Occupational Exposure Assessment Tools in Europe: A Comprehensive Inventory Overview

Abstract

Objectives

The Network on the Coordination and Harmonisation of European Occupational Cohorts (OMEGA-NET) was set up to enable optimization of the use of industrial and general population cohorts across Europe to advance aetiological research. High-quality harmonized exposure assessment is crucial to derive comparable results and to enable pooled analyses. To facilitate a harmonized research strategy, a concerted effort is needed to catalogue available occupational exposure information. We here aim to provide a first comprehensive overview of exposure assessment tools that could be used for occupational epidemiological studies.

Methods

An online inventory was set up to collect meta-data on exposure assessment tools. Occupational health researchers were invited via newsletters, editorials, and individual e-mails to provide details of job-exposure matrices (JEMs), exposure databases, and occupational coding systems and their associated crosswalks to translate codes between different systems, with a focus on Europe.

Results

Meta-data on 36 general population JEMs, 11 exposure databases, and 29 occupational coding systems from more than 10 countries have been collected up to August 2021. A wide variety of exposures were covered in the JEMs on which data were entered, with dusts and fibres (in 14 JEMs) being the most common types. Fewer JEMs covered organization of work (5) and biological factors (4). Dusts and fibres were also the most common exposures included in the databases (7 out of 11), followed by solvents and pesticides (both in 6 databases).

Conclusions

This inventory forms the basis for a searchable web-based database of meta-data on existing occupational exposure information, to support researchers in finding the available tools for assessing occupational exposures in their cohorts, and future efforts for harmonization of exposure assessment. This inventory remains open for further additions, to enlarge its coverage and include newly developed tools.

What’s Important About This Paper?

OMEGA-NET was set up to enable optimization of the use of industrial and general population cohorts across Europe to advance aetiological research. This inventory forms the basis for a searchable web-based database of meta-data on existing occupational exposure information, to support researchers in finding the available tools for assessing occupational exposures in their cohorts.

Introduction

Prospective cohort studies are considered the strongest design in occupational epidemiology, as it often allows for obtaining information directly from individuals and updating information over time (Blair et al., 2015). Although many large cohorts with occupational information exist in Europe (Kogevinas et al., 2020), these are not yet used to their full potential in the study of occupational risks. Pooling these data would increase statistical power, offering opportunities such as looking at rare outcomes and interactions between risk factors, as well as enabling the exploration of between-countries differences (Turner and Mehlum, 2018). A major limiting factor is the lack of large-scale systematic and harmonized exposure assessment that is required for coordinated occupational health research (Peters et al., 2020). Good quality exposure assessment is essential to detect and characterize relevant exposure–disease associations.

The first step in the exposure assessment process often involves translating narrative descriptions of occupational histories into occupational codes, either manually or by using (semi-)automated systems to code such free text. These codes then offer the opportunity to link a study to a job-exposure matrix (JEM), i.e. a cross-tabulation between occupational title and workplace hazards.

JEMs are an important tool in large-scale and systematic exposure assessment (Kromhout and Vermeulen, 2001; Peters, 2020). JEMs are often based on expert judgement, but exposure (measurement) data can also be used to develop a JEM (Ge et al., 2018). Numerous JEMs have been developed and described over the years, all with their own coding systems and definitions of exposure. However, many different coding systems exist, and vary between countries and over time. Due to these differences in coding systems, a preliminary step using crosswalks that represent correspondence between systems may be required to link a study to a specific JEM (‘t Mannetje and Kromhout, 2003).

Occupational exposure measurements of a wide range of occupations have been collected in several national exposure databases during recent decades (Peters et al., 2012). Based on these data, exposure levels for all types of jobs and time periods could potentially be estimated by statistical modelling. Although several large occupational exposure databases exist, their use in population-based research has been limited due to a lack of FAIR (findable, accessible, interoperable, and re-useable) principles.

Although comparisons between individual JEMs have been published (e.g. Offermans et al., 2012), JEMs have not been systematically compiled and compared. Additionally, a clear overview of all available JEMs, exposure databases, job coding systems, and crosswalks has been lacking, and knowledge about these tools is not easily available. More accessible information on job coding systems and crosswalks between different systems will also support harmonization of exposure data and tools across countries.

Our aim was to collate regional (i.e. continental) and country-specific exposure assessment tools, with an initial focus on Europe, that can be applied to large general population cohorts to allow for risk analyses and to facilitate health impact analyses.

Methods

OMEGA-NET (the Network on the Coordination and Harmonisation of European Occupational Cohorts, http://omeganetcohorts.eu/) is an EU COST Action that started in 2017 and will continue into 2022 (Turner and Mehlum, 2018). OMEGA-NET includes members from over 40 countries, including European and neighbouring countries (Belarus, Morocco, Palestinian Authority, Russian Federation) as well as international partner countries Australia, the United States, and the United Arab Emirates. The project was set up to enable optimization of the use of industrial and general population cohorts across Europe to advance aetiological research (Turner and Mehlum, 2018).

Within the scope of OMEGA-NET, an online inventory was created to collect meta-data on various exposure assessment tools (https://occupationalexposuretools.net). Occupational health researchers were invited to provide details on JEMs, exposure databases, occupational coding systems, and the associated crosswalks, with a focus on Europe. The inventory was promoted, and contributions were sought, by mailings within consortia with a focus on occupational health research [OMEGA-NET, and the EU-H2020 Exposome Project for Health and Occupational Research (EPHOR) (Pronk et al., unpublished data) including representatives from 12 European countries], the newsletter of the International Commission on Occupational Health, an editorial in this journal (Peters et al., 2020), conference presentations, and by directly approaching individual researchers that were identified via searches in PubMed and Google.

Combining all meta-data, an open resource for occupational exposure assessment tools has been built. Data entries were checked for inconsistencies and clarifications were sought where necessary. Here, we describe the characteristics of the meta-data available as of August 2021. As we focus on tools that can be used in general population cohorts, we have excluded data that were provided on industry-specific JEMs (n = 3) for current descriptive analysis.

Results

Meta-data on 36 general population JEMs, 11 exposure databases, and 29 occupational coding systems have been collected from individual researchers up to August 2021.

A wide variety of exposures were covered, with dusts and fibres (in 14 JEMs) being the most common types (Table 1). Among dusts and fibres, asbestos was the most assessed exposure, with presence in 10 JEMs, followed by quartz and wood dust (both in 7 JEMs). Fewer JEMs covered organization of work including working time (5) or biological factors (4). Other exposures that were relatively often covered included benzene, chromium, nickel, and physical workload (each in seven JEMs). Many JEMs were originally developed for the Nordic countries, with FINJEM being the earliest JEM in our inventory (Kauppinen et al., 1998). There was also overlap in development of JEMs: FINJEM formed the basis for three later JEMs (i.e. NOCCA, INTEROCC, and MatEmEsp), whereas SYN-JEM used DOM-JEM as input. Furthermore, 27 JEMs were based on expert assessment, 17 included direct measurements, and 13 relied on self-reported data, with many reporting a combination of these sources. Five JEMs had an industry axis, 13 were time varying, and 7 were sex specific.

Table 1.

Overview of the 36 JEMs entered in the OMEGA-NET inventory of exposure assessment tools by August 2021.

JEM name	Job coding	Exposure metrics	Time period covered (time intervals)	Exposures	Data source(s)	Region for which the JEM was originally developed
ALOHA + JEM (Skorge et al., 2009)	ISCO 1988	Intensity: semi-quantitative	n.s.	Dusts and fibres: mineral dust; organic dust Solvents: chlorinated solvents; aromatic solvents; other solvents Metals: metals (n.s.) Pesticides: fungicides; herbicides; insecticides Other chemicals: gas and fumes (n.s.)	Expert assessment	Europe, North America
Asbestos JEMa,b (Swuste et al., 2008)	ISCO 1968	Intensity: semi-quantitative; probability	1945–1994 (5-year intervals)	Asbestos	Expert assessment Direct measurements	The Netherlands
AsbJEMa,b (van Oyen et al., 2015)	N/A	Intensity: quantitative	1943–present (1943–1966, 1967–1986, 1987–2003, 2004+)	Asbestos	Expert assessment Direct measurements	Australia
BEN-JEMb (Spycher et al., 2017)	ISCO 1998	Intensity: quantitative; probability	1945–2009 (1945–1959, 1960–1984, 1985–1994, 1 995–1997, 1998–2000, 2001–2003, 2004–2006, 2007–2009)	Benzene	Expert assessment Direct measurements	Europe, North America
CANJEMa,b (Sauvé et al., 2018)	ISCO 1968 SOC 2010 CITP 1968 NOC 2011 CCDO1971	Intensity: semi-quantitative; probability; frequency	1930–2000 (varying intervals)	CANJEM included 258 agents from the selected categories (not further specified in online inventory)	Expert assessment	North America
Constances JEM (Yung et al., 2020)	PCS	Intensity	n.s.	Awkward work postures; physical work load; repetitive work movements; sedentary work; standing work; work with video display units (VDU)	Expert assessment Self-reported data	France
COVID-19-JEM (Oude Hengel et al., 2021)	ISCO 2008	Probability	2020	Biological factors: infection risk (number of contacts; type of contacts; indirect contact; location; social distancing; face covering) Organization of work: job insecurity; migrants	Expert assessment	Denmark, The Netherlands, UK
dBAR-JEMb,c (Stokholm et al., 2020)	ISCO 1988	Intensity: quantitative	n.s.	Noise	Expert assessment Direct measurements	Denmark
DEE-JEM (Ge et al., 2020)	ISCO 1968	Intensity: quantitative; probability	n.s.	Diesel engine exhaust	Expert assessment Direct measurements	Europe, North America
DOM-JEM (Peters et al, 2011)	ISCO 1968	Intensity: semi-quantitative	n.s.	Dusts and fibres: asbestos; biological dust; quartz Metals: chromium; nickel Other chemicals: diesel engine exhaust; PAHs (n.s.) Biological factors: animal contact; endotoxins	Expert assessment	Europe
FINJEMb (Kauppinen et al., 1998)	ISCO 1958	Intensity: quantitative; probability	1945–1997 (1945–1959, 1960–1984, 1985–1994, 1995–1997)	Dusts and fibres: asbestos; man-made mineral fibres; inorganic dust (n.s.); quartz; animal dust; flour dust; plant dust; pulp or paper dust; synthetic polymer dust; textile dust; wood dust (hardwood); wood dust (softwood); wood dust (n.s.); leather dust Solvents: aliphatic and alicyclic hydrocarbon solvents (n.s.); benzene; styrene and styrene oxide; toluene; xylene; aromatic solvents (n.s.); methylene chloride; perchloroethylene; trichloroethanes; trichloroethylene; chlorinated hydrocarbon solvents (n.s.); formaldehyde; organic solvents (n.s.) Pesticides: fungicides; herbicides; insecticides Metals: arsenic; cadmium; chromium; iron; lead; nickel Other chemicals: carbon monoxide; diesel engine exhaust; gasoline engine exhaust; isocyanates; benzo(a)pyrene; bitumen fumes; oil mist; PAHs (n.s.); environmental tobacco smoke; sulphur dioxide and trioxide; welding fumes (n.s.) Biological factors: Gram-negative bacteria of human origin; moulds Physical agents: cold; hand-arm vibration; heat; noise; ionizing radiation; non-ionizing radiation; solar and ultraviolet radiation; ultrasound; noise; impulsiveness; hand vibration Ergonomics, physical workload, and injury related: accident risk Psychosocial domains: psychological job demands; social support at work from supervisors Organization of work: night (permanent or rotating)	Expert assessment Self-reported data Direct measurements	Finland
INTEROCC Chemical-JEM (van Tongeren et al., 2013)	ISCO 1968 ISCO 1988	Intensity: quantitative; probability	n.s.	Dusts and fibres: asbestos; quartz; animal dust; wood dust (n.s.) Solvents: gasoline; benzene; toluene; methylene chloride; perchloroethylene; trichloroethanes; trichloroethylene Metals: cadmium; chromium; iron; lead; nickel Other chemicals: diesel engine exhaust; benzo(a)pyrene; bitumen fumes; sulphur dioxide and compounds; welding fumes	Expert assessment FINJEM	International
INTEROCC ELF-JEM (Turner et al., 2014)	ISCO 1968 ISCO 1988	Intensity: quantitative	n.s.	Non-ionizing radiation	Expert assessment Direct measurements	Europe, North America
Lower Body JEM (Rubak et al., 2014)	ISCO 1988	Intensity: quantitative; probability; duration; frequency	n.s.	Awkward work postures; physical work load; standing work; whole-body vibration	Expert assessment	Denmark
LUXAR-JEMa (Vested et al., 2019)	ISCO 1988	Intensity: quantitative (lux), peaks	n.s.	Light at day	Expert assessment Direct measurements	Southern Scandinavia
MatEmESpb (García et al., 2013)	CNO-94	Intensity: qualitative, quantitative; probability; frequency; peaks	1996–2005	Dusts and fibres: asbestos; man-made mineral fibres; quartz; animal dust; flour dust; wood dust (n.s.) Solvents: gasoline; aliphatic and alicyclic hydrocarbon solvents (n.s.); benzene; aromatic hydrocarbon solvents (n.s.); methylene chloride; perchloroethylene; trichloroethanes; trichloroethylene; chlorinated hydrocarbon solvents (n.s.); formaldehyde; organic solvents (n.s.) Pesticides: thiram; captam; 2,4-D or 2,4,5-T; atrazine; diquat; diuron; chlorpyriphos; endosulfan; methomyl; pyrethrins Metals: arsenic; cadmium; chromium; iron; lead; nickel Other chemicals: benzo(a)pyrene; bituminous fumes; PAHs (n.s.); oil mist; sulphur dioxide; isocyanates; welding fumes (n.s.) Physical agents: heat; noise Ergonomics, physical workload, and injury related: awkward work postures; physical work load; repetitive work movements; sedentary work; standing work; work with video display units (VDU); vibrations; safety hazards Psychosocial domains: violence; job control, autonomy; psychological job demands; role conflict/ambiguity/clarity; social support at work from supervisors; skill use opportunities; work engagement; job insecurity; esteem; sociodemographic characteristics of working force Organization of work: contract duration; job insecurity; low pay; work contract type; night (permanent or rotating); duration; regular/variable working hours; shift work; working weekends; employment situation	Expert assessment Self-reported data FINJEM	Spain
Matgénéa,b (Marant Micallef et al., 2021)	ISCO 1968 PCS 1994	Intensity: semi-quantitative; probability	1950–2010 (varying intervals)	Dusts and fibres: asbestos; ceramic fibres; mineral wools; cement; quartz; flour dust; leather dust Solvents: aliphatic and alicyclic hydrocarbon solvents (n.s.); aromatic solvents (n.s.); benzene; gasoline; methylene chloride; perchloroethylene; trichloroethylene; carbon tetrachloride; chloroform; formaldehyde; ketones; ethers; alcohols; ethylene and propylene glycols Other chemicals: PAHs	Expert assessment	France
