Tijuana childhood lead risk assessment revisited: validating a GIS model with environmental data

The objective of this research was to determine the spatial distributions of childhood lead poisoning and soil lead contamination in urban Tijuana. The Bocco-Sanchez model of point-source emissions was evaluated in terms of validity and reliability. We compared the model's predicted vulnerable populations with observed cases of childhood lead poisoning in Tijuana, identified fixed point sources in the field, and analyzed 76 soil samples from 14 sites. The soil lead results were compared to the blood lead analyses performed on Tijuana children whose blood lead levels were > or =10 microg/dL, who reported that they did not use lead-glazed ceramics for cooking or storing food (n = 63). Using GIS, predicted vs observed risk areas were assessed by examining spatial patterns, including the distribution of cases per designated risk area. Chi-square analysis of expected vs observed values did not differ significantly at the p = 0.02 level, showing that the model was strikingly accurate in predicting the distribution of subjects with elevated blood lead. Results reveal that while point sources are significant, other sources of lead exposure are also important. The relative public health risk from exposure to lead in an urban setting may be assessed by distinguishing among sources of exposure and associating concentrations to blood lead levels. The results represent an iterative approach in environmental health research by linking environmental and human biomarker lead concentrations and using these results to validate an environmental model of risk to lead exposure.