Ann. occup. Hyg., Vol. 48, No. 1, pp. 21-27, 2004
© 2004 British Occupational Hygiene Society
Published by Oxford University Press
Contact Allergy in Construction Workers: Results of a Multifactorial Analysis
1 Department of Medical Informatics, Biometry and Epidemiology, Friedrich-Alexander-University Erlangen-Nürnberg, Waldstrasse 6, 91054 Erlangen; 2 Bau-Berufsgenossenschaft Frankfurt am Main, An der Festeburg 27–29, 60389 Frankfurt/Main; 3 Information Network of Departments of Dermatology (IVDK), Department of Dermatology, Georg-August University Göttingen, von-Siebold-Strasse 3, 37075 Göttingen, Germany
Received 21 March 2003; in final form 9 June 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objectives: To quantify the risk of contact allergy (CA) to important ubiquitous allergens associated with certain occupations, in particular the construction industry, and to identify possible time trends, controlling for potential confounding variables. Methods: Bivariate as well as Poisson regression analysis of standardized anamnestic and patch test data comprising 82 561 patients assessed in the 33 German and Austrian contact dermatitis units of the Information Network of Departments of Dermatology (IVDK) between 1992 and 2000. Results: CA to five of the 18 (groups of) allergens considered here were observed significantly more often in construction workers: dichromate, epoxy resin (BADGE), cobalt, thiurams and N-isopropyl-p-phenylenediamine (IPPD). Multifactorial analysis confirmed an increased risk of CA to these allergens in construction workers, compared with other occupations. A very strong association between cobalt and chromate allergy was found in construction workers (OR 39.1, 95% CI 21.1–79.6). Conclusions: Dichromate is still an prominent allergen in construction workers; as yet, there is only weak evidence of a decrease in Germany. Therefore, the addition of ferrous sulphate to cement, which has been a successful intervention in other countries, should be promoted further. Although CA to other important occupational allergens like thiurams, IPPD and epoxy resin or cobalt (very often associated with dichromate CA) is less frequent, prevention should address these allergens, too. The use of protective gloves with minimal intrinsic CA risk, e.g. due to thiurams in (synthetic) rubber or chromate in leather gloves, should be promoted.
Keywords: contact allergy; construction industry; chromate; cobalt; epoxy resin; clinical epidemiology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Contact dermatitis ranks high in occupational disease statistics in many countries. Its aetiology may be allergic or irritant, sometimes combined, and often aggravated by constitutional factors. Allergic contact sensitization is considered to last life-long, thus making life-long allergen avoidance necessary, which is often only possible by a change of occupation. Contact allergy (CA) is diagnosed by patch testing, i.e. by exposing the skin of the back to minute amounts of the potentially offending substance(s) and observing whether a small patch of allergic contact dermatitis develops which will indicate the causative agent(s) (Gawkrodger, 2001; Wahlberg, 2001). As a final and often most difficult step, the individual significance of each positive (allergic) patch test reaction has to be evaluated in the light of the patient’s previous exposures and the course and pattern of dermatitis, i.e. whether the positive test reaction is ‘clinically relevant’ or has been relevant previously (Bruze, 1990; Wahlberg, 2001).
