The preventable burden of work-related ill-health

The preventable burden of work-related ill-health Abstract Background The fraction of ill-health overall attributable to occupational conditions has not been extensively evaluated, thus contributing to the perception of a lesser relevance of education and research in occupational health in respect to other fields of medical research and practice. Aims To assess the relevance of work-related conditions on the aetiology of human ill-health in different health domains. Methods We extracted the risk estimates associated with heritability and with occupational risk factors for chronic lymphocytic leukaemia (CLL), major depressive disorder (MDD) and long QT syndrome (LQTS) from 13 published international reports. The selection criteria for the eligible studies were: genome-wide studies, or studies of the occupational risk factors associated with one of the three diseases of interest. We calculated and compared the respective population attributable fraction for the combined occupational risk factors, and for heritability. Results We estimated that occupational risk factors would account for 12% (95% confidence interval (CI) 4–19) of CLL, 11% (95% CI 7–15) of MDD and 10% (95% CI 2–13) of LQTS burden in the general population. The corresponding figures for heritability would be 16% (95% CI 11–22), 28% (95% CI 20–5) and 17% (95% CI 7–27). Conclusions More efforts in capacity building and research in occupational health are warranted aiming to prevent ill-health and to preserve a productive life for the ageing work population. Genetics, health impact assessment, medical education, occupational health, public health policy Introduction In recent years, Rushton et al. [1] confirmed the 1981 Doll and Peto estimate of 4% of all cancers combined (excluding non-melanoma skin cancer) attributable to occupational factors [2]. Despite a substantial literature on risk of cardiovascular diseases and exposure to particulate matter, lead, noise, shiftwork and psychosocial stress, the burden of occupational exposures on cardiovascular diseases and diseases other than cancer in general has not been adequately evaluated [3], perhaps contributing to a diminishing interest towards education and research in occupational health [4]. To provide examples of the extent to which occupational medicine can contribute to reduce the burden of human ill-health, we conducted an exercise on three diseases for which both occupational and genetic risk factors had been investigated in detail. These diseases were: 1. Chronic lymphocytic leukaemia (CLL): a malignancy of the lymphohaemopoietic system due to a monoclonal expansion of mature small circulating lymphocytes. Occupational exposures reportedly associated with an increased CLL risk include contact with chicken meat, ethylene oxide (a sterilizing agent used in the chemical industry, food industry and hospitals), organophosphate insecticides, and organic solvents including benzene [5–8]. An estimate of the heritability due to common polygenic polymorphisms, and of its overall impact at the population level on CLL aetiology, was published based on a comprehensive pooled analysis of genome-wide association studies (GWAS) [9]. 2. Major depressive disorder (MDD): evidence linking work-related stress to MDD risk comes from a review of international cohort studies on two known work-related stress determinants, namely lack of control and effort–reward imbalance [10]. Two papers estimated MDD heritability, one based on the Scottish Family Health Study GWAS analysis [11], and the second on the Psychiatrics Genomics Consortium study [12]. 3. Long QT syndrome (LQTS): a hereditary or acquired condition of prolonged and inhomogeneous ventricular repolarization detected by non-invasive electrocardiography, and known to be associated with an increased risk of sudden death [13]. Congenital forms of LQTS have been associated with mutation in cLQTS genes encoding for ion channels [14]. When resulting in >50% reduction in channel current, mutations in the LQTS genes are significant predictors of LQT events, such as major arrhythmia, cardiac arrest and sudden death [15,16]. Acquired conditions associated with a prolonged QT interval include work-related stress from high job strain, effort–reward imbalance and nightshift work [17,18]. Based on GWAS results, the proportion of variance in QT length explained by autosomal single nucleotide polymorphisms (SNPs) was 0.168 (SE 0.052) [19], which we considered as representative of LQTS heritability. The aims of our study were 2-fold: (i) to estimate the weight of occupational factors in the aetiology of diseases in various domains of human health and (ii) to use such estimates to support the importance of occupational health in healthcare, research and the academy. We thus emphasize that, among the environmental conditions contributing to disease, occupational exposures require due acknowledgement as a major independent factor. Methods The three diseases we selected for our exercise were those satisfying the following criteria: (i) belonging to diverse medical domains, such as oncology, psychiatry and cardiology; (ii) not being acknowledged as typically related to occupational factors and (iii) to have been the object of publications covering both occupational risk factors, identified though a careful exposure assessment, and genetic factors investigated through GWAS. For GWAS studies, when available, we selected the publications on heritability as derived from the common gene polymorphisms associated with risk of the disease in question. To compare the weight of occupational and genetic factors in the aetiology of CLL, MDD and LQTS, we calculated the combined attributable fraction for occupational risk factors associated with each disease and derived the estimates of heritability from published reports. The conceptual analogy between the two metrics makes such comparison reasonable: in fact, both express the fraction of disease explained by a given risk factor, which in the case of heritability, consists of transmissible common gene polymorphisms [20]. Two studies provided estimates of MDD heritability: in the Scottish Family Health Study, the fraction of MDD heritability based on GWAS results was 0.251 (SE 0.099) [11]; in the Psychiatrics Genomics Consortium study [12] it was 0.285 (SE 0.022). We assumed the weighted average of these two estimates to represent the rate of MDD heritability in the general population. As we were unable to find publications on LQTS heritability, we considered the proportion of variance in QT length explained by autosomal SNPs in the only published GWAS study [13] as representative of LQTS heritability. To calculate the disease fraction attributable to occupational exposures at the population level, we used the prevalence of the exposed to the CLL risk factors among the controls from the respective studies [5–8], and the prevalence of the two MDD occupational risk factors, namely lack of control and effort–reward imbalance, among the non-depressed study subjects of the European SHARE survey [10]. To estimate the burden of LQTS due to occupational risk factors, we extracted the prevalence of exposed to high job strain and effort–reward imbalance among the controls from reference N. 17, and the prevalence of night shift workers in the general population from the US Bureau of Labor Statistics [21]. In all three examples, the procedure for calculating the population attributable fraction was the following: 1. Calculation of the attributable fraction for each specific occupational exposure among the exposed (AFe), based on the published risk estimates (relative risk (RR)):  AFe = (RR−1)/RR (1) 2. Calculation of the attributable fraction at the population level (AFp) for each occupational factor, using their estimated prevalence among the general population (prevalence of the exposed [Pe]) and the AFe value calculated as described above:  AFp =AFe × CF (2) 3. Calculation of the burden of disease attributable to occupation as a whole by combining the individual AFp values, as suggested by Steenland and Armstrong [22]:  AFoverall = [1−∏k(1−AFk)] (3) where the attributable fraction of the occupational exposures combined is the complementary probability of the overall product of the complementary probabilities of the k exposures in the set. Approval from an ethics committee was not sought, as our study used data from published reports, all of which had been approved by the competent ethics committee. Results Heritability would explain 16% (95% confidence interval (CI) 11–22%) of the CLL burden in the general population. On the other hand, we estimated that four occupational exposures account for 12% (95% CI 4–19) (Table 1). Table 1. Estimate of the attributable fraction of CLL, MDD and LQTS associated with occupational factors and heritability from common gene polymorphisms at the general population level Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  View Large Table 1. Estimate of the attributable fraction of CLL, MDD and LQTS associated with occupational factors and heritability from common gene polymorphisms at the general population level Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  View Large Heritability would account for 28% (95% CI 20–35) of MDD cases occurring among the general population. On the other hand, the attributable fraction of the disease related to effort–reward imbalance and low job control would account for 11% (95% CI 7–15). GWAS-based estimates suggest 17% (95% CI 7–27) LQTS cases to result from common gene polymorphisms, while working conditions, including irregular 6-h rotating nightshifts, job strain and effort–reward imbalance, would explain 10% (95% CI 2–13). Discussion Our results suggest that a few work-related factors account for a sizeable share of disease burden in disparate domains of clinical medicine. In our examples, heritability, as derived from genome-wide scans, would account for larger shares. However, the estimates of the occupational share and the genetic share are within the same order of magnitude, and only the former is feasible to prevent. In our study, the role of occupational risk factors might be underestimated for the following reasons: (i) only a handful of occupational risk factors for the selected diseases were documented properly enough to contribute to our analysis, compared with the extensive GWAS coverage; (ii) the smaller size of occupational studies results in lesser precision of the risk estimates compared with GWAS risk estimates and (iii) in case of true associations, the inter-individual variability of occupational exposure, and the uncertainty in defining it at toxicologically relevant levels in retrospective studies, result in underestimating the strength of the association. On the other hand, overestimating the role of occupational exposures might have resulted from combining the AFp associated with each individual occupational exposure, as we assumed mutual independence between such exposures. This is most likely for the diverse occupational risk factors for CLL. However, when considering psychosocial risk factors, these would more likely occur jointly, so that the attributable fraction for the combined effects would be overestimated. Nevertheless, our findings point to the need of reversing the current trend of lesser consideration towards education and research in occupational health in respect to other medical disciplines. In the last few decades, the interest of academic bodies and research funding agencies in occupational health has been progressively diminishing, as testified also by the relative weight of occupational medicine in the major medical journals being at its lowest, with only 0.5% of their articles dedicated to this specialty [23]. Several academic programmes have been shut down, and attracting a reasonable share of research funding is becoming increasingly difficult in respect to other mainstream science disciplines. Such circumstances have resulted in decreasing educational opportunities, which in turn has affected the ability to detect and diagnose occupational diseases. On the other hand, acknowledgement of the relevance of occupational health research and education would translate into a longer and more productive life for the working population. Pursuing these objectives implies investments for education and research in occupational health. Key points In this study, we estimated that work-related factors accounted for a sizeable share of the disease burden. Occupational determinants of cardiovascular, malignant and psychological disease diseases are potentially preventable. However, this is poorly recognized and warrants better academic education. Competing interest Neither co-author has any financial and personal relationships with other people or organizations that could inappropriately influence (bias) their work. Both co-authors research and teach occupational health in their respective academic bodies, and both are members of the International Commission of Occupational Health, the UK Faculty of Occupational Medicine and the UK Society of Occupational Medicine. References 1. Rushton L, Hutchings SJ, Fortunato Let al.   Occupational cancer burden in Great Britain. Br J Cancer  2012; 107( Suppl. 1): S3– S7. Google Scholar CrossRef Search ADS PubMed  2. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst  1981; 66: 1191– 1308. Google Scholar CrossRef Search ADS PubMed  3. Rushton L. The global burden of occupational disease. Curr Environ Health Rep  2017; 4: 340– 348. Google Scholar CrossRef Search ADS PubMed  4. Gehanno JF, Rollin L, Ladner J, Darmoni SJ. How is occupational medicine represented in the major journals in general medicine? Occup Environ Med  2012; 69: 603– 605. Google Scholar CrossRef Search ADS PubMed  5. Moore T, Brennan P, Becker Net al.   