Hardy, Vanessa Amar, Derek Ruths, Nicholas B. King

Abstract

Objective: Industry-funded research is a threat to the validity of scientific inference on carcinogenic hazards. Scientists require tools to better identify industry-funded studies and minimize the influence of industry bias in evidence synthesis reviews. We applied a novel large language model (LLM)-based tool named InfluenceMapper to demonstrate and evaluate its performance in identifying industry sponsored research on the carcinogenicity of two occupational exposures of commercial interest by major industries, benzene and cobalt. Materials and Methods: We identified all epidemiological, animal bioassay, and mechanistic studies included in systematic reviews on the carcinogenicity of benzene and cobalt conducted by the WHO/IARC Monographs Programme. InfluenceMapper extracted sponsoring entities disclosed in study publications and classified 40 possible relationship types between all entities and the study and each author. A human classified entities as industry or industry-funded. Positive predictive values described the extent of false positive relationships. Results: Analyses included 1,520 studies for both agents. We identified 184 disclosed industry or industry-funded entities from InfluenceMapper output that were involved in 453 distinct study-entity and author-entity relationships. For each agent, between 4-8% of studies were funded by industry and 1-4% of studies had at least one author that disclosed receiving industry funding. Industry trade associations funded 19 studies published in 14 journals over a 37-year span. After funding, the most prevalent disclosed relationships with industry were receiving data, holding employment, paid consulting, and providing expert testimony. Industry sponsored research consisted predominantly of mechanistic studies. Positive predictive values were excellent (>98%) for study-entity relationships but declined for relationships with individual authors. Conclusion: LLM-based tools can significantly expedite and bolster the detection of industry sponsored research in cancer prevention. Possible use cases include facilitating the assessment of bias from industry studies in evidence synthesis reviews and alerting scientists to the influence of industry on scientific inference about carcinogenic hazards.

Manon Truchon, Mahée Gilbert-Ouimet

Abstract

Objective While psychosocial risks at work (PRW) are known to negatively impact health and productivity, their interrelations and cumulative effects remain poorly understood. This study aims to test an integrative explanatory model—grounded in both the Job Demands-Resources model and the theory of Psychosocial Safety Climate (PSC)—to better understand how PRW interact and influence outcomes. By modeling the interrelations between PRW, this study seeks to identify key risks that act as amplifiers or triggers, offering organizations evidence-based guidance on which PRW to address first to maximize impact. Material and Methods In October-November 2024, 3000 French-speaking workers from a demographically diverse populational panel completed the French-language Questionnaire on Psychosocial Risks, Well-being, and Health at Work (QRBEST), a validated instrument measuring 22 PRW and 12 outcomes of well-being, health, and productivity. Structural equation modeling will be used to assess the fit and plausibility of the proposed explanatory model, where six latent factors (demands, resources, violence/harassment, PSC, health, and productivity) are expected to interact. Confirmatory factor analyses will first validate the factor structure; if needed, theory- and data-driven adjustments will be made. Once the measurement model is satisfactory, the structural model will be tested to evaluate the hypothesized pathways among factors. Results Preliminary analyses show good to excellent internal consistency for all QRBEST dimensions (α>.75). The factorial structure is supported by confirmatory factor analyses. Upcoming analyses will examine the model’s fit and hypothesized pathways and identify key PRW linked to adverse outcomes; findings will be ready for presentation at EPICOH. Conclusion This research will advance understanding of how various PRW interact to influence well-being, health, and productivity. The findings might support the development of more targeted and effective prevention strategies across diverse organizational contexts, especially those with limited occupational health resources. Funding: Fonds institutionnel de recherche de l’Université du Québec à Rimouski.