As CA cannot be treated up to now, primary prevention is of utmost importance. However, to efficiently target prevention, risk factors for CA to a specific substance have to be identified as a first step; these may vary substantially between different allergens due to different (occupational) exposure profiles. Population-based studies are often preferable over patient-based studies by virtue of a potentially more unbiased estimation of risk. However, in the case of CA they have limited power to detect more than the most obvious associations because (i) sample sizes are relatively small in relation to the usually low prevalence of specific sensitization and (ii) the obvious necessity to at least adjust for age and sex to reduce confounding will inevitably lead to a situation of sparse data. Thus, only a large collection of sufficiently standardized clinical patch test data allows for detailed multifactorial analysis. While risk estimates obtained in such clinical data should be interpreted cautiously, because of unknown underlying selection processes, this apparent drawback is put into perspective by the observation that participation rates in population-based studies, especially if employing more invasive diagnostics such as patch testing, range far from 100% and thus also carry the risk of selection bias. As a collection of clinical CA data of unique size, the Information Network of Departments of Dermatology (IVDK) data pool was analysed to assess occupational risk factors for CA to an array of ubiquitous allergens (Geier et al., 2002). This paper presents results pertaining to construction workers, who have a notoriously high profile of occupational skin disease (Avnstorp, 2000), mainly due to exposure to wet cement (Rühl and Hadrich, 2001).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The IVDK (www.ivdk.org) is a multi-centre project for the epidemiological surveillance of CA (Uter et al., 1992, 1998). Briefly, results of all patch tests, which are performed according to national (Schnuch et al., 2001) and international guidelines (Wahlberg, 2001), and a standardized history of all patients patch tested in the participating centres are recorded. At regular intervals, this data is anonymously transferred to the data centre in Göttingen for pooled analysis. Thus, the IVDK surveillance is an instrument of clinical epidemiology, i.e. analyses are based on data of patients referred for the evaluation of suspected CA, as opposed to population-based epidemiological approaches. Data management and analysis is performed with the statistical program package SAS (version 8.2; SAS Institute, Cary, NC).
In the present analyses, CA to selected allergens of the so called ‘standard series’ is considered, focusing on those substances encountered in an occupational context. The ‘standard series’ is a set of allergens applied to almost all patients patch tested, because it contains the most important allergens on a general level. Results are based on test readings 3 days after application of patches. Allergens were obtained by Hermal, Reinbek, Germany. All patients patch tested in the participating centres of the IVDK between January 1992 and December 2000 were included in this analysis. If there were multiple consultations with one patient, the data from the consultation with the strongest reaction to the respective allergen was selected for the present analysis, in order to increase the frequency of a positive outcome and to avoid the statistical problems of dependent data arising when analysing the full data set.
Anamnestic data include the anatomical site of current dermatitis (aggregated to 14 categories in the present analysis), data on previous or current atopic dermatitis, age (categorized according to the quartiles of the overall age distribution for this analysis), sex, the time of the patch test (considered in four categories here) and the current occupation. Occupations are registered within the IVDK by using the code numbers of the Federal Statistical Bureau (Anonymous, 1992). To avoid problems of small sample sizes, single occupations were aggregated to 52 meaningful occupational groups with regard to a homogeneous occupational exposure profile, including one category for ‘missing data’ and one for ‘undetermined exposure’ (mainly old age pensioners and students), which is not considered in the tables, although included in the multifactorial analyses. The single job titles aggregated to ‘construction work’ are shown in Table 1 of the supplementary information included in the version of this paper in the on-line edition. If contact allergy to a certain substance was rare in the group of patients (<1.5% overall prevalence, as for IPPD and epoxy resin), the 52 occupational groups were further aggregated to 15 major occupational groups.
Bivariate analyses were performed utilizing Fisher’s exact test employing Bonferroni adjustment to account for multiple testing, trend tests analysing dichotomous outcomes with the Cochran–Armitage trend test (two-sided). To estimate the risk associated with each risk factor considered here, Poisson regression analysis was performed. The pre-specified model included the six variables in the degree of detail mentioned above. The adjusted prevalence ratios (PR) were derived from the parameter estimates of the Poisson model to quantify the strength of association, accompanied by 95% confidence intervals (CI) to address the precision of the estimates. The association between CA to dichromate and CA to cobalt, stratified for construction workers and non-construction workers, respectively, was quantified with odds ratios (ORs) derived from logistic regression analysis, with accompanying profile likelihood 95% CIs, controlling for age and sex as potential confounders.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the study period, 82 561 patients were patch tested in the 33 participating centres. Important demographic characteristics according to the MOAHLFA index (Schnuch et al., 1997) are given in Table 1; a very high proportion of men and a relatively high proportion of patients with occupational skin disease and with hand dermatitis, respectively, are evident among construction workers (n = 1238). Due to the fact that not every patient is tested with the standard series containing the set of allergens considered here and because there may be omissions of certain allergens individually, the actual number of patients exposed to the allergens varies (Table 2). In Table 2, crude proportions of patients reacting positively to the allergens listed are given. After adjusting for multiple testing, five allergens exhibited significantly higher proportions of CA responses in construction workers (P < 0.05, one-sided, adjusted Fisher’s exact test): dichromate (tested as potassium dichromate [CAS 7778-50-9]), epoxy resin (BADGE, diglycidylether of bisphenol A [CAS 1675-54-3], cobalt (tested as cobalt chloride [CAS 7646-79-9]), thiuram mix (tetramethylthiuram monosulphide [CAS 97-74-5], tetramethylthiuram disulphide [CAS 137-26-8], tetraethylthiuram disulphide [CAS 97-77-8] and dipentamethylenethiuram disulphide [CAS 94-37-1] in equal parts) and N-isopropyl-N'-phenyl-p-phenylenediamine (IPPD, CAS 101-72-4). The following multifactorial analyses address these allergens.
