Sharing the knowledge gained from occupational cohort studies: a call for action.

OBJECTIVES: An immense body of knowledge has been created by establishing various job-exposure matrices (JEMs) to assess occupational exposures in community- and industry-based cohort studies. These JEMs could be made available to occupational epidemiologists using knowledge-sharing technologies, thereby saving considerable amounts of time and money for researchers investigating occupation-related research questions. In this paper, the authors give an example of how a detailed JEM can be easily transformed into a job-specific module (JSM) for use in community-based studies. METHODS: OccIDEAS is operationalised as a web-based software, combining the use of JSMs with an individual expert exposure assessment to assess occupational exposures in various industries according to a set of predefined rules. The authors used a JEM focusing on endocrine-disrupting chemicals from a German study on testicular cancer in the automobile industry to create a JSM in OccIDEAS. RESULTS: The JEM was easily translated into OccIDEAS requiring about 50 h of work by an epidemiologist familiar with the German JEM to learn about the OccIDEAS structure, establish the required set of exposure rules and to translate the JEM into OccIDEAS. Language did not represent an obstacle for translation either. To make the data available in an international context, an interpreter had to translate the German tasks and exposures after they were coded into OccIDEAS. CONCLUSIONS: JEMs which are constructed based on identifying tasks that determine exposure can be easily transformed into a JSM. Occupational epidemiologists are invited to contribute to the international scope of OccIDEAS by providing their previously established JEMs to make existing data on occupational exposures widely available to the epidemiological community.

Exposure to metal-working fluids in the automobile industry and the risk of male germ cell tumours.

INTRODUCTION: In a previous analysis of a case-control study of testicular cancer nested in a cohort of automobile workers, we observed an increased risk for testicular cancer among workers who had ever been involved in occupational metal-cutting tasks. We investigated whether this risk increase was due to exposure to metal-working fluids (MWF). METHODS: Occupational exposure to MWF was assessed in detail using a job-specific questionnaire for metal-cutting work. We calculated ORs and associated 95% CIs individually matched for age (±2 years) and adjusted for a history of cryptorchidism by conditional logistic regression. RESULTS: The prevalence of exposure to MWF was 39.8% among cases and 40.1% among controls. For total germ cell tumours and seminomas we did not observe risk increases for metal-cutting tasks or occupational exposure to MWF (OR 0.95; 95% CI 0.69 to 1.32 and OR 0.88; 95% CI 0.58 to 1.35, respectively). However, dermal exposure to oil-based MWF was associated with an increased risk for non-seminomatous testicular cancer. Dermal exposure to oil-based MWF for more than 5000 h showed particularly high risk estimates (OR 4.72; 95% CI 1.48 to 15.09). DISCUSSION: Long-term dermal exposure to oil-based MWF was a risk factor for the development of non-seminomatous testicular germ cell cancer. Possible measures to reduce exposure include the introduction of engineering control measures such as venting or enclosing of machines, and enforcing the use of personal protective equipment during metal cutting.

Development and evaluation of a tool for retrospective exposure assessment of selected endocrine disrupting chemicals and EMF in the car manufacturing industry.

OBJECTIVES: A system for retrospective occupational exposure assessment combining the efficiency of a job exposure matrix (JEM) and the precision of a subsequent individual expert exposure assessment (IEEA) was developed. All steps of the exposure assessment were performed by an interdisciplinary expert panel in the context of a case-control study on male germ cell cancer nested in the car manufacturing industries. METHODS: An industry-specific JEM was developed and automatic exposure estimation was performed based on this JEM. A subsample of exposure ratings was done by IEEA to identify determinants of disagreement between the JEM and the individual review. Possible determinants were analyzed by calculating odds ratios (ORs) of disagreement between ratings with regard to different dimensions (e.g. high versus low intensity of exposure). Disagreement in ≥20% of the sampled exposure ratings with a statistically significant OR was chosen as a threshold for inclusion of the exposure ratings into a final IEEA. RESULTS: The most important determinants of disagreement between JEM and individual review were working outside of the production line (disagreement 80%), low probability of exposure (disagreement 25%), and exposure depending on specific activities like usage of specific lacquers (disagreement 32%) for jobs within the production line. These determinants were the selection criteria of exposure ratings for the subsequent final IEEA. CONCLUSIONS: Combining a JEM and a subsequent final IEEA for a selected subset of exposure ratings is a feasible and labor-saving approach for exposure assessment in large occupational epidemiological studies.

Asbestos exposure and malignant lymphoma: a multicenter case-control study in Germany and Italy.

AIMS: To analyze the relationship between asbestos exposure and malignant lymphoma in a multicenter case-control study conducted in Germany and Italy according to a common core protocol. METHODS: Male and female patients with malignant lymphoma (n = 1,034) between 18 and 80 years of age were prospectively recruited in six study areas in Germany (Ludwigshafen/Upper Palatinate, Heidelberg/Rhine-Neckar-County, Würzburg/Lower Frankonia, Hamburg, Bielefeld/East Westphalia, and Munich) and in two study areas in Sardinia, Italy (Cagliari and Nuoro provinces). A total of 1,173 population control subjects were drawn from population registers. In a structured personal interview, we elicited a complete occupational history, including every occupational period that lasted at least 1 year. On the basis of job task-specific supplementary questionnaires, trained experts assessed the exposure to asbestos. As a measure of cumulative asbestos exposure on a time by intensity scale, fiber-years were calculated. 12 cases (1.2%) and 12 control subjects (1.0%) had a cumulative asbestos exposure of more than 2.6 fiber-years (highest exposure category according to the 90th percentile of exposed control subjects). Odds ratios (OR) and 95% confidence intervals (CI) were calculated using unconditional logistic regression analysis adjusted for age, sex and study region. Patients with specific lymphoma sub-entities were additionally compared with the entire control group. RESULTS: We observed no statistically significant association between cumulative asbestos exposure and the risk of any lymphoma subtype. An elevated risk was found for the association between exposure to more than 2.6 fiber-years and multiple myeloma (OR = 6.0; 95% CI 1.4-25.1); however, numbers were small (n = 3 cases, all of them from Italy; n = 12 control subjects). CONCLUSIONS: Our study does not support an association between asbestos exposure and risk of malignant lymphoma.