Delepee Raphael – Lebailly Pierre

Abstract

Epidemiological studies have already established a link between occupational agricultural exposures and cancers. However, despite extensive epidemiological study, the biological plausibility of the link between these exposures and cancer development has not been definitively confirmed. The aim of this study were i) to determine the effect of current smoking on DNA damage in blood cells and ii) the effect of various agricultural exposures including pesticides in a French agricultural cohort. Blood samples were collected from n=375 farmers, with 97 were current smokers, 97 former smokers and 185 never smokers. The alkaline comet assay version was applied on cryopreserved isolated lymphocytes isolated to evaluate DNA damage levels, evaluated by scoring a mean 800 nuclei per sample, classified in four categories by an AI Yolo recognition algorithm using a neural network developed in the laboratory. Socio-demographic information including smoking related questions, livestock and crop production of the farm and pesticide application schedule were collected by face-to face questionnaires. Univariate linear and logistic regression tests and multivariate analysis were performed. the initial results on smoking status show no clear difference in any of the qualitative criteria considered in our population. Smoking status should not be considered as a major cofounding factor in occupational studies using the alkaline comet assay. The 168 different molecules listed in the treatment schedules will be considered individually and grouped by Mode of Action (MoA), and farmers will be considered individually and clustered by treatment types of cultures and livestock activities All of this data will be correlated with the levels of DNA damage estimated by the comet assay. With very little literature evaluating genotoxic damages linked with agricultural activities, this study tends to be an exploratory study. We hope to find specific agricultural activities or treatment patterns links to an increase of DNA damages evaluated by comet assay.

José Rodolfo Tenório Lima, Felipe Seiti Sekiya, Rodrigo Pereyra de Sousa Coelho

Abstract

Objective: To analyze the incidence and characteristics of temporary disability benefits among workers in the Brazilian sugarcane agroindustry, comparing manual and mechanized labor processes. Methods: This ecological study analyzed secondary data from 2012 to 2021. Formal labor data were obtained from the Annual Social Information Report (RAIS) database, and data on social security benefits were sourced from the Smartlab platform. Workers were categorized into two groups: manual laborers (e.g., sugarcane field workers) and mechanized workers (e.g., machine operators). The incidence rates of temporary disability benefits (B31 – non-work-related and B91 – work-related) were calculated. Relative risk (RR) was used to compare groups, and causes of disability were classified using ICD-10. Results: During the period, 70,339 temporary disability benefits were granted in the sector, 75% to manual workers and 25% to mechanized ones. The average annual incidence was 28.8 per 1,000 for manual labor and 25.3 per 1,000 for mechanized labor. The overall RR for disability was 1.14, indicating a 14% higher risk for manual workers. The B91-specific RR was 1.71. Musculoskeletal disorders (MSDs) were more prevalent among manual workers, accounting for 33.7% of cases vs. 24.6% among mechanized workers. Among B91 cases, dorsalgia (M54) was the leading cause in manual workers (19%), while wrist/hand fractures (S62) predominated in mechanized roles (13%). Three of the five leading B91 causes in manual workers were MSDs, compared to only one in mechanized jobs. Conclusion: Despite a general decline in benefit concessions, significant inequalities remain. Manual workers face a higher burden of MSDs, while mechanized workers are more affected by severe traumatic injuries. The persistent high incidence of work-related benefits underscores the link between job conditions and health outcomes. Addressing these disparities requires stronger enforcement of labor standards, broader application of NR-31, enhanced training and safety measures, and policies tailored to the realities of rural labor in Brazil’s sugarcane sector.

Tosca O.E. de Crom, Iris Eekhout, Ellian Lebbink, Caren Lanting, Susan Peters, Anjoeka Pronk

Abstract

Objective: It is well established that several occupational exposures have carcinogenic properties; the precise extent to which these exposures contribute to the recent cancer incidence in the Netherlands remains unclear. We estimated the annual cancer burden attributable to six occupational carcinogens in the Netherlands. Material and Methods: We selected occupational exposures classified as Group 1 carcinogens by the International Agency for Research on Cancer (IARC) with high relevance to the Dutch working population. Selected exposures included asbestos, crystalline silica, diesel engine exhaust, and solar ultraviolet radiation (UV), as well as being employed as a welder or painter. Population attributable fractions (PAFs) were calculated using Levin’s equation for each exposure and associated cancer. Input data for the calculations included the proportions of the population exposed during the relevant exposure period (1971–2011) derived from Statistics Netherlands, and relative risks for each occupational exposure-cancer type combination obtained from existing literature. Cancer incidences from 2021, obtained from the Netherlands Cancer Registry, were combined with the PAFs to estimate the absolute numbers of attributable cases. Results: Of 126,717 new cancer diagnosis in the Netherlands in 2021, 3,932 (3.1%) were attributable to the occupational exposures in our study. Occupational cancer burden in men (5.4%) were higher than in women (0.5%). Solar UV exposure resulted in the largest cancer burden (1720 basal cell skin cancers), followed by asbestos (925 lung cancers, 463 mesotheliomas, 10 ovarian cancers, and 9 laryngeal cancers), occupation as painter (156 lung cancers, 83 bladder cancers), occupation as welder (201 lung cancers, 11 ocular melanomas), crystalline silica (233 lung cancers), and diesel engine exhaust (187 lung cancers). Conclusion: Occupational exposures contribute to thousands of cancer diagnosis in the Netherlands annually. Our estimates of cancer burden attributable to six important occupational carcinogens highlight areas for potential efforts for exposure and disease prevention.

