Exposure to chlorpyrifos and use of pesticides in Arizona

This study of pesticide use and exposure in the Arizona NHEXAS sample explored demographic variation in pesticide use and the feasibility of predicting the subjects' urinary TCPY (a chlorpyrifos metabolite) using questionnaire data. A variety of demographic differences in pesticide use patterns such as frequency of personal application, use of professional exterminators, and frequency of use outside the home were found. In contrast, there were few demographic differences in TCPY, although mean TCPY was higher than in earlier epidemiological studies. Two mulitvariate (multiple regression) methods of predicting TCPY from questionnaire responses were tested. The first method combined pesticide use questions into a scale; the second method treated them separately. R2s were similar and below .25 using either method. Including only subjects reporting some pesticide use raised the R2 to .35. These results suggest that questionnaire responses cannot predict exposure accurately enough to be useful proxies for biological samples.