Potential Exposure of Hands Inside Protective Gloves a Summary of Data from Non-Agricultural Pesticide Surveys

Transcription

1 PII: S (00) Ann. occup. Hyg., Vol. 45, No. 1, pp , 2001 Crown Copyright 2001 Published by Elsevier Science Ltd on behalf of BOHS. All rights reserved Printed in Great Britain /01/$20.00 Potential Exposure of Hands Inside Protective Gloves a Summary of Data from Non-Agricultural Pesticide Surveys A. N. I. GARROD*, A. M. PHILLIPS and J. A. PEMBERTON Health and Safety Executive, Magdalen House, Stanley Precinct, Bootle L20 3QZ, UK The results of individual surveys are compared and general observations are made about the protective nature of protective gloves. A total of 190 measurements of contamination by involatile components of pesticide products were made inside the gloves. Protective gloves were nearly always found to be contaminated inside, except for public hygiene insecticide uses. Disposable gloves were sometimes used when applying public hygiene insecticides, while gloves worn for other applications were normally re-used. The data suggest that when expressing inner glove exposure as amount of product per unit time, the data distributions of the non-zero results appeared similar for all activities except for the application of antifoulings. A median value for all non-zero data was indicated at 1.36 mg product per minute; a 75th percentile at 4.21 mg min 1 and a 95th percentile at 71.9 mg min 1, assuming product densities of 1.0 g ml 1. In default of better information, these indicators can be used in healthbased risk assessment. A further 47 data related to exposure inside new gloves, and 21 data to deposition on outer gloves. These indicate that in general, gloves provide a reasonable degree of protection to non-agricultural pesticides. Due regard should be paid to human factors when considering contamination inside gloves. When people take off dirty gloves, it is difficult to avoid touching the glove exterior. With gloves off, the subject can touch contaminated surfaces. Donning dirty gloves furnishes an inevitable contact with contamination on the outside of the gloves, and can introduce that contamination inside. A common mode of contamination could contribute to the similar distributions for hand contamination inside gloves, from very different uses. Crown Copyright 2001 Published by Elsevier Science Ltd on behalf of BOHS. All rights reserved. Keywords: hand exposure; protective gloves; pesticides; non-agricultural; dermal exposure INTRODUCTION Since 1992, the Health and Safety Executive has conducted research into exposure to non-agricultural pesticide products: public hygiene insecticides (Llewellyn et al., 1996), remedial products timber preservatives and masonry biocides (Garrod et al., 1998), timber pre-treatment preservatives (Garrod et al., 1999) and antifoulant products (unpublished). For public hygiene insecticides and remedial products, the mixing and loading element of the work was brief and was not sampled separately. For timber pre-treatment preservatives there was no mixing and loading. For Received 14 July 1999; in final form 9 February *Author to whom correspondence should be addressed. Tel.: ; fax: antifoulant products, product loading was a separate task. In each of these studies, hand exposure was estimated by the users wearing cotton gloves inside their protective gloves. This paper compares the results of the individual surveys and makes general observations about the protective nature of gloves. The tasks undertaken included diluting product concentrates and loading applicator equipment; spraying narrow or wide areas, indoors and outdoors; using solvent- or water-based products and at high or low pressures; applying pesticide dusts; and handling treated articles which remain wetted with water- or solvent-based products. Products contained several pesticide active substances, both inorganic and organic, in a range of concentrations, and exposure ranged from a few minutes to several hours. The protective gloves in use varied from disposable latex (for some public hygiene insecticide surveys) to well-used 55