Marine Renier , Juliette Hippert , Mathilde Bureau , Nicolas Vigneron , AGRICAN group, Isabelle Baldi , Pierre Lebailly ,

Abstract

Objective: Prostate cancer is the second most common and the 5th cause of cancer death among men worldwide. Several meta-analyses have linked the incidence of prostate cancer to agriculture and pesticide use. Among these, organophosphate insecticides are a large family that has been widely used throughout the world for decades and have been seldom studied. We assessed association between occupational exposure to 26 organophosphate insecticides and prostate cancer risk in the Agriculture and Cancer cohort (AGRICAN). Material and Methods: The PESTIMAT crop-exposure matrix was used to assess exposure to the 26 organophosphates studied, based on the probability, frequency and intensity of their use. A Cox regression model with age as the time scale was used to estimate the hazard ratio with 95% confidence intervals among men enrolled from 2005 to 2007 in the AGRICAN cohort, with follow-up until 31st December 2017. Results: With a total of 4,102 incident cases of prostate cancer out of 80,038 men, an increased prostate cancer risk was observed with duration and/or exposure score for the following 12 organophosphates: diazinon, malathion, dichlorvos, parathion-ethyl, chlorpyrifos ethyl, fonofos, phorate, terbufos, azinphos, mevinphos, parathion-methyl and trichlorfon. We also observed some inverse association with prostate cancer risk for a few molecules such as methidathion and oxydemeton-methyl. No confounding effects were detected in the relationship between exposure to each organophosphate and prostate cancer risk. Conclusion: Our study provides new informations about exposure to organophosphates and prostate cancer risk. Consideration of data from follow-up questionnaire will enable us to complete and refine the results.

Johanneke E. Oosterman, Suzan Wopereis, Martijn E.T. Dollé, Alex Burdorf, Linda W.M. van Kerkhof , Heidi M. Lammers-Van der Holst

Abstract

Objective Shift work disrupts sleep-wake rhythms, causing poor sleep, circadian misalignment, sleepiness, and reduced performance. This study evaluated the effect of personalized sleep and nutrition strategies on sustained attention in night shift workers. Methods In this 12-week intervention, 57 healthy male shift workers (18–60 years) were allocated to a personalized sleep (n=25), nutrition (n=22), or control (n=10) arm. The sleep group received guidance on sleep timing and strategic napping; the nutrition group received meal plans with timed calorie and macronutrient distribution. At baseline, early intervention (first shifts post-advice), and end of intervention, participants completed 5-minute Psychomotor Vigilance Tests (NASA PVT+ app) at the start and end of two consecutive shifts. Attentional lapses (reaction time >500 ms) was the primary outcome. Negative binomial mixed models included participant and shift as random intercepts, adjusted for age, chronotype, and household composition. Results Baseline lapse rates did not differ between groups. Across all groups and timepoints, lapse rates were over twice as high at shift end vs. start (Rate Ratio = 2.19, 95% CI [1.84, 2.61]). In controls, lapse rates increased from baseline to the early (RR = 2.19 [1.37, 3.49]) and end of the intervention (RR = 2.43 [1.45, 4.07]), with a similar pattern in the nutrition group. However, the sleep intervention mitigated this increase. The sleep group had 51% fewer lapses than controls during early intervention (RR = 0.49 [0.28, 0.88]); and this protective effect persisted at the end of the intervention (RR = 0.53 [0.28, 1.00]). Conclusion The sleep intervention reduced attentional lapses during night shifts relative to control and nutrition groups. While sustained attention declined in the other groups, the sleep group remained stable over time. These findings support integrating personalized sleep strategies into occupational health policy to improve safety and reduce error risk in 24-hour operations.