Modelling of Inhalation Exposures to Pharmaceutical Agents
Mc Donnell, Patricia
MetadataShow full item record
This item's downloads: 1360 (view details)
Assessment of worker's exposure is becoming increasingly critical in the pharmaceutical industry as drugs of higher potency are being manufactured. Also the batch nature of operations often makes it difficult to obtain sufficient numbers of exposure measurements to adequately characterise exposure levels. This research aims to investigate the ability of two occupational exposure models to predict inhalable dust exposures in this industry and specifically to: refine and validate an existing deterministic source-receptor exposure model; and to refine, calibrate, validate and investigate the reliability of the inhalable dust exposure form of the newly developed Advanced REACH Tool (ART). A broad range of workplace exposure assessment data containing all the contextual information required for the exposure models was collated from a multinational pharmaceutical company. Within both exposure models, input parameters such as localised controls and handling activities, were refined to reflect pharmaceutical situations. The source-receptor model was validated and the relationship between model estimates and measured pharmaceutical data was investigated (n=381 measurements). The inhalable dust exposure form of the ART was calibrated with a pharmaceutical dataset and linear mixed effects regression analysis was used to translate the relative model scores to quantitative exposure levels (in mg/m3). The pharmaceutical dataset was also included in the generic dataset for the calibration of the inhalable dust model and results of both calibrations are compared to investigate the applicability of the generic model for pharmaceutical scenarios. As part of the validation of the generically calibrated ART, relative bias and uncertainty around geometric mean exposure estimates were calculated for 16 pharmaceutical exposure scenarios (n=192 measurements). To investigate the reliability of the online ART, 18 health and safety professionals assessed four exposure scenarios representative of the industry; information and a demonstration of the ART were provided at two stages during the one-day workshop. Inter-rater agreement was investigated and also the participants¿ assessment per determinant and their ART exposure estimates were compared with the corresponding gold-standard assessments. The refined source-receptor exposure model resulted in good correlations between the log-transformed model predictions and the actual measurement data at scenario level (rs=0.69, n=48, p<0.001). The model overestimated scenarios ix with measured exposure levels <0.1 mg m3 (rs=0.69, bias=0.71, n=46, p < 0.001), and underestimated scenarios with higher measured concentrations (>0.1 mg m3) (rs=0.59, bias= -4.9, n=33, p < 0.001). Including information on the refined sub-parameters improved the correlations. The pharmaceutical and generically calibrated ART were able to estimate geometric mean exposure levels (with 90% confidence) for a given workplace exposure scenario within a factor of 4.6 and 4.4 respectively of the measured geometric mean exposure level. The calibrations resulted in comparable models which were able to explain similar levels of total exposure variance (69 and 64% respectively). The validation of the generic model showed that for 12 of the 16 scenarios, the ART geometric mean exposure estimates were lower than measured exposure levels, with on average a one-third underestimation of exposure (relative bias=-32%). For 75% of the scenarios the exposure estimates were, within the 90% uncertainty factor of 4.4. Results from the reliability study showed that the average not-chance corrected agreement values of the determinants with gold standard for the increased information stages were 58, 67, and 83% respectively. While the provision of information on ART improved inter-rater agreement and agreement with the gold-standard for most determinants, there was a broad range between the participants and gold-standard exposure estimates, with approximately 60% within ten-fold of the gold-standard. The source-receptor model provides a useful basis for an exposure assessment tool for the pharmaceutical industry; however the ART will likely have more useful applications for the this industry where it is anticipated that it can be used as part of the exposure assessment strategy under the Chemical Agents Directive (98/24/EC) or to assist with developing risk evaluations within the scope of REACH. The ART is not recommended as a suitable tool to assess a quantitative exposure level at a specific workplace. It is an expert tool and extensive user training in required. As models are increasingly used in the context of REACH and beyond, this study emphasises that proper validation and reliability studies are required.
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. Please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.
The following license files are associated with this item: