Non-analytic problems in detecting arsenic and cadmium in children living near a cadmium refinery in Denver, Colorado.

The aim of the present study was to determine urinary arsenic (N = 322) and cadmium (N = 366) levels in children aged six months to six years who live near a working cadmium refinery and to compare their values with those of children from comparison neighborhoods. A questionnaire designed to identify exposure pathways was administered to the parents. There were unexpected problems in the study. Eighty-four percent of the arsenic samples were below the detection limit of 10 micrograms/l and summary statistics could not be calculated. Urinary arsenic and cadmium values could not be standardized for volume and concentration of urine because a large proportion of the samples had very low creatinine values. The original round of cadmium testing was afflicted with contamination problems, possibly due to the mishandling of pediatric urine bags by the parents during the collection procedure. A retest for cadmium levels under clinical conditions showed lower cadmium levels, all but two were below the detection limit. While biological monitoring of exposure to metals can be undertaken indirectly by measuring the concentration of the metals in urine, the analyses in this study were complicated by (1) not using more sensitive analytical tests for arsenic and cadmium determination, (2) not being able to standardize children's urinary values with creatinine, and (3) allowing in-home urine collection.