Predicting Low-Level Childhood Lead Exposure in Metro Atlanta Using Ensemble Machine Learning of High-Resolution Raster Cells

Low-level lead exposure in children is a major public health issue. Higher-resolution spatial targeting would significantly improve county and state-wide policies and programs for lead exposure prevention that generally intervene across large geographic areas. We use stack-ensemble machine learning, including an elastic net generalized linear model, gradient-boosted machine, and deep neural network, to predict the number of children with venous blood lead levels (BLLs) ≥2 to <5 µg/dL and ≥5 µg/dL in ~1 km2 raster cells in the metro Atlanta region using a sample of 92,792 children ≤5 years old screened between 2010 and 2018. Permutation-based predictor importance and partial dependence plots were used for interpretation. Maps of predicted vs. observed values were generated to compare model performance. According to the EPA Toxic Release Inventory for air-based toxic release facility density, the percentage of the population below the poverty threshold, crime, and road network density was positively associated with the number of children with low-level lead exposure, whereas the percentage of the white population was inversely associated. While predictions generally matched observed values, cells with high counts of lead exposure were underestimated. High-resolution geographic prediction of lead-exposed children using ensemble machine learning is a promising approach to enhance lead prevention efforts.

Moving Morehouse School of Medicine Translation Tx Research through MDTTs-Multidisciplinary Translational Teams

Morehouse School of Medicine (SOM) works to achieve its vision of advancing health equity through conducting transformational, translation science (Tx). Tx describes our translational research continuum, symbolizing a method and scientific philosophy that intentionally promotes and supports convergence of interdisciplinary approaches and scientists to stimulate exponential advances for the health of diverse communities. Morehouse SOM actualizes Tx through multidisciplinary translational teams (MDTTs). We chronicle the identification of MDTTs by documenting formation, composition, functioning, successes, failures, and sustainability. Data and information were collected through key informant interviews, review of research documents, workshops, and community events. Our scan identified 16 teams that meet our Morehouse SOM definition of an MDTT. These team science workgroups cross basic science, clinical, and public health academic departments, and include community partners and student learners. We present four MDTTs, in various stages of progress, at Morehouse SOM and how they are advancing translational research.

Increased Risk of Sub-Clinical Blood Lead Levels in the 20-County Metro Atlanta, Georgia Area-A Laboratory Surveillance-Based Study

Lead (Pb) is a naturally occurring, highly toxic metal that has adverse effects on children across a range of exposure levels. Limited screening programs leave many children at risk for chronic low-level lead exposure and there is little understanding of what factors may be used to identify children at risk. We characterize the distribution of blood lead levels (BLLs) in children aged 0–72 months and their associations with sociodemographic and area-level variables. Data from the Georgia Department of Public Health’s Healthy Homes for Lead Prevention Program surveillance database was used to describe the distribution of BLLs in children living in the metro Atlanta area from 2010 to 2018. Residential addresses were geocoded, and “Hotspot” analyses were performed to determine if BLLs were spatially clustered. Multilevel regression models were used to identify factors associated with clinical BBLs (≥5 µg/dL) and sub-clinical BLLs (2 to <5 µg/dL). From 2010 to 2018, geographically defined hotspots for both clinical and sub-clinical BLLs diffused from the city-central area of Atlanta into suburban areas. Multilevel regression analysis revealed non-Medicaid insurance, the proportion of renters in a given geographical area, and proportion of individuals with a GED/high school diploma as predictors that distinguish children with BLLs 2 to <5 µg/dL from those with lower (<2 µg/dL) or higher (≥5 µg/dL) BLLs. Over half of the study children had BLLs between 2 and 5 µg/dL, a range that does not currently trigger public health measures but that could result in adverse developmental outcomes if ignored.

Hematological Differences among Malaria Patients in Rural and Urban Ghana

BACKGROUND

Scarce studies have addressed hematological differences of malaria in urban and rural regions.

METHODS

Full or complete blood cell counts from 46 and 75 individuals (age range from < 1 to 92 years) with uncomplicated malaria infection living in urban (Accra) and rural (Dodowa) Ghana, respectively, were assessed. Sickle cell trait and patients were excluded from the study.

RESULTS

Between overall groups, patients from Accra had significantly lower parasite count (p < 0.0001) and granulocyte number (p = 0.026). Children in Accra had a significantly lower parasitemia (p = 0.0013), hemoglobin (p = 0.0254), platelet count (p = 0.0148) and red blood cell levels (p = 0.0080) when compared with the children of Dodowa. In adults, mean cell hemoglobin (p = 0.0086) and parasite count (p < 0.0001) were significantly higher in Dodowa.

CONCLUSION

These results indicate that children living in urban setting may experience a greater anemic effect to malaria as compared with those living in a rural setting.