Dr. Jacqueline MacDonald Gibson seeks proactive, preventative approach to identifying home lead hazards
The U.S. Department of Housing and Urban Development (HUD) recently indicated it will fund a new study led by Dr. Jacqueline MacDonald Gibson of Indiana University’s School of Public Health-Bloomington. MacDonald Gibson, who chairs the Department of Environmental and Occupational Health, will serve as principal investigator for “Targeting Homes with High Lead Exposure Risks by Leveraging Big Data and Advanced Machine-Learning Algorithms.”
The HUD 2020 Lead Technical Studies grant of $450,000 will enable MacDonald Gibson and her colleagues to develop new and more cost-effective methods for identifying residences with children at high risk for lead exposure.
MacDonald Gibson’s co-principal investigators include IU alumnus Dr. Emmanuel Obeng-Gyasi of North Carolina Agricultural and Technical State University and Drs. James Harrington and Jennifer Hoponick Redmon of RTI International in Research Triangle Park, North Carolina. Partner organizations also involved with the study include Greensboro, North Carolina; Fort Wayne, Indiana; and the State of Indiana's Housing and Community Development Authority.
While federal policies enacted in the 1970s to mitigate the public's exposure to lead have helped, lead still can be found in old paint, plumbing pipes and fixtures, consumer goods, and more. Overall, blood lead levels in U.S. residents have decreased; however, three percent of children still have elevated levels of lead in their blood. What’s more, low-income and Black children are at higher risk of exposure. Studies over the past several decades have linked children’s exposure to lead to permanent cognitive damage, even at low levels of exposure.
Presently, many state and local agencies use a home's age, the results of children’s blood lead level tests, or both to determine which homes should be targeted for lead hazard intervention. But relying on a home’s age alone may be overly limiting and relying on children to serve as “canaries in the coal mine” unnecessarily subjects them to potential cognitive harm.
The researchers plan to use machine-learning techniques to leverage existing data on potential residential lead exposure risk factors in order to create a Web-based mapping application. The app—to be called “Pb Reduction In Critical Homes” (PREDICT)—will aim to precisely match potential lead interventions to specific at-risk households. The study’s various partner organizations will test the PREDICT app.