Occupational exposure to meat and risk of lymphoma: a multicenter case-control study from Europe. Int J Cancer  2007; 121: 2761– 2766. Google Scholar CrossRef Search ADS PubMed  6. Cocco P, t’Mannetje A, Fadda Det al.   Occupational exposure to solvents and risk of lymphoma subtypes: results from the Epilymph case-control study. Occup Environ Med  2010; 67: 341– 347. Google Scholar CrossRef Search ADS PubMed  7. Kiran S, Cocco P, Mannetje Aet al.   Occupational exposure to ethylene oxide and risk of lymphoma. Epidemiology  2010; 21: 905– 910. Google Scholar CrossRef Search ADS PubMed  8. Cocco P, Satta G, Dubois Set al.   Lymphoma risk and occupational exposure to pesticides: results of the Epilymph study. Occup Environ Med  2013; 70: 91– 98. Google Scholar CrossRef Search ADS PubMed  9. Sampson JN, Wheeler WA, Yeager Met al.   Analysis of heritability and shared heritability based on genome-wide association studies for thirteen cancer types. J Natl Cancer Inst  2015; 107: djv279. Google Scholar CrossRef Search ADS PubMed  10. Siegrist J, Lunau T, Wahrendorf M, Dragano N. Depressive symptoms and psychosocial stress at work among older employees in three continents. Global Health  2012; 8: 27. Google Scholar CrossRef Search ADS PubMed  11. Fernandez-Pujals AM, Adams MJ, Thomson Pet al.   Epidemiology and heritability of major depressive disorder, stratified by age of onset, sex, and illness course in Generation Scotland: Scottish Family Health Study (GS:SFHS). PLoS One  2015; 10: e0142197. Google Scholar CrossRef Search ADS PubMed  12. Zeng Y, Navarro P, Fernandez-Pujals AMet al.  ; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium. A combined pathway and regional heritability analysis indicates NETRIN1 pathway is associated with major depressive disorder. Biol Psychiatry  2017; 81: 336– 346. Google Scholar CrossRef Search ADS PubMed  13. Straus SM, Kors JA, De Bruin MLet al.   Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol  2006; 47: 362– 367. Google Scholar CrossRef Search ADS PubMed  14. Viskin S. Long QT syndromes and torsade de pointes. Lancet  1999; 354: 1625– 1633. Google Scholar CrossRef Search ADS PubMed  15. Moss AJ, Shimizu W, Wilde AAet al.   Clinical aspects of type-1 long-QT syndrome by location, coding type, and biophysical function of mutations involving the KCNQ1 gene. Circulation  2007; 115: 2481– 2489. Google Scholar CrossRef Search ADS PubMed  16. Noseworthy PA, Havulinna AS, Porthan Ket al.   Common genetic variants, QT interval, and sudden cardiac death in a Finnish population-based study. Circ Cardiovasc Genet  2011; 4: 305– 311. Google Scholar CrossRef Search ADS PubMed  17. Hintsa T, Määttänen I, Hintsanen Met al.   Work stress and the long QT syndrome: high job strain and effort-reward imbalance at work associated with arrhythmic risk in the long QT syndrome. J Occup Environ Med  2013; 55: 1387– 1393. Google Scholar CrossRef Search ADS PubMed  18. Meloni M, Setzu D, Del Rio A, Campagna M, Cocco P. QTc interval and electrocardiographic changes by type of shift work. Am J Ind Med  2013; 56: 1174– 1179. Google Scholar CrossRef Search ADS PubMed  19. Yang J, Manolio TA, Pasquale LRet al.   Genome partitioning of genetic variation for complex traits using common SNPs. Nat Genet  2011; 43: 519– 525. Google Scholar CrossRef Search ADS PubMed  20. Ramakrishnan V, Thacker LR. Population attributable fraction as a measure of heritability in dichotomous twin data. Commun Stat Simulat Comput  2012; 41. 21. Bureau of Labor Statistics. Beginning and Ending Hours: Full Time Wage and Salary Workers, May 2004 . Washington, DC: United States Department of Labor; July, 2005. http://www.bls.gov/news.release/flex.t07.htm. 22. Steenland K, Armstrong B. An overview of methods for calculating the burden of disease due to specific risk factors. Epidemiology  2006; 17: 512– 519. Google Scholar CrossRef Search ADS PubMed  23. Gehanno JF, Rollin L, Ladner J, Darmoni SJ. How is occupational medicine represented in the major journals in general medicine? Occup Environ Med  2012; 69: 603– 605. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Occupational Medicine Oxford University Press

The preventable burden of work-related ill-health

Occupational Medicine , Volume Advance Article – Apr 12, 2018

Loading next page...
 
/lp/ou_press/the-preventable-burden-of-work-related-ill-health-GrTnOcM38J
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0962-7480
eISSN
1471-8405
D.O.I.