Case-control study of male germ cell tumors nested in a cohort of car-manufacturing workers: Findings from the occupational history.

BACKGROUND: To examine whether the previously observed excess risk of male germ cell cancer in a cohort of car-manufacturing workers can be attributed to occupational activities inside and/or outside the car industry. METHODS: A nested case-control study among workers in six plants included 205 cases of germ cell cancer and 1,105 controls, individually matched by year of birth (±2 years). Job periods of the individual occupational histories were coded based on the International Standard Classification of Occupations (ISCO) and the industrial classification of economic activities (NACE). Odds ratios (ORs) and corresponding 95%-confidence intervals (CI) for ever-never and cumulative employment were calculated by conditional multivariate logistic regression adjusted for cryptorchidism. RESULTS: Significantly increased risks were observed for machinery fitters and assemblers (A) (OR = 1.8, 95% CI 1.25-2.53) and "workers not elsewhere classified" (OR = 2.10, 95% CI 1.27-3.54), but no trend was observed for employment duration in either occupational group. Stratification of job group A by metal-cutting and non-cutting jobs yielded ORs of 1.87 (95% CI 1.31-2.67) and of 1.24 (95% CI 0.68-2.28), respectively. Among "plumbers, welders, sheet & structural metal workers" (adjusted OR 1.4, 95% CI 0.99-1.95) only "structural metal preparers and erectors" showed a substantially increased risk (OR = 2.30; 95% CI 1.27-4.27). CONCLUSIONS: Our results do not fully explain the increased incidence of germ cell cancer in the cohort, but support previous findings showing increased risks among metal workers. These risks were most strongly pronounced in metal-cutting activities.

Sun exposure and malignant lymphoma: a population-based case-control study in Germany.

Although some causes for malignant lymphoma are known their etiology is not well understood so far. We analyze the relationship between sun exposure and malignant lymphoma in a multicenter, population-based case-control study. Patients with malignant lymphoma (n = 710, 18-80 years) were prospectively recruited in 6 study regions in Germany. For each case, a gender, region and age-matched control was drawn from population-registers. In personal interviews, lifetime holidays spent in sunny climate, outdoor leisure activities and sunbed or sunlamp use were recorded. On basis of job task-specific supplementary questionnaires, an occupational physician assessed the cumulative working time outside. Odds ratios (OR) and 95%-confidence-intervals (CI) were calculated using conditional logistic regression analysis, adjusted for smoking and alcohol consumption. To increase statistical power, patients with specific lymphoma subentities were additionally compared with the entire control group using unconditional logistic regression. We observed a reduced overall lymphoma risk among subjects having spent vacations at sunny climates or frequently used sunbeds or sunlamps. The analysis of lymphoma subentities revealed similar results with the exception of T-NHL and follicular lymphoma which were positively associated with outdoor leisure activities. While cumulative working time outside appeared unrelated to NHL overall and most subentities, it was negatively associated with follicular lymphoma and weakly positively to HL. This data suggest that exposure to natural and artificial ultraviolet radiation may reduce the OR for lymphoma in this study population.

Solvent exposure and malignant lymphoma: a population-based case-control study in Germany.

AIMS: To analyze the relationship between exposure to chlorinated and aromatic organic solvents and malignant lymphoma in a multi-centre, population-based case-control study. METHODS: Male and female patients with malignant lymphoma (n = 710) between 18 and 80 years of age were prospectively recruited in six study regions in Germany (Ludwigshafen/Upper Palatinate, Heidelberg/Rhine-Neckar-County, Würzburg/Lower Frankonia, Hamburg, Bielefeld/Gütersloh, and Munich). For each newly recruited lymphoma case, a gender, region and age-matched (+/- 1 year of birth) population control was drawn from the population registers. In a structured personal interview, we elicited a complete occupational history, including every occupational period that lasted at least one year. On the basis of job task-specific supplementary questionnaires, a trained occupational physician assessed the exposure to chlorinated hydrocarbons (trichloroethylene, tetrachloroethylene, dichloromethane, carbon tetrachloride) and aromatic hydrocarbons (benzene, toluene, xylene, styrene). Odds ratios (OR) and 95% confidence intervals (CI) were calculated using conditional logistic regression analysis, adjusted for smoking (in pack years) and alcohol consumption. To increase the statistical power, patients with specific lymphoma subentities were additionally compared with the entire control group using unconditional logistic regression analysis. RESULTS: We observed a statistically significant association between high exposure to chlorinated hydrocarbons and malignant lymphoma (Odds ratio = 2.1; 95% confidence interval 1.1-4.3). In the analysis of lymphoma subentities, a pronounced risk elevation was found for follicular lymphoma and marginal zone lymphoma. When specific substances were considered, the association between trichloroethylene and malignant lymphoma was of borderline statistical significance. Aromatic hydrocarbons were not significantly associated with the lymphoma diagnosis. CONCLUSION: In accordance with the literature, this data point to a potential etiologic role of chlorinated hydrocarbons (particularly trichloroethylene) and malignant lymphoma. Chlorinated hydrocarbons might affect specific lymphoma subentities differentially. Our study does not support a strong association between aromatic hydrocarbons (benzene, toluene, xylene, or styrene) and the diagnosis of a malignant lymphoma.