Public contributions to exposure assessment, occupational hygiene and exposure control.
IOHA 2018 Washington: Progress in the alignment of control banding tools
AIHCe 2017 Seattle: Testing Hazard Banding Performance and Strength
BOHS OH2017 Harrogate: Filling the OELV gap: Optimise Hazard Banding
NVvA Conference 2017: The alignment of parallel OH tools. An IOHA project?
IOHA 2015 Conference: The need for international alignment of Occupational Hygiene tools
The IOHA2024 Basic Characterisation PDC (Dublin 9 June). Download the zip-files
with handouts
and the Excel application IH-Aligner.
IH-Aligner is an Excel tool to process and export results of PAS measurements performed within a Similar Exposure Group (SEG) or an Exposure Scenario (ES). It performs the arithmetic adjustment of measurement results to the correct time-weighted average (TWA) matching with the OELV reference period. Furthermore, tests are performed on the goodness-of-fit (log)normality of the TWA set and several UTL95%,## calculated for testing compliance. The main benefit is that measurement results are converted into a format that can be imported (copied and pasted) into most freely available statistical programmes used in occupational hygiene. The use of IH-Aligner is free of charge. However, we would welcome feedback.
IOHA2024 Dublin 13 June. on the Rehab of measurement uncertainty in compliance testing.
NYF 2022 EN689 Conference. OELV compliance decisions and the EN689 preliminary test.
If the 95% percentile (C95%) of a log-normal shaped exposure distribution is around the Occupational Exposure Limit Value (OELV) and PAS measurements are taken within a SEG to test compliance with the OELV, then the EN689 5.5.2 preliminary test can unnecessarily delay compliance decisions and is less cost-effective for employers compared to the EN689 5.5.3 statistical test, which measures with 70% confidence whether fewer than 5% of exposures in the SEG exceed the OELV (C95,70% > OELV). The preliminary test decision schedule also underestimates workers' health risk and delays necessary control measures for GSD greater than 6.
IOHA 2018 Conference. Towards a library of exposure profile related GSD values
Occupational exposure assessment is normally characterised by measurement series with small sample sizes, compared with other quality-control areas such as environmental monitoring and product safety. A library of SEG or exposure profile (EN689 3.1) specific GSDs may support the appraiser (the EN689 synonym for assessor) to test the confidence of their sample GSD (is it too small or too large for the SEG examined) when six or fewer PAS measurements are available. Such a library provides scientists with a tool to make their Bayesian compliance testing methods more credible. Existing GSD info already debunks existing, historical myths that the long-term GSD must be smaller than 3 (IHSTat).
The pressure to publish and use only positive and 'significant' findings in OELV limit setting and exposure modelling drives journals and researchers to rearrange or reject data, support unvalidated methods, and selectively cite publications regardless of their validity or relevance. Institutes and journals in occupational exposure sciences should be transparent (make raw data available) and work according to the codes on research integrity.
SOPHYT Forum (Lille). Statistics and representative measurements
AEHI Seminario (Barcelona). The new CEN 689 Compliance testing guidance
CEN137 TC1 (Delft). Lognormality and the handling of non-detects
Deutsche Gesellschaft fuer Arbeitshygiene. BW_Statv1.xlsx The Excel tool for the BOHS-NVvA compliance testing guidance.
BOHS Conference (Manchester). Between-worker differences and sampling strategy flaws
NVvA Guidance on exposure data recording
Long-term exposure sampling at a Dutch (bio)chemical production site
On the compliance control of long-term OELVs. Vinyl chloride monomer (in Dutch, English summary)
WEEL Committee. Workplace Environmental Exposure Levels (OARS WEELs). Melamine. American Industrial Hygiene Association. Fairfax Virginia 1999.
HYGINIST: Compliance testing software for workplace exposure within a Similar Exposure Group (SEG) with goodness-of-fit, regression-based handling of non-detects, comparing sample GM and GSD with priors, and optimising sample size.
Scientific publication On the Strength and Validity of Hazard Banding and its supplementary data
NVvA Conference presentation (Scheveningen). Testing strength and validity of Hazard Band engines
AIHCe (San Antonio). Validation of health hazard grouping in Control Banding
BOHS Poster (Nottingham). A hierarchy of OELVs
Presentation H3##-statement based kick-off levels. Bilthoven NVT-AT 140925
Careful with that DNEL, Occupational Hygienist! A quantitative comparison with higher hierarchy OELVs
Dohsbase, a comprehensive database with hierarchical information on phys-chem, health hazard info, OELVs, sampling methods and risk ranking of chemicals used in the workplace.
Observations and analysis: 1% lost to follow-up. Mortality 39%. Overall Standardized Mortality Ratio (SMR) 85.9% (95% CI: 71.3-102.6%).
Of 275 cohort members, exposure could be quantified, resulting in 43,725 ppm-years.
One leukaemia case (non-acute myeloid) was observed in the second of three exposure categories, compared with 1.17 expected.
3.3 and 7.5 Additional leukaemia cases predicted by extreme benzene risk models (Austin, 1988 ; Rinsky, 1981) were rejected with probabilities due to chance of 0.03 and 0.0001 respectively.
Conclusion: The absence of leukaemia excess deaths in this cohort with long-term benzene exposure levels of 0.5-20 ppm per day, 159 ppm-years (520 mg/m3-years) per worker on the average, is consistent with the ECHA/RAC (2018) mode-of-action-based threshold approach for benzene-induced leukaemia.
Strengths: a well-defined industrial cohort with stable employment, lower overall mortality, and pure benzene exposure. A predefined research protocol and separated assessment of individual exposure and cause of death reduce the risk of observer bias, data dredging and post hoc manipulation of the dose-response.
The relatively high baseline health status, access to healthcare and lifestyle factors in the wealthy coal area in the Netherlands may have contributed to differences in observed risk estimates compared to other benzene cohorts.
Limitations: the cause of death was unknown for 5 of the 121 deaths, and cumulative exposure data were unavailable for 36 workers.
Noteworthy finding: a statistically significant excess of brain tumours was observed (6 cases versus 0.95 expected), a finding that has also been reported in industrial cohorts from countries with developed healthcare systems.
Mortality among laboratory assistants in a Coke plant [1928-1946] (in Dutch, English summary). The first properly conducted retrospective occupational cohort study in the Netherlands.