10.1093/occmed/kqy063
Publisher site
See Article on Publisher Site

Abstract

Abstract Background The fraction of ill-health overall attributable to occupational conditions has not been extensively evaluated, thus contributing to the perception of a lesser relevance of education and research in occupational health in respect to other fields of medical research and practice. Aims To assess the relevance of work-related conditions on the aetiology of human ill-health in different health domains. Methods We extracted the risk estimates associated with heritability and with occupational risk factors for chronic lymphocytic leukaemia (CLL), major depressive disorder (MDD) and long QT syndrome (LQTS) from 13 published international reports. The selection criteria for the eligible studies were: genome-wide studies, or studies of the occupational risk factors associated with one of the three diseases of interest. We calculated and compared the respective population attributable fraction for the combined occupational risk factors, and for heritability. Results We estimated that occupational risk factors would account for 12% (95% confidence interval (CI) 4–19) of CLL, 11% (95% CI 7–15) of MDD and 10% (95% CI 2–13) of LQTS burden in the general population. The corresponding figures for heritability would be 16% (95% CI 11–22), 28% (95% CI 20–5) and 17% (95% CI 7–27). Conclusions More efforts in capacity building and research in occupational health are warranted aiming to prevent ill-health and to preserve a productive life for the ageing work population. Genetics, health impact assessment, medical education, occupational health, public health policy Introduction In recent years, Rushton et al. [1] confirmed the 1981 Doll and Peto estimate of 4% of all cancers combined (excluding non-melanoma skin cancer) attributable to occupational factors [2]. Despite a substantial literature on risk of cardiovascular diseases and exposure to particulate matter, lead, noise, shiftwork and psychosocial stress, the burden of occupational exposures on cardiovascular diseases and diseases other than cancer in general has not been adequately evaluated [3], perhaps contributing to a diminishing interest towards education and research in occupational health [4]. To provide examples of the extent to which occupational medicine can contribute to reduce the burden of human ill-health, we conducted an exercise on three diseases for which both occupational and genetic risk factors had been investigated in detail. These diseases were: 1. Chronic lymphocytic leukaemia (CLL): a malignancy of the lymphohaemopoietic system due to a monoclonal expansion of mature small circulating lymphocytes. Occupational exposures reportedly associated with an increased CLL risk include contact with chicken meat, ethylene oxide (a sterilizing agent used in the chemical industry, food industry and hospitals), organophosphate insecticides, and organic solvents including benzene [5–8]. An estimate of the heritability due to common polygenic polymorphisms, and of its overall impact at the population level on CLL aetiology, was published based on a comprehensive pooled analysis of genome-wide association studies (GWAS) [9]. 2. Major depressive disorder (MDD): evidence linking work-related stress to MDD risk comes from a review of international cohort studies on two known work-related stress determinants, namely lack of control and effort–reward imbalance [10]. Two papers estimated MDD heritability, one based on the Scottish Family Health Study GWAS analysis [11], and the second on the Psychiatrics Genomics Consortium study [12]. 3. Long QT syndrome (LQTS): a hereditary or acquired condition of prolonged and inhomogeneous ventricular repolarization detected by non-invasive electrocardiography, and known to be associated with an increased risk of sudden death [13]. Congenital forms of LQTS have been associated with mutation in cLQTS genes encoding for ion channels [14]. When resulting in >50% reduction in channel current, mutations in the LQTS genes are significant predictors of LQT events, such as major arrhythmia, cardiac arrest and sudden death [15,16]. Acquired conditions associated with a prolonged QT interval include work-related stress from high job strain, effort–reward imbalance and nightshift work [17,18]. Based on GWAS results, the proportion of variance in QT length explained by autosomal single nucleotide polymorphisms (SNPs) was 0.168 (SE 0.052) [19], which we considered as representative of LQTS heritability. The aims of our study were 2-fold: (i) to estimate the weight of occupational factors in the aetiology of diseases in various domains of human health and (ii) to use such estimates to support the importance of occupational health in healthcare, research and the academy. We thus emphasize that, among the environmental conditions contributing to disease, occupational exposures require due acknowledgement as a major