Background: To assess infertility knowledge and treatment beliefs among African American women in an urban community in Atlanta, Georgia. Methods: This was a cross sectional study at a safety net hospital. A convenience sample of a total of 158 women receiving outpatient obstetrical or gynecologic care from March–April 2017 were recruited. Infertility knowledge and treatment beliefs were assessed using a previously applied and field-tested survey from the International Fertility Decision Making Study. Results: The mean infertility knowledge score was 38.15% for total subjects. Those with a higher level of education (p < 0.0001) and those with paid employment (p = 0.01) had a significantly higher level of infertility knowledge. Those who had a history of infertility therapy were significantly more likely to agree with negative treatment beliefs (p = 0.01). There was no significant difference in infertility knowledge or treatment beliefs based on age, sexuality, parity or being pregnant at the time of survey completion. Conclusions: African American women in our urban clinic setting seem to have a limited level of knowledge pertaining to infertility. Further research is needed to understand how differences in knowledge and beliefs translate into infertility care decision-making and future childbearing.

The molecular mechanisms underlying chemoresistance in some newly diagnosed multiple myeloma (MM) patients receiving standard therapies (lenalidomide, bortezomib, and dexamethasone) are poorly understood. Identifying clinically relevant gene networks associated with death due to MM may uncover novel mechanisms, drug targets, and prognostic biomarkers to improve the treatment of the disease. This study used data from the MMRF CoMMpass RNA-seq dataset (N = 270) for weighted gene co-expression network analysis (WGCNA), which identified 21 modules of co-expressed genes. Genes differentially expressed in patients with poor outcomes were assessed using two independent sample t-tests (dead and alive MM patients). The clinical performance of biomarker candidates was evaluated using overall survival via a log-rank Kaplan–Meier and ROC test. Four distinct modules (M10, M13, M15, and M20) were significantly correlated with MM vital status and differentially expressed between the dead (poor outcomes) and the alive MM patients within two years. The biological functions of modules positively correlated with death (M10, M13, and M20) were G-protein coupled receptor protein, cell–cell adhesion, cell cycle regulation genes, and cellular membrane fusion genes. In contrast, a negatively correlated module to MM mortality (M15) was the regulation of B-cell activation and lymphocyte differentiation. MM biomarkers CTAG2, MAGEA6, CCND2, NEK2, and E2F2 were co-expressed in positively correlated modules to MM vital status, which was associated with MM’s lower overall survival.

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.

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.