Annals of Occupational Hygiene Advance Access originally published online on July 21, 2009
Annals of Occupational Hygiene 2009 53(7):677-689; doi:10.1093/annhyg/mep046
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Quantification and Statistical Modeling—Part I: Breathing-Zone Concentrations of Monomeric and Polymeric 1,6-Hexamethylene Diisocyanate
1 Department of Environmental Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7431, USA
2 Department of Biostatistics and Carolina Population Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7420, USA
3 Safety & Health Assessment and Research for Prevention Program, Washington State Department of Labor and Industries, Olympia, WA 98504-4330, USA
* Author to whom correspondence should be addressed. Tel: +1-919-966-3826; fax: +1-919-966-7911; e-mail: leena_french{at}unc.edu
We conducted a repeated exposure-assessment survey for task-based breathing-zone concentrations (BZCs) of monomeric and polymeric 1,6-hexamethylene diisocyanate (HDI) during spray painting on 47 automotive spray painters from North Carolina and Washington State. We report here the use of linear mixed modeling to identify the primary determinants of the measured BZCs. Both one-stage (N = 98 paint tasks) and two-stage (N = 198 paint tasks) filter sampling was used to measure concentrations of HDI, uretidone, biuret, and isocyanurate. The geometric mean (GM) level of isocyanurate (1410 µg m–3) was higher than all other analytes (i.e. GM < 7.85 µg m–3). The mixed models were unique to each analyte and included factors such as analyte-specific paint concentration, airflow in the paint booth, and sampler type. The effect of sampler type was corroborated by side-by-side one- and two-stage personal air sampling (N = 16 paint tasks). According to paired t-tests, significantly higher concentrations of HDI (P = 0.0363) and isocyanurate (P = 0.0035) were measured using one-stage samplers. Marginal R2 statistics were calculated for each model; significant fixed effects were able to describe 25, 52, 54, and 20% of the variability in BZCs of HDI, uretidone, biuret, and isocyanurate, respectively. Mixed models developed in this study characterize the processes governing individual polyisocyanate BZCs. In addition, the mixed models identify ways to reduce polyisocyanate BZCs and, hence, protect painters from potential adverse health effects.
Keywords: air sampling • exposure determinants • hexamethylene diisocyanate • isocyanate • statistical modeling
Received April 11, 2008; in final form April 26, 2009