independent factor. Methods The three diseases we selected for our exercise were those satisfying the following criteria: (i) belonging to diverse medical domains, such as oncology, psychiatry and cardiology; (ii) not being acknowledged as typically related to occupational factors and (iii) to have been the object of publications covering both occupational risk factors, identified though a careful exposure assessment, and genetic factors investigated through GWAS. For GWAS studies, when available, we selected the publications on heritability as derived from the common gene polymorphisms associated with risk of the disease in question. To compare the weight of occupational and genetic factors in the aetiology of CLL, MDD and LQTS, we calculated the combined attributable fraction for occupational risk factors associated with each disease and derived the estimates of heritability from published reports. The conceptual analogy between the two metrics makes such comparison reasonable: in fact, both express the fraction of disease explained by a given risk factor, which in the case of heritability, consists of transmissible common gene polymorphisms [20]. Two studies provided estimates of MDD heritability: in the Scottish Family Health Study, the fraction of MDD heritability based on GWAS results was 0.251 (SE 0.099) [11]; in the Psychiatrics Genomics Consortium study [12] it was 0.285 (SE 0.022). We assumed the weighted average of these two estimates to represent the rate of MDD heritability in the general population. As we were unable to find publications on LQTS heritability, we considered the proportion of variance in QT length explained by autosomal SNPs in the only published GWAS study [13] as representative of LQTS heritability. To calculate the disease fraction attributable to occupational exposures at the population level, we used the prevalence of the exposed to the CLL risk factors among the controls from the respective studies [5–8], and the prevalence of the two MDD occupational risk factors, namely lack of control and effort–reward imbalance, among the non-depressed study subjects of the European SHARE survey [10]. To estimate the burden of LQTS due to occupational risk factors, we extracted the prevalence of exposed to high job strain and effort–reward imbalance among the controls from reference N. 17, and the prevalence of night shift workers in the general population from the US Bureau of Labor Statistics [21]. In all three examples, the procedure for calculating the population attributable fraction was the following: 1. Calculation of the attributable fraction for each specific occupational exposure among the exposed (AFe), based on the published risk estimates (relative risk (RR)):  AFe = (RR−1)/RR (1) 2. Calculation of the attributable fraction at the population level (AFp) for each occupational factor, using their estimated prevalence among the general population (prevalence of the exposed [Pe]) and the AFe value calculated as described above:  AFp =AFe × CF (2) 3. Calculation of the burden of disease attributable to occupation as a whole by combining the individual AFp values, as suggested by Steenland and Armstrong [22]:  AFoverall = [1−∏k(1−AFk)] (3) where the attributable fraction of the occupational exposures combined is the complementary probability of the overall product of the complementary probabilities of the k exposures in the set. Approval from an ethics committee was not sought, as our study used data from published reports, all of which had been approved by the competent ethics committee. Results Heritability would explain 16% (95% confidence interval (CI) 11–22%) of the CLL burden in the general population. On the other hand, we estimated that four occupational exposures account for 12% (95% CI 4–19) (Table 1). Table 1. Estimate of the attributable fraction of CLL, MDD and LQTS associated with occupational factors and heritability from common gene polymorphisms at the general population level Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  View Large Table 1. Estimate of the attributable fraction of CLL, MDD and LQTS associated with occupational factors and heritability from common gene polymorphisms at the general population level Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  Author, year, reference  Risk factor  Risk estimate (95% CI)  AFe (95% CI)  Proportion exposed in the general population  AFp (95% CI)  CLL   Moore et al., 2007 [5]  Contact with chicken meat  1.55 (1.01–2.37)  0.35 (0.01–0.58)  0.054  0.019 (0.0005–0.031)   Cocco et al., 2010 [6]  Organic solvents, including benzene  1.30 (1.10–1.60)  0.15 (0.09–0.38)  0.410  0.062 (0.037–0.154)   Kiran et al., 2010 [7]  Ethylene oxide  2.00 (0.80–4.70)  0.50 (-0.25–0.79)  0.011  0.006 (-0.003–0.009)   Cocco et al., 2013 [8]  Organophosphorus insecticides  2.70 (1.20–6.00)  0.55 (0.17–0.83)  0.007  0.004 (0.001–0.006)    The four occupational risk factors combined  –  –  –  0.121 (0.036–0.192)   Sampson et al., 2015 [9]  Heritability from GWAS  –  –  –  0.164 (0.105–0.222)  MDD   Siegrist et al., 2012 [10]  Effort–reward imbalance  1.51 (1.28–1.78)  0.34 (0.22–0.44)  0.316  0.071 (0.046–0.092)   Siegrist et al., 2012 [10]  Low control  1.42 (1.20–1.68)  0.30 (0.17–0.40)  0.164  0.049 (0.027–0.066)    The two occupational risk factors combined  –  –  –  0.112 (0.072–0.152)   Zeng et al., 2017 [12]  Heritability from GWAS  –  –  –  0.276 (0.203–0.349)  LQTS   Meloni et al., 2013 [18]  Six-hour rotating night work shifts  10.5 (1.10–99.3)  0.90 (0.09–0.99)  0.04  0.036 (0.004–0.039)   Hintsa et al., 2013 [17]  Job strain  1.51 (1.13–2.02)  0.34 (0.12–0.50)  0.10  0.032 (0.011–0.048)   Hintsa et al., 2013 [17]  Effort–reward imbalance  1.86 (1.00–3.46)  0.46 (0.00–0.71)  0.07  0.032 (0.000–0.049)    The three occupational risk factors combined  –  –  –  0.097 (0.015–0.130)   Yang et al., 2011 [19]  Quote of variance due to common gene variation  –  –  –  0.168 (0.066–0.270)  View Large Heritability would account for 28% (95% CI 20–35) of MDD cases occurring among the general population. On the other hand, the attributable fraction of the disease related to effort–reward imbalance and low job control would account for 11% (95% CI 7–15). GWAS-based estimates suggest 17% (95% CI 7–27) LQTS cases to result from common gene polymorphisms, while working conditions, including irregular 6-h rotating nightshifts, job strain and effort–reward imbalance, would explain 10% (95% CI 2–13). Discussion Our results suggest that a few work-related factors account for a sizeable share of disease burden in disparate domains of clinical medicine. In our examples, heritability, as derived from genome-wide scans, would account for larger shares. However, the estimates of the occupational share and the genetic share are within the same order of magnitude, and only the former is feasible to prevent. In our study, the role of occupational risk factors might be underestimated for the following reasons: (i) only a handful of occupational risk factors for the selected diseases were documented properly enough to contribute to our analysis, compared with the extensive GWAS coverage; (ii) the smaller size of occupational studies results in lesser precision of the risk estimates compared with GWAS risk estimates and (iii) in case of true associations, the inter-individual variability of occupational exposure, and the uncertainty in defining it at toxicologically relevant levels in retrospective studies, result in underestimating the strength of the association. On the other hand, overestimating the role of occupational exposures might have resulted from combining the AFp associated with each individual occupational exposure, as we assumed mutual independence between such exposures. This is most likely for the diverse occupational risk factors for CLL. However, when considering psychosocial risk factors, these would more likely occur jointly, so that the attributable fraction for the combined effects would be overestimated. Nevertheless, our findings point to the need of reversing the current trend of lesser consideration towards education and research in occupational health in respect to other medical disciplines. In the last few decades, the interest of academic bodies and research funding agencies in occupational health has been progressively diminishing, as testified also by the relative weight of occupational medicine in the major medical journals being at its lowest, with only 0.5% of their articles dedicated to this specialty [23]. Several academic programmes have been shut down, and attracting a reasonable share of research funding is becoming increasingly difficult in respect to other mainstream science disciplines. Such circumstances have resulted in decreasing educational opportunities, which in turn has affected the ability to detect and diagnose occupational diseases. On the other hand, acknowledgement of the relevance of occupational health research and education would translate into a longer and more productive life for the working population. Pursuing these objectives implies investments for education and research in occupational health. Key points In this study, we estimated that work-related factors accounted for a sizeable share of the disease burden. Occupational determinants of cardiovascular, malignant and psychological disease diseases are potentially preventable. However, this is poorly recognized and warrants better academic education. Competing interest Neither co-author has any financial and personal relationships with other people or organizations that could inappropriately influence (bias) their work. Both co-authors research and teach occupational health in their respective academic bodies, and both are members of the International Commission of Occupational Health, the UK Faculty of Occupational Medicine and the UK Society of Occupational Medicine. References 1. Rushton L, Hutchings SJ, Fortunato Let al.   