|
|
Bivariate, unadjusted analyses could not confirm a trend over time for epoxy resin (P = 0.77), thiurams (P = 0.60) or cobalt (P = 0.33) and only a weak, non-significant decrease for dichromate (P = 0.22) and IPPD (P = 0.06) in construction workers. The five Poisson regression models, with a positive test reaction to the allergens listed above as outcome, revealed several factors which were significantly associated with CA risk (online supplement, Tables S3a–S7b). An excerpt of results is given in Table 3 regarding non-occupational risk factors. Significant variation of risk with age or the timing of the patch test is evident, including a steady overall increase in sensitization prevalence over time for epoxy resin and a decrease for dichromate.
|
Results regarding occupation as risk factor are given in Table 3b of the on-line supplement: estimates for the ‘top five’ occupations and major occupational groups, respectively, and the respective reference group for each allergen are given. All factors were always included in the model, even if only part of the results is presented. Regarding dichromate, cobalt and epoxy resin, construction workers have the highest risk, whereas medical occupations and rubber manufacturers have an even higher risk for thiuram CA and farming and service occupations ‘not elsewhere classified’ (NEC) are associated with a higher risk for IPPD CA. It should be noted, however, that while the ranking of occupations according to CA risk estimates is definite, the actual estimates critically depend on the definition of the reference group, which is arbitrary. In our presentation, we have chosen occupational groups with a CA prevalence very close to the average as reference in each of the five models. The variation in risk associated with occupation differs between the allergens: the quotient of highest occupational PR/lowest occupational PR is 10.2 for dichromate, 11.0 for cobalt and 28.1 for thiurams. As expected, the quotients for IPPD (3.4) and epoxy resin (8.5) are lower, because the aggregation to major occupational groups necessary for these two allergens diminishes variation.
Of those 89 construction workers who had an allergic reaction to cobalt, 74 also reacted positively to chromate; of the 15 remaining patients, six were allergic to nickel. Conversely, 74 of 216 chromate allergic construction workers reacted positively to cobalt and 142 did not. Thus, in construction workers a very strong association between CA to chromate and CA to cobalt is found (OR 39.1, 95% CI 21.1–79.6). In contrast, the association in non-construction workers was not as strong (OR 11.8, 95% CI 10.8–12.9), in both cases controlled for sex and age.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
With the multifactorial analysis presented here, an elevated risk for specific CAs associated with current employment in the construction industry was identified in contact dermatitis patients patch tested over a period of 9 yr in a national CA surveillance network. The results add important evidence to the few available population-based studies on dermatitis risk in the construction industry and the even fewer attempts to estimate prevalence or incidence of specific CAs in this trade.
The overall incidence of ‘occupational skin disease’, which is almost equivalent to occupational contact dermatitis (OCD, irritant or allergic) has been estimated to be 19.0/10 000 workers/yr in tile setters and terrazzo workers and 5.4 in construction and cement workers in northern Bavaria (average of the 24 occupational groups 6.7) (Dickel et al., 2001). The German surveillance scheme provided by the workers’ compensation board identified 2702 cases of suspected OCD in the construction industry in 1999 (14.3% of all reported cases) (Plinske et al., 2001), which, related to 2 384 856 persons working in the construction industry in 1999 (Plinske et al., 2001), would correspond to a notification rate of 11 cases/10 000 working yr.