2 56 A. N. I. Garrod et al. heavy duty gauntlets (generally in a reasonable condition). RESULTS Pesticide active substances such as permethrin, bendiocarb, copper, arsenic and 2-phenyl phenol were measured on sampling gloves inside protective gloves. The resulting exposure data were expressed in terms of the amount of product in a standardized form, since in-use concentrations of the active substance varied between 20 mg l. 1 and 250 g l. 1, and the durations of daily use between 10 and 400 min. The standardized form of expression chosen was amount of in-use product per minute, to normalize for the concentration of active substance and the job duration. Also, the product density was assumed as 1.0 g ml 1. The findings suggest that when the exposure of hands inside protective gloves is expressed in a standardized form, there appeared to be little difference between the distributions for the types of use. Figures 1 4 present the data for the four main non-agricultural pesticide product types in terms of mg product per minute, as histograms on a logarithmic scale (log base 10). Each data set was too sparse to be able to draw firm conclusions about whether or not the data follow a log-normal distribution. Statistical tests (chisquared, Shapiro Wilk) suggest that log-normality Fig. 1. Insecticide product in gloves. Fig. 2. Remedial product in gloves.

3 Protective gloves and non-agricultural pesticide surveys 57 Fig. 3. Antifoulant product in gloves. Fig. 4. Wood pre-treatment preservative in gloves (light = new gloves). cannot be rejected for the insecticide, antifoulant and pre-treatment (existing gloves) data sets and can be rejected for the combined data sets (p 0.01). Figure 5 compares the data sets. Comparison by eye suggests that there are two main groups: that including insecticide, remedial and pre-treatment applications, and that for antifouling applications (and pre-treatment using new gloves). One-way analysis of variance (between scenarios versus within scenarios) showed statistically significant differences (p 0.01), but comparisons of each pair of scenarios showed that many were not significantly different. The antifoulants data set is significantly different from the insecticide and remedial data sets, and the remedial data set is significantly different from the pre-treatment (new gloves) data set. There is not a significant difference between pre-treatment with existing gloves and with new gloves. However, all of these statements need to be read with caution as the data sets are small. The full data set histogram for existing gloves appears in Fig. 6, with a log-normal distribution curve for comparison. Statistical tests indicated that the full data set distribution did not match log-normality. Figure 7 compares the distributions on a logarithmic scale (log base 10), for solvent-based products and for all other products (water-based, dusts, etc.). Comparison by eye shows little clear difference between these distributions. Although not presented in this paper, histogram distributions for the quantities of active substances

4 58 A. N. I. Garrod et al. Fig. 5. Confidence limits of exposure inside gloves comparison of different data sets. Fig. 6. Distribution of exposure inside gloves (all data sets). detected on sampling gloves showed no pattern and were distributed over more than five orders of magnitude. While there are too few data for reliable statistical analysis, Table 1 presents some values drawn from the data sets. These are compared with data relating to deposition on the exterior of gloves, acquired in studies by the Institute of Occupational Medicine on behalf of the Health and Safety Executive (Tannahill et al., 1996). The survey of timber pre-treatment indicated the differences in exposure when new, clean gloves were worn for a second cycle of timber treatment. The IOM data related to deposition on outer gloves. When compared with the in-glove exposures, in general, gloves provide a reasonable degree of protection to non-agricultural pesticides. DISCUSSION The four use sectors that have been studied comprise very different tasks. Insecticide application for public hygiene purposes involved a few litres or tens of grams; where diluted, this is usually with water, and sprayed or dusted at pressures of around 1 3 bar. Remedial treatment of masonry or timber involved tens to one hundred litres, diluted with water or supplied as a solution in solvent, and sprayed at pressures of around 4 8 bar. Antifouling application involved using hundreds of litres of viscous solvent-based product, either loading the sprayer reservoir, or airless spraying at 100 bar or more. The industrial pre-treatment of timber involved the intermittent handling of solvent-damp or water-wet articles, and contact with industrial plant and equipment.

