Neighborhood characteristics of poisoning

Background: Analysis of variations in poisoning prevalence and intensity by environmental/neighborhood characteristics is a useful tool for health policy and targeting of interventions. Methods: Records of calls in 2005 to the designated poison center for North Carolina (N=83,303) were extracted. Those with valid ZIP Codes (N=48,740) were merged with ZIP Code-level demographic characteristics from the Claritas Demographic database (Claritas Inc). Calls for poisonings due to household products (N=4,546), prescription medications (N=8,772), and herbicides/pesticides (N=1,889) were analyzed. Data for poisonings per unit population and demographic characteristics potentially of interest were mapped by ZIP Code using ArcGIS ArcInfo 9.2 (ESRI Inc.). Results: Overall poisoning rates showed regional concentrations, especially in urban and suburban areas. Household product poisonings were inversely associated with the proportion of the population under 15 (p=0.0037), and were not associated with the proportions of the population with less than high school education or with income below poverty. Prescription drug poisonings were related directly to the proportion of the population over 65 (p=0.0001) and inversely with the proportion under 15 (p=0.0021), but were not related to education or poverty. Herbicide/pesticide poisonings showed trends toward association with higher poverty and with service/farm employment. Intentionality was not addressed in this exploratory analysis. Numerical data presented here will be amplified with maps at presentation. Conclusions: These findings, combined with geographic data, suggest that analysis of poison center data can determine places and populations to target for poison prevention educational interventions, potentially making the best use of public health resources.