In the UK, the construction industry is not among the ‘top 10’ high incidence occupations for men for OCD, according to the surveillance schemes EPIDERM and OPRA (Cherry et al., 2000). An analysis of data of the US Supplementary Data System found workers in the construction industry (SIC 1972/77 numbers 15–17) in the third rank of industries regarding occupational skin disease, after agriculture and manufacturing, in accordance with 1981 annual BLS survey data (O'Malley et al., 1988). However, data from these registers are difficult to compare due to very different methodological approaches.
A Finnish cross-sectional study performed in 1987 found chromate CA, verified by patch test, in nine of 117 workers with hand dermatitis (total sample size 1620) (Roto et al., 1996). The prevalence of work-related hand dermatitis (allergic or irritant) was 6.8% among the construction workers and 8.9% among the concrete element prefabrication workers. A Polish cross-sectional study found dermatitis in 28.2% of 905 workers exposed to wet cement in the prefabrication of concrete building elements; 23.3% of these exposed workers were found to be allergic to dichromate and 12.9% to cobalt (Kiec-Swierczynska, 1990). In a southern Taiwanese cross-sectional study (telephone interview of 1147 construction workers and medical assessment of those reporting skin problems), 13.9% of men reported skin problems possibly related to work. Cases of OCD were regarded as under-diagnosed and inadequately managed (Guo et al., 1999). While this latter observation from Taiwan may not be directly transferable to Europe, it highlights the possibility of under-reporting in those surveillance schemes relying on notification.
Based on data in the Finnish Register of Occupational Diseases between 1991 and 1997, Kanerva et al. calculated an incidence of chromate CA of 6.94 (notified!) cases/10 000 working yr in cast concrete product workers, 3.44 cases in bricklayers, 2.79 cases in reinforcement concreters and related and 1.32 in building workers, while very few cases of ‘occupational cobalt CA’ were reported in these four occupations (Kanerva et al., 2000). However, in Finland, ferrous sulphate has been added to cement since 1987, which led to a decline in chromate CA by supposedly two-thirds (while irritant OCD remained constant) (Roto et al., 1996); a similar beneficial effect has been seen in Denmark (Avnstorp, 1992; Zachariae et al., 1996). Thus, CA incidence is presumably much lower in Finland than in Germany; a direct estimation is yet not possible.
In Germany, the chromium content has been regulated only recently by an amendment to the ‘technical rule for hazardous substance no. 613’ (Anonymous, 2002). In addition, manufacturers and suppliers of cement have agreed to add ferrous sulphate to cement delivered in sacks since January 2000 and to dry and wet mortar since July 2002, while ready mixed concrete, used on site or in prefabrication, still has its natural concentration of soluble chromate. These fairly recent, slight improvements cannot be reflected in our data as yet. In fact, chromate CA prevalence has so far remained almost stable in our data.
One main message of our results is not to disregard other CAs as causes of OCD, such as epoxy resin and, to a lesser extent, thiurams and IPPD. Once chromate is largely eliminated as a cause of allergic OCD, these occupational allergens will become relatively more important.
Cobalt allergy induced by cement is considered to be secondary to pre-existing (chromate) dermatitis (Fregert and Gruvberger, 1978). Monovalent cobalt allergy is also quite rare in construction workers according to other studies (e.g. Garcia and Armisen, 1985). The reason for such strong concordance is, according to clinical (Cavelier and Foussereau, 1995) and experimental (Liden and Wahlberg, 1994) studies, coupled exposure rather than structural, antigenetic similarity. The observation of a very low incidence of cobalt allergy in construction work (after introduction of ferrous sulphate to cement) (Kanerva et al., 2000) might indicate that the reduction in the content of soluble chromate and chromate allergy, respectively, will lead to a subsequent decrease in cobalt allergy, e.g. indirectly by lowering the number of cases of severe chromate dermatitis predisposing to cobalt allergy.