5 Protective gloves and non-agricultural pesticide surveys 59 Fig. 7. Product in gloves (light = solvent based, dark = others). Table 1. Exposure inside gloves: all units as mg min 1 in-use product Data set Insecticide Remedial Antifoulant Pre-treatment Existing gloves New gloves No. of non-zero data Frequency a 38% 94% 93% 97% 95% Range Median th percentile value th percentile value Arithmetic mean Comparison: deposition on glove exterior (IOM data) No. of non-zero data Range Median a Frequency indicates the likelihood of exposure occurring: this accounts for none detected or below limit of quantitation results. It might be expected that several distinct sets of data would be obtained: this paper suggests that the expectation is erroneous. In public hygiene insecticide uses, disposable latex gloves were sometimes used and exposure inside the gloves was found in only 38% of samples, compared with % of samples for the other types of product. The 190 non-zero results indicated that most exposure distributions inside gloves appeared to coincide, for quite different activities and natures of product used, when exposure is expressed as a rate of contamination. A median value for the entire data set was indicated at 1.36 mg product per minute; a 75th percentile at 4.21 mg min 1 and a 95th percentile at 71.9 mg min 1. Where no better data are available, these values are proposed for use in risk assessment. Comparison of the median values for deposition on the glove exterior and interior suggests that better than 20-fold protection is offered by wearing protective gloves, type unspecified. Hitherto, research has concentrated on the permeability of glove fabrics, in respect of breakthrough times and rates of transfer through glove material: for solvents that is a reasonable concern. However, the data in this paper suggest that the in-glove distributions for most types of product are similar for water-based and solvent-based products in respect of time-weighted in-glove exposure, based on involatile substances. Regression analysis of the full set of data indicates that there is a poor correlation between the amount of product on the sampling glove and the time that the sampling glove was worn (correlation with a line of best fit, 0.03). This suggests that irrespective of glove material, the potential for exposure of hands

6 60 A. N. I. Garrod et al. inside gloves to involatile substances has neither the duration of work nor the type of product as important factors. The in-glove exposure distribution is significantly different for antifouling applications. A reason for this may be that applying antifouling requires workers to follow stringent personal protective procedures, due to the irritant nature of many of these products. Glove removal during a spraying session is uncommon. Surveys have shown that workers take care to avoid antifoulant products contacting their skin. In general and for longer spells of work, it is proposed that workers are more likely to remove and put on again their protective gloves, and transfer contamination inside. Wearing gloves already contaminated inside will tend to give exposures towards the top end of the range, while regularly replacing gloves will reduce the potential for such exposure. Wearing gloves just once and then disposing of them reduces the likelihood of exposure, and of top-end exposure. The human factor is common to all glove uses. Simple observation of workers putting on and taking off protective gloves indicates the potential for hand exposure. This supports an obvious conclusion: that controlling exposure to workers hands inside their protective gloves relies on training to recognize how substances get inside gloves, and on a programme for regular glove replacement. Acknowledgements The authors thank Ms Suzi Curtis and Mr Alan Spence of HSE, who conducted the statistical tests and analyses. REFERENCES Garrod, A. N. I., Rimmer, D. A., Robertshaw, L. and Jones, T. (1998) Occupational exposure through spraying remedial pesticides. Annals of Occupational Hygiene 42(3), Garrod, A. N. I., Martinez, M., Pearson, J., Proud, A. and Rimmer, D. A. (1999) Exposure to preservatives used in the industrial pre-treatment of timber. Annals of Occupational Hygiene 43(8), Llewellyn, D. M., Brazier, A., Brown, R., Cocker, J., Evans, M. L., Hampton, J., Nutley, B. P. and White, J. (1996) Occupational exposure to permethrin during its use as a public hygiene insecticide. Annals of Occupational Hygiene 40(5), Tannahill, S. N., Robertson, A., Cherrie, B., Donnan, P., Mac- Connell, W. L. A. and Macleod, G. J. (1996) A comparison of two different methods for assessment of dermal exposure to non-agricultural pesticides in three sectors. IOM report TM 96/107, Institute of Occupational Medicine, Roxburgh Place, Edinburgh.