Geographic Analysis of Blood Lead Levels and Neighborhood-Level Risk Factors Among Children Born in 2008–2010


Childhood lead exposure remains a public health concern, as it can lower a child’s intelligence quotient and cause permanent neurological damage. The objective of this study was to identify census tracts with the highest risk for blood lead levels (BLLs) ≥5 μg/dL in Philadelphia, Pennsylvania. We analyzed BLLs among children born in 2008–2010 who had at least one venous BLL test. A multivariable mixed effects logistic regression model was used to create risk scores and estimate adjusted odds ratios (OR) for BLLs ≥5 µg/dL and census tract-level characteristics, including median age of housing, percent of Black residents, vacant properties, rental properties, adults living in poverty, and adults with a high school education. Of 49,246 children, 14.37% had more than one BLL ≥5 µg/dL. Census tracts with a higher percentage of pre-1950 housing (OR = 1.82, 95% confidence interval [CI] [1.63, 2.04]), Black residents (OR = 3.22, 95% CI [2.70, 3.84]), vacant properties (OR = 1.45, 95% CI [1.23, 1.72]), and poverty (OR = 1.49, 95% CI [1.18, 1.89]) were associated with the highest risk. The mean risk score was 0.12 (range: 0.02–0.35). Our findings show that risk scores can help target prevention activities.


Speaker / Author: 
Elizabeth Taggert, MPH, Philadelphia Department of Public Health
Mary Figgatt, MPH, Philadelphia Department of Public Health
Lucy Robinson, PhD, Department of Epidemiology and Biostatistics, Dornsife School of Public Health, Drexel University
Raynard Washington, PhD, Philadelphia Department of Public Health
Natalie Kotkin, Philadelphia Department of Public Health
Caroline Johnson, MD, Philadelphia Department of Public Health
Month Year: 
October 2019
Page #: 
Publication Month: 
October 2019