Annals of Occupational Hygiene Advance Access originally published online on September 18, 2007
Annals of Occupational Hygiene 2007 51(7):646-647; doi:10.1093/annhyg/mem043
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTER TO THE EDITOR |
Reply
We thank Mr Walker for his interest in our paper (Creely et al., 2006) and are disappointed to learn that it failed to live up to his expectations!Unfortunately, data collected for one purpose often fail to enlighten other people with different aims and objectives. While we sympathize with the problems in handling litigation cases, the purpose of our study was neither to find improvements to the MDHS 25 (HSE, 1999) air sampling methodology nor to make barristers' lives any easier (or more difficult). Rather, the context of our work is within the general occupational hygiene approach to find ways of preventing occupational ill health.
As stated in our paper, the purpose of our study was to determine the exposure levels that are achievable using recognized good working practices and exposure control methods. We concluded that biological monitoring methods could be used more effectively to control workplace exposure than by using air monitoring methods. In fact, we indicated the shortcomings of the air sampling method in our paper and referred to other studies where this had also been reported. We used the best air-sampling method available to us but are unable to respond to the particular criticisms and uncertainties in the MDHS 25 analytical methodology. These points are covered in an accompanying reply by John White.
Mr Walker stresses in numbered paragraph 1 of his letter that air-sampling results are a critical issue in determining negligence and liability of an employer. This is elaborated in paragraph 3 where the issue of non-detectable samples is covered. Firstly, in our opinion, the points raised simply illustrate the limitations of the adversarial approach used in the British courts rather than being a valid criticism of our study. Secondly, it is a tenet of occupational hygiene practice that air monitoring should be used as a backup to any workplace exposure control programme for the very reasons implied in the letter. Employers would be well advised not to rely on air sampling measurements as an excuse for failing to implement primary control measures. In this context they should also avoid relying on air-sampling measurements as a defence in court or they will no doubt face a grilling along the lines described.
With regard to the technical queries relating to the limits of detection and quantification (LOD and LOQ) respectively, it cannot be inferred that by setting an LOQ at a certain level that airborne isocyanates were definitely present at lower levels and we do not imply this by setting a nominal value of 
LOQ for levels that we cannot quantify. As Mr Walker has rightly stated, all that can be concluded is that the method could not reliably detect anything below this level. It would therefore be wrong to use a zero level for non-detectable samples. Since our air samples were collected over a range of different time periods, it does not make much sense to specify the LOD for each sample in terms of an air concentration. Air sample concentrations are not accurate down to the analytical LOD, and it would be misleading to report them as if they were. While we found that we could achieve an acceptable level of detection (0.02 µg) for the analytical method, there was considerable random variability found in some field blank samples, which were used as quality control samples. Unfortunately, this seems to be a feature of the treated filter sample method for which we have no explanation to offer. This meant that it was possible to detect isocyanates in field samples, but when the result of field blanks were taken into account, the corrected results could not be reliably quantified at the analytical LOD quoted. Hence, this was one further reason why we set the LOQ at 0.001 mg m–3.
Using a level of LOQ for statistical analysis is common practice in exposure assessment for the simple reason that it is not possible to use a value of zero when performing statistical analyses for data that are log normally distributed. This is a common problem when dealing with data sets, where the majority of measurements are below the LOD, but we do not believe that our conclusions are in any way invalidated by this finding. Further explanation of these principles has been discussed by Finkelstein and Verma (2001) and Glass and Gray (2001) among others.
Paragraph 4 criticized our sampling strategy, but we feel this is unfounded for the following reasons. Firstly, an impinger is only necessary when isocyanate aerosol is likely to be present. The great majority of samples collected were in the polyurethane foam industry where there was no such aerosol. For the tasks selected for monitoring, we mainly collected samples using treated filters, although some impinger/filter combination samples were collected as a check. We observed no difference between the filter-only samples and the check samples. The same approach was adopted for spray painting and polyurethane spray application but there were too few samples to make a meaningful comparison. However, we do make the point that these particular workers were all using supplied-air breathing apparatus, so it is not possible to correlate the air samples with their biological monitoring data. To do so would be illogical and misleading. With regard to dermal exposure, we do make the point several times that this may lead to systemic absorption and be indicated by biological monitoring.
We hope that this has gone some way to answering Mr Walker's criticisms. By way of summary, we would agree that the issue of isocyanate exposure assessment is a difficult matter, which poses some challenges for occupational hygiene professionals. We look forward to seeing more research carried out in this area which will hopefully lead to improved exposure assessment methodologies and control options.
1 Institute of Occupational Medicine, Research Avenue North, Riccarton, Edinburgh EH14 4AP, UK
2 Health & Safety Laboratory, Harpur Hill, Buxton, Derbyshire SK17 9JN, UK
* E-mail: graeme.hughson{at}abdn.ac.uk
Received June 20, 2007;
REFERENCES
Creely KS, Hughson GW, Cocker J, et al. Assessing isocyanate exposures in polyurethane industry sectors using biological and air monitoring methods. Ann Occup Hyg (2006) 50:609–21.
Finkelstein MM, Verma DK. Exposure estimation in the presence of nondetectable values: another look. Am Ind Hyg Assoc J (2001) 62:195–8.
Glass DC, Gray CN. Estimating mean exposures from censored data: exposure to benzene in the Australian petroleum industry. Ann Occup Hyg (2001) 45:275–82.
HSE. MDHS 25/3. Organic isocyanates in air (1999) Sudbury, UK: Health and Safety Executive Books.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||