Citation:

Xue, J., V. Zartarian, R. Tornero-Velez, L. Stanek, A. Poulakos, A. Walts, K. Triantafillou, M. Suero, AND N. Grokhowsky. A Generalizable Evaluated Approach, Applying Advanced Geospatial Statistical Methods, to Identify High Lead Exposure Locations at Census Tract Scale: Michigan Case Study. ENVIRONMENTAL HEALTH PERSPECTIVES. National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, NC, 130(7):1-14, (2022). https://doi.org/10.1289/EHP9705

Impact/Purpose:

Lead mapping to inform targeting of risk reduction efforts is a priority research area in EPA's Lead Coordination Plan and the Federal Lead Action Plan.¿ Through a collaborative case study with Region 5 and Michigan Department of Health and Human Services, ORD developed and applied a generalizable approach utilizing geospatial statistical methods and models to (1) analyze and map available children’s blood lead levels (BLL) data (obtained from the state via data use agreement); and (2) explore contributions to elevated BLL “hot spots” from old housing and sociodemographic variables (using convergence of EPA/OEJ’s EJSCREEN Pb Index; ORD’s Schultz et al. 2017 BLL model; and HUD’s Deteriorated Paint Index). This research supports multiple Agency drivers for lead, including regulatory efforts, actions on environmental justice, and building and maintaining partnerships with states. The analysis presented in this paper was conducted through a technical assistance request from Region 5 to ORD, to inform joint planning discussions on Pb between Region 5 and their state Pb/EJ partners. Results include data, maps, and analyses at the census tract scale. This paper provides methods and results to more effectively target and prioritize Pb risk reduction, prevention and mitigation efforts in the most vulnerable locations/populations. This approach could be applied to other states to inform lead mitigation and prevention efforts.

Description:

Background: Despite great progress in reducing environmental lead levels, many children in the U.S. are still being exposed, and there is no safe level. Objective: Develop a generalizable approach for systematically identifying, verifying, and analyzing locations with high prevalence of children’s elevated blood lead levels (EBLLs) and assess available lead models/indices as surrogates, with a Michigan (MI) case study. Methods: We obtained and geocoded 1.8 million BLL test results of children <6 years in MI from 2006-2016, then evaluated them for data representativeness by comparing two % EBLL rates (# children tested with EBLL divided by both # children tested and total population). We analyzed % EBLLs across census tracts over three time periods and between two reference levels (≥5 vs. ≥10 µg/dL) to evaluate consistency. Locations with high % EBLLs were identified by top 20 percentiles and Getis-Ord Gi* geospatial cluster “hot spot” analysis. For the locations identified, we analyzed convergences with three available lead exposure models/indices based on old housing and sociodemographics. Results: Analyses of 2014-2016 % EBLL data identified 11 MI locations via cluster analysis and 80 additional locations via top 20 percentiles, and associated census tracts. Data representativeness and consistency were supported by a 0.93 correlation coefficient between the two EBLL rates over 11 years, and Kappa score ~0.8 of % EBLL hot spots across the time periods and reference levels (2014-2016). Many EBLL hot spot locations converge with lead exposure models/indices; others suggest additional sources. In many census tracts, % EBLLs have reduced over time, especially since 2011. Discussion: Some identified locations were confirmed by previous efforts; we identified additional ones at finer geographic resolution using advanced geospatial statistical methods in conjunction with mapping/visualization, and assessed utility of surrogates in the absence of BLL data. This approach could be applied to other states to inform lead mitigation and prevention. More analyses and data are needed to prioritize specific environmental drivers of high % EBLL locations for protecting public health.

URLs/Downloads:

Record Details: