysis to identify communities that were consistent, sporadic, and emergent hotspots for overdose deaths. Third, we identified characteristics that differentiated discordant versus non-discordant overdose deaths. RESULTS: Authority communities had lower housing stability and were younger, more impoverished, and less educated relative to hubs and county-wide numbers. White communities were more likely to be hubs, while Hispanic communities were more likely to be authorities. Geographically discordant deaths more commonly involved fentanyl, cocaine, and amphetamines and were more likely to be accidental. Non-discordant deaths more commonly involved opioids other than fentanyl or heroin and were more likely to be the result of suicide. CONCLUSION: This study is the first to examine the journey to overdose and demonstrates that such analysis can be applied in metropolitan areas to better understand and guide community responses.

Impact of the COVID-19 Pandemic on Opioid Overdose Deaths: a Spatiotemporal Analysis.

The effects of the opioid crisis have varied across diverse and socioeconomically defined urban communities, due in part to widening health disparities. The onset of the COVID-19 pandemic has coincided with a spike in drug overdose deaths in the USA. However, the extent to which the impact of the pandemic on overdose deaths has varied across different demographics in urban neighborhoods is unclear. We examine the influence of COVID-19 pandemic on opioid overdose deaths through spatiotemporal analysis techniques. Using Milwaukee County, Wisconsin as a study site, we used georeferenced opioid overdose data to examine the locational and demographic differences in overdose deaths over time (2017-2020). We find that the pandemic significantly increased the monthly overdose deaths. The worst effects were seen in the poor, urban neighborhoods, affecting Black and Hispanic communities. However, more affluent, suburban White communities also experienced a rise in overdose deaths. A better understanding of contributing factors is needed to guide interventions at the local, regional, and national scales.

Examining Opioid Overdose Deaths across Communities Defined by Racial Composition: a Multiscale Geographically Weighted Regression Approach.

To provide data that can guide community-targeted practices, policies, and interventions in urban metropolitan areas, we used geospatial analysis to examine the community-level opioid overdose death determinants and their spatial variation across a study area. We obtained spatial datasets containing multiple, high-quality measures of socioeconomic conditions, public health status, and demographics for analysis and visualization in geographic information systems. We employed a multiscale modeling approach (multiscale geographically weighted regression; MGWR) to provide a comprehensive and robust analysis of opioid overdose death determinants, explain how geospatial patterns vary across scales across Milwaukee County in 2019, and examine the differential influence of factors locally, regionally, and globally. We subsequently examined how associations varied with the racial/ethnic composition of communities by dividing Milwaukee County into White-majority, Black-majority, and Hispanic-majority regions according to census data and conducting separate, independent modeling processes. Overall, the multiscale model explained 83% of opioid overdose death variability across neighborhoods in Milwaukee County using 12 selected variables. Statistical analysis and geovisualization of patterns, trends, and clusters using MGWR unveiled dramatic racialized health disparities in Milwaukee, showing how factors that influenced opioid overdose deaths varied across diverse communities in Milwaukee. The observed geographic variation in relationships included the impact of naloxone availability and incarceration rates on overdose deaths with pronounced differences between White communities and communities of color. Understanding, community-level factors that contribute to overdose risk should guide targeted community-level solutions. Overall, our findings demonstrate the value of precision epidemiology using MGWR analysis for defining and guiding responses to public health challenges.

Geospatial analysis of misinformation in COVID-19 related tweets.

COVID-19 has emerged as a global pandemic caused by its highly transmissible nature during the incubation period. In the absence of vaccination, containment is seen as the best strategy to stop virus diffusion. However, public awareness has been adversely affected by discourses in social media that have downplayed the severity of the virus and disseminated false information. This article investigates COVID-19 related Twitter activity in May and June 2020 to examine the origin and nature of misinformation and its relationship with the COVID-19 incidence rate at the state and county level. A geodatabase of all geotagged COVID-19 related tweets was compiled. Multiscale Geographically Weighted Regression was employed to examine the association between social media activity and the spatial variability of disease incidence. Findings suggest that MGWR could explain 80% of the COVID-19 incidence rate variations indicating a strong spatial relationship between social media activity and spread of the Covid-19 virus. Discourse analysis was conducted on tweets to index tweets downplaying the pandemic or disseminating misinformation. Findings indicate that sites of Twitter misinformation showed more resistance to pandemic management measures in May and June 2020 later experienced a rise in the number of cases in July.