Probability of causation in individual workers: Lung cancer due to occupational exposure to asbestos

Abstract Background Compensation systems for asbestos-related lung cancer commonly apply the Helsinki criteria, which assume a 4% excess relative risk per fibre-year. Under linearity, the relative risk (RR) doubles at 25 fibre-years, where the probability of causation (PoC), estimated as (RR−1)/ RR, reaches 50%. Recent large case-control studies suggest steeper exposure-response relations at lower exposures. We aimed to estimate PoC of asbestos-related lung cancer to evaluate exposure thresholds for compensation of lung cancer cases occupationally exposed to asbestos. Methods Relative risk of asbestos-related lung cancer was estimated using two approaches: A meta-regression of 22 occupational studies forming the core evidence on cumulative asbestos exposure and lung cancer since the 1980s (130,341 participants). A meta-analysis of the recently conducted SYNERGY pooled case-control study (14 studies, 37,866 participants), adjusted for age, sex, smoking, and study. PoC was estimated with 95% prediction intervals (95%PI) and interpreted as a lower bound. Results Occupational cohort studies showed a shallow non-linear exposure-response relation at low exposure ranges (0–65 fibre-years) with substantial heterogeneity (I2 = 92.7%). SYNERGY showed a steeper relation (6.8% increase in risk per fibre-year; 95%PI: 0%–17.5%) and lower heterogeneity (I2 = 63.4%). PoC ≥50% occurred at 62.9 (point estimate) and 18.2 fibre-years (upper 95%PI) for occupational studies, versus 10.5 and 4.3, respectively, in SYNERGY. Conclusions The SYNERGY study, a high-quality study with harmonised exposure assessment, provided exposure-response estimates relevant to current exposure to lower levels of mixed asbestos fibres in the Netherlands, supporting lower exposure thresholds than the existing Helsinki criteria when estimating PoC in compensation contexts. Key Messages What is already known on this topic Probability of causation (PoC) is applied in workers’ compensation with a PoC ≥50% as criterion, typically corresponding to a doubling in baseline disease risk; An exposure threshold of 25 fibre-years is used as the doubling in risk criterion for lung cancer attribution (2014 Helsinki criteria) in compensation systems worldwide; Upper confidence intervals (statistical uncertainty) of the PoC have been suggested as eligibility thresholds for compensation to weigh uncertainties in favour of workers. What this study adds This study uses recent and transparent quantitative meta-analyses for the relation between lung cancer and asbestos exposure as alternative for the Helsinki criterion of 25 fibre-years (4% excess relative risk per fibre-year); This study applies the upper bound of 95% prediction intervals of relative risk to estimate PoC, considering uncertainty because of between study differences in risk, which could be a more appropriate measure than statistical uncertainty of the meta-risk-estimate alone; Together with the use of recent general population studies, a lower exposure threshold (4.3 fibre-years) was obtained than when the Helsinki criterion was applied (25 fibre-years); In contrast, historical occupational cohort studies resulted in a threshold of 18.2 fibre-years. How this study might affect research, practice or policy Prediction intervals are a potentially useful tool to account for uncertainty arising from meta-analyses in compensation processes; Caution is warranted when interpreting PoC as it does not account for earlier cancer onset due to exposure. Future research could aim to address this gap.