In view of a whole variety of allergens and irritants, the availability of suitable protective gloves, like nitrile-coated resilient textile gloves (Struppek and Ludwig, 1999), and promotion of actually wearing personal protective equipment can be regarded as the most suitable overall means of primary prevention. Suitable gloves, i.e. those which fit well, which are technically adequate and which have a low intrinsic risk to sensitize, afford protection not only against an array of potential allergens, but also against irritants. Irritant contact dermatitis, even if minimal initially, is, in turn, a pacemaker for CA and subsequent allergic OCD, which makes prevention even more important.
Acknowledgements—The study was supported by grant no. F5156 of the German Federal Institute for Occupational Safety and Health. The following centres of the IVDK contributed data to this analysis (in alphabetical order): Aachen (H. Dickel), Augsburg (O. Hirschsteiner, A. Ludwig), Berlin B.-Frank. (B. Tebbe, R. Treudler), Berlin Charité (J. Grabbe, T. Zuberbier), Berlin UKRV (J. Grabbe, T. Zuberbier), Bochum (C. Szliska), Dortmund (P.J. Frosch, C. Pirker), Dresden (G. Richter), Duisburg (J. Schaller), Erlangen (K.-P. Peters, M. Fartasch), Essen (H.-M. Ockenfels, U. Hillen), Graz (W. Aberer, B. Kränke), Göttingen (Th. Fuchs, J. Geier), Halle (G. Gaber, D. Lübbe), Hamburg (M. Kiehn, D. Vieluf), Heidelberg (A. Schulze-Dirks, M. Hartmann), Homburg / Saar (P. Koch), Jena (M. Gebhardt, A. Bauer), Kiel (J. Brasch), Lübeck (J. Kreusch, J. Grabbe), Magdeburg (U. Jappe, E. Weisshaar), Mainz (D. Becker), Mannheim (C. Bayerl), Marburg (I. Effendy), München LMU (B. Przybilla, F. Rueff), München Schwabing (M. Agathos), München TU (J. Rakoski), Nürnberg (I. Müller), Osnabrück (W. Uter, S.John), Rostock (H. Heise), Tübingen (G. Lischka), Ulm (H. Gall
), Wuppertal (O. Mainusch, J. Raguz).
|
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: +49 9131 8522750; fax: +49 9131 8522721; e-mail: wolfgang.uter{at}rzmail.uni-erlangen.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Anonymous. (1992) Klassifizierung der Berufe: systematisches und alphabetisches Verzeichnis der Berufsbenennungen. Wiesbaden: Bundesamt für Statistik.
Anonymous. (2002) Technische Regeln für Gefahrstoffe—TRGS 613: Ersatzstoffe, Ersatzverfahren und Verwendungsbeschränkungen für chromathaltige Zemente und chromathaltige zementhaltige Zubereitungen (Ausgabe Oktober 2002). Bundesarbeitsblatt 72–81.
Avnstorp C. (1992) Cement eczema. An epidemiological intervention study. Acta Derm Venereol Suppl (Stockh); 179: 1–22.[Medline]
Avnstorp C. (2000) Construction workers. In Kanerva L, Elsner P, Wahlberg J, Maibach HI, editors. Handbook of occupational dermatology. Berlin: Springer. pp. 890–2.
Bruze M. (1990) What is a relevant contact allergy? Contact Dermatitis; 23: 224–5.[Medline]
Cavelier C, Foussereau J. (1995) Kontaktallergie gegen Metalle und deren Salze. Teil II: Nickel, Kobalt, Quecksilber und Palladium. Derm Beruf Umwelt; 43: 152–62.
Cherry N, Meyer JD, Adisesh A, Brooke R, Owen-Smith V, Swales C, Beck MH. (2000) Surveillance of occupational skin disease: EPIDERM and OPRA. Br J Dermatol; 142: 1128–34.[CrossRef][Web of Science][Medline]
Dickel H, Kuss O, Blesius CR, Schmidt A, Diepgen TL. (2001) Report from the register of occupational skin diseases in northern Bavaria (BKH-N). Contact Dermatitis; 44: 258–9.[Medline]
Fregert S, Gruvberger B. (1978) Solubility of cobalt in cement. Contact Dermatitis; 4: 14–8.[Medline]
Garcia J, Armisen A. (1985) Cement dermatitis with isolated cobalt sensitivity. Contact Dermatitis; 12: 52.[Medline]
Gawkrodger DJ. (2001) Patch testing in occupational dermatology. Occup Environ Med; 58: 823–8.
