Response to “A Probabilistic Arsenic Exposure Assessment for Children Who Contact CCA-Treated Playsets and Decks, Part 1: Model Methodology, Variability Results, and Model Evaluation”

As stated in the report from the EPA SAP from the December 3–5, 2003 meeting regarding the SHEDS-Wood CCA assessment: “It was the consensus of the Panel that, by and large, the best information on input variables at this time has been used” (SAP, 2004). The Exponent comments on Zartarian et al. (2006) suggest that the values used in the SHEDS-Wood model for maximum dermal loading should be replaced with observational children's hand loading data from the Kwon et al. (2004), Wang et al. (2005), and Shalat et al. (2006) studies. While the 2001 and 2004 SAPs recommended that empirical data be collected for children actually playing on treated CCA-treated structures, the recommended studies have some major limitations, which are summarized below.

Kwon et al. (2004) and Wang et al. (2005)—Study Limitations

In addition, the objective of the Kwon et al. (2004) study and Wang et al. (2005) was to determine the quantitative amount of arsenic on the hands of children in contact with CCA-treated wood structures and soil in playgrounds; the studies were not designed to measure maximum dermal loading and the reported data are likely to lead to underestimates of inputs for that variable. Furthermore, there are 66 data points in the Kwon et al. (2004) study; this is relatively small when compared with more than 700 data points used in the SHEDS-Wood model. The values used in the SHEDS-Wood CCA assessment for transfer efficiency and maximum dermal loading were measured under well-controlled experimental conditions designed to collect data for those variables.

Shalat et al . (2006)—Study Limitations

Of key importance is the point that the information on children's hand loadings and on duration of children's play time in these studies are insufficient to directly inform the estimate of maximum dermal hand loading or related SHEDS model inputs; time spent on playgrounds is not equivalent to hand contact with wood. Without knowing actual dermal contact time with the CCA-treated wood, one cannot know whether the maximum dermal loading was reached. Because of this and the other limitations listed above, we do not believe that the data from these studies would necessarily improve the accuracy of the SHEDS-Wood CCA exposure assessment presented in Zartarian et al. (2006) and Xue et al. (2006). As indicated in the article, the means of the SHEDS-Wood distributions are very similar to the values used in most of the other models, and the distributions capture the other values in many cases. The comparison presented in the paper indicates that SHEDS-Wood estimates are consistent with, or in the range of, other CCA models. As stated in the SAP (2004) report: “The general consensus of the Panel was that the current SHEDS-Wood model implementation represented a good faith effort on the part of the Agency. Even though one can question specific choices of distributional assumptions, overall the work seemed a reasonable effort and a sound basis for risk assessment within the limitations of available information.”