Assessment of occupational exposure to chemicals by air sampling for comparison with limit values: the influence of sampling strategy.

This study was aimed at quantifying the impact of sampling duration and the number of measurements taken on the quality of assessing occupational exposure to toluene. To this end, a measurement database was built, based on four campaigns carried out in an industrial printing facility. Five homogeneous exposure groups (HEGs) were set up and between 120 and 290 individual measurements lasting from 2 to 8 h were collected for each of them. These measurements were performed with the objective of comparing them to the 8-h Occupational Exposure Limit (8-h OEL). The resulting data were used to define a reference exposure profile per HEG: the 'gold standard'. This exposure profile corresponds to a log-normal distribution of measurements from which compliance/non-compliance with the 8-h OEL decision is derived. To simulate the possible sampling strategies used by industrial hygienists, six scenarios were defined, each containing a different number of measurements: 2, 3, 4, 6, 9, and 12 measurements performed per HEG, over different working days and different seasons of the year. The measurement values per scenario were simulated by sampling from the real measurement per HEG. For each scenario, 1000 simulated exposure profiles and corresponding simulated compliance decisions were computed. They were compared to the gold standard compliance decision using statistical indicators. Three methods were used for computing the simulated compliance decision: (i) the 95th percentile must be lower than the 8-h OEL, (ii) the exceedance fraction with respect to the 8-h OEL must be <0.1% (as defined by standard CEN 689, Appendix D), and (iii) the 70% upper confidence limit of the exceedance fraction with respect to the 8-h OEL must be <5% (as defined by French regulations). The results show that exposure assessment quality increases with both the number of measurements and sampling duration when using the 95th percentile and exposure assessment based on French regulations, whereas it decreases when using the standard. Moreover, guidelines for the efficient evaluation of chemical exposure in the workplace can be drawn up to help professional occupational hygienists. Indeed, boundaries can be recommended regarding the number of measurements and sampling duration necessary to obtain a reliable exposure assessment while minimizing effort devoted to sampling and analysis.