Occupational cancer burden in Great Britain. Br J Cancer  2012; 107( Suppl. 1): S3– S7. Google Scholar CrossRef Search ADS PubMed  2. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst  1981; 66: 1191– 1308. Google Scholar CrossRef Search ADS PubMed  3. Rushton L. The global burden of occupational disease. Curr Environ Health Rep  2017; 4: 340– 348. Google Scholar CrossRef Search ADS PubMed  4. Gehanno JF, Rollin L, Ladner J, Darmoni SJ. How is occupational medicine represented in the major journals in general medicine? Occup Environ Med  2012; 69: 603– 605. Google Scholar CrossRef Search ADS PubMed  5. Moore T, Brennan P, Becker Net al.   Occupational exposure to meat and risk of lymphoma: a multicenter case-control study from Europe. Int J Cancer  2007; 121: 2761– 2766. Google Scholar CrossRef Search ADS PubMed  6. Cocco P, t’Mannetje A, Fadda Det al.   Occupational exposure to solvents and risk of lymphoma subtypes: results from the Epilymph case-control study. Occup Environ Med  2010; 67: 341– 347. Google Scholar CrossRef Search ADS PubMed  7. Kiran S, Cocco P, Mannetje Aet al.   Occupational exposure to ethylene oxide and risk of lymphoma. Epidemiology  2010; 21: 905– 910. Google Scholar CrossRef Search ADS PubMed  8. Cocco P, Satta G, Dubois Set al.   Lymphoma risk and occupational exposure to pesticides: results of the Epilymph study. Occup Environ Med  2013; 70: 91– 98. Google Scholar CrossRef Search ADS PubMed  9. Sampson JN, Wheeler WA, Yeager Met al.   Analysis of heritability and shared heritability based on genome-wide association studies for thirteen cancer types. J Natl Cancer Inst  2015; 107: djv279. Google Scholar CrossRef Search ADS PubMed  10. Siegrist J, Lunau T, Wahrendorf M, Dragano N. Depressive symptoms and psychosocial stress at work among older employees in three continents. Global Health  2012; 8: 27. Google Scholar CrossRef Search ADS PubMed  11. Fernandez-Pujals AM, Adams MJ, Thomson Pet al.   Epidemiology and heritability of major depressive disorder, stratified by age of onset, sex, and illness course in Generation Scotland: Scottish Family Health Study (GS:SFHS). PLoS One  2015; 10: e0142197. Google Scholar CrossRef Search ADS PubMed  12. Zeng Y, Navarro P, Fernandez-Pujals AMet al.  ; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium. A combined pathway and regional heritability analysis indicates NETRIN1 pathway is associated with major depressive disorder. Biol Psychiatry  2017; 81: 336– 346. Google Scholar CrossRef Search ADS PubMed  13. Straus SM, Kors JA, De Bruin MLet al.   Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol  2006; 47: 362– 367. Google Scholar CrossRef Search ADS PubMed  14. Viskin S. Long QT syndromes and torsade de pointes. Lancet  1999; 354: 1625– 1633. Google Scholar CrossRef Search ADS PubMed  15. Moss AJ, Shimizu W, Wilde AAet al.   Clinical aspects of type-1 long-QT syndrome by location, coding type, and biophysical function of mutations involving the KCNQ1 gene. Circulation  2007; 115: 2481– 2489. Google Scholar CrossRef Search ADS PubMed  16. Noseworthy PA, Havulinna AS, Porthan Ket al.   Common genetic variants, QT interval, and sudden cardiac death in a Finnish population-based study. Circ Cardiovasc Genet  2011; 4: 305– 311. Google Scholar CrossRef Search ADS PubMed  17. Hintsa T, Määttänen I, Hintsanen Met al.   Work stress and the long QT syndrome: high job strain and effort-reward imbalance at work associated with arrhythmic risk in the long QT syndrome. J Occup Environ Med  2013; 55: 1387– 1393. Google Scholar CrossRef Search ADS PubMed  18. Meloni M, Setzu D, Del Rio A, Campagna M, Cocco P. QTc interval and electrocardiographic changes by type of shift work. Am J Ind Med  2013; 56: 1174– 1179. Google Scholar CrossRef Search ADS PubMed  19. Yang J, Manolio TA, Pasquale LRet al.   Genome partitioning of genetic variation for complex traits using common SNPs. Nat Genet  2011; 43: 519– 525. Google Scholar CrossRef Search ADS PubMed  20. Ramakrishnan V, Thacker LR. Population attributable fraction as a measure of heritability in dichotomous twin data. Commun Stat Simulat Comput  2012; 41. 21. Bureau of Labor Statistics. Beginning and Ending Hours: Full Time Wage and Salary Workers, May 2004 . Washington, DC: United States Department of Labor; July, 2005. http://www.bls.gov/news.release/flex.t07.htm. 22. Steenland K, Armstrong B. An overview of methods for calculating the burden of disease due to specific risk factors. Epidemiology  2006; 17: 512– 519. Google Scholar CrossRef Search ADS PubMed  23. Gehanno JF, Rollin L, Ladner J, Darmoni SJ. How is occupational medicine represented in the major journals in general medicine? Occup Environ Med  2012; 69: 603– 605. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Occupational MedicineOxford University Press

Published: Apr 12, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off