Geier J, Uter W, Lessmann H, Schnuch A. (2002) Forschungsvorhaben ‘Frühzeitige Erkennung allergener Stoffe bei beruflicher und nicht-beruflicher Exposition’ (FaSt). Abschlussbericht des IVDK. Göttingen: IVDK.
Guo YL, Wang BJ, Yeh KC, Wang JC, Kao HH, Wang MT, Shih HC, Chen CJ. (1999) Dermatoses in cement workers in southern Taiwan. Contact Dermatitis; 40: 1–7.[Web of Science][Medline]
Kanerva L, Jolanki R, Estlander T, Alanko K, Savela A. (2000) Incidence rates of occupational allergic contact dermatitis caused by metals. Am J Contact Dermatitis; 11: 155–60.[CrossRef][Medline]
Kiec-Swierczynska M. (1990) Occupational dermatoses and allergy to metals in Polish construction workers manufacturing prefabricated building units. Contact Dermatitis; 23: 27–32.[CrossRef][Web of Science][Medline]
Liden C, Wahlberg JE. (1994) Cross-reactivity to metal compounds studied in guinea pigs induced with chromate or cobalt. Acta Derm Venereol; 74: 341–3.[Medline]
O’Malley M, Thun M, Morrison J, Mathias CGT, Halperin WE. (1988) Surveillance of occupational skin disease using the supplementary data system. Am J Ind Med; 13: 291–9.[Web of Science][Medline]
Plinske W, Drexel G, Drexel-Schlund C, Butz M. (2001) BK-DOK ’99. Dokumentation des Berufskrankheiten-Geschehens in Deutschland. St Augustin: HVBG.
Roto P, Sainio H, Reunala T, Laippala P. (1996) Addition of ferrous sulfate to cement and risk of chromium dermatitis among construction workers. Contact Dermatitis; 34: 43–50.[CrossRef][Web of Science][Medline]
Rühl R, Hadrich D. (2001) Cement induced skin damage. Croner’s Occup Hyg Mag; 29: 17–23.
Schnuch A, Geier J, Uter W, Frosch PJ, et al. (1997) National rates and regional differences in sensitization to allergens of the standard series. Population adjusted frequencies of sensitization (PAFS) in 40,000 patients from a multicenter study (IVDK). Contact Dermatitis; 37: 200–9.[CrossRef][Web of Science][Medline]
Schnuch A, Aberer W, Agathos M, Brasch J, Frosch PJ, Fuchs T, Richter G. (2001) Leitlinien der Deutschen Dermatologischen Gesellschaft (DDG) zur Durchführung des Epikutantests mit Kontaktallergenen. Hautarzt; 52: 864–866.[Medline]
Struppek K, Ludwig S. (1999) Prävention des Chromat-Ekzems im Baugewerbe. Derm Beruf Umwelt; 47: 13–15.
Uter W, Diepgen TL, Arnold R, Hillebrand O, Pietrzyk PM, Stüben O, Schnuch A. (1992) The Informational Network of Departments of Dermatology in Germany—a multicenter project for computer-assisted monitoring of contact allergy—electronic data processing aspects. Derm Beruf Umwelt; 40: 142–9.
Uter W, Schnuch A, Geier J, Frosch PJ. (1998) The epidemiology of contact dermatitis. The information network of departments of dermatology (IVDK) in Germany. Eur J Dermatol; 8: 36–40.[Web of Science][Medline]
Wahlberg JE. (2001) Patch testing. In Rycroft RJG, Menné T, Frosch PJ, Leppoittevin J-P, editors. Textbook of contact dermatitis. Berlin: Springer. pp. 435–468.
Zachariae CO, Agner T, Menné T. (1996) Chromium allergy in consecutive patients in a country where ferrous sulfate has been added since 1981. Contact Dermatitis; 35: 83–5.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. OGDEN Annals of Occupational Hygiene at Volume 50: Many Achievements, a Few Mistakes, and an Interesting Future Ann. Hyg., November 1, 2006; 50(8): 751 - 764. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||