NOCCA-JEMb (Kauppinen et al., 2009)	NYK	Intensity: semi-quantity; probability	1945–1994 (1945–1959, 1960–1974, 1975–1984, 1985–1994)	Dusts and fibres: asbestos; quartz; animal dust; wood dust (n.s.) Solvents: gasoline; benzene; toluene; chloroform; methylene chloride; perchloroethylene; trichloroethanes; trichloroethylene; formaldehyde Metals: chromium; iron; lead; nickel Other chemicals: diesel engine exhaust; gasoline engine exhaust; benzo(a)pyrene; bitumen fumes; sulphur dioxide and compounds; welding fumes Physical agents: light at night; ionizing radiation; solar and UV radiation Ergonomics, physical workload, and injury related: physical work load Organization of work: night (permanent or rotating)	Expert assessment Direct measurements FINJEM	Nordic countries
NORJEM—mechanicalc (Hanvold et al., 2019)	ISCO 1988 STYRK-98	Intensity: qualitative	n.s.	Awkward work postures; physical work load; standing work; hands above shoulder height; standing/walking	Self-reported data	Norway
NORJEM—psychosocialc (Hanvold et al., 2019)	ISCO 1988 STYRK-98	Intensity	n.s.	Job control, autonomy; psychosocial job demands; social support at work from supervisors; skill use opportunities; monotonous work; job strain	Self-reported data	Norway
OAsJEM (Le Moual et al., 2018)	ISCO 1988	Intensity: semi-quantitative	n.s.	Dusts and fibres: animal dust; flour dust; plant dust; textile dust; wood dust (n.s.) Solvents: organic solvents Pesticides: fungicides; herbicides Metals: metal (n.s.) Other chemicals: detergents and cleaning products; isocyanates	Expert assessment Literature data	France
Physical workload factors JEMc (Solovieva et al., 2012)	ISCO 1988	Probability, duration, frequency	n.s.	Awkward work postures; physical work load; repetitive work movements; standing work	Self-reported data	Finland
POLLEK (Szemik et al., 2020)	N/A	Intensity, probability, duration, frequency	n.s.	Career advancements opportunities; psychological job demands; work–family interface	Self-reported data Direct measurements	Poland
Psychosocial JEMc (Solovieva et al., 2012)	ISCO 1988	Intensity: quantitative	n.s.	Job control, autonomy; psychosocial job demands; social support at work from supervisors; skill use opportunities	Self-reported data	Finland
RF-JEM (Migault et al., 2019)	ISCO 1988	Intensity: quantitative, probability	n.s.	Non-ionizing radiation	Expert assessment Self-reported data Direct measurements Literature data	Europe, North America, Oceania
Shiftwork JEM (Fernandez et al., 2014)	ISCO 1968	Probability	n.s.	Exposure to light at night; phase shift; sleep disruption; poor diet; lack of physical activity; lack of vitamin D; graveyard shifts; early morning shifts	Expert assessment Self-reported data	Australia
SHOCK-JEM (Huss et al., 2013)	ISCO 1988	Intensity	n.s.	Electric shock	Expert assessment Direct measurements	Europe, North America
Shoulder JEM (Dalbøge et al., 2016)	ISCO 1988	Intensity, duration, frequency	n.s.	Awkward work postures; physical work load; repetitive work movements; hand-arm vibration; computer work	Expert assessment Direct measurements	Denmark
SIOPS-JEM (Behrens et al., 2016)	ISCO 1968 ISCO 1988	Intensity: quantitative	n.s.	Social prestige	Expert assessment	Europe, North America
Swedish noise JEMb (Sjöström et al., 2013)	ISCO 1958 ISCO 1988 ISCO 2008 NYK SSYK 96	Intensity	1970–2004 (5-year intervals)	Noise	Expert assessment Direct measurements	Sweden
Swedish physical workload JEMb,c,d	ISCO 1988 SSYK 96	Duration	1989–2013 (1989–1997, 1997–2013)	Heavy lifting (at least 15 kg); physically strenuous work; fast breathing; forward bent position; twisted position; hands above shoulder level; repetitive work; frequent bending or twisting; physical load index	Self-reported data	Sweden
Swedish psychosocial JEMb,c,d	ISCO 1988 SSYK 96	Duration	1989–2013 (1989–1997, 1997–2013)	Job control, autonomy; psychosocial job demands; social support at work from supervisors; skill use opportunities; job strain	Self-reported data	Sweden
SWEJEM Chemicals and Particlesd	FOB 80 SSYK 96	Intensity: quantitative (mg/m3); probability	n.s.	Dusts and fibres: asbestos; man-made mineral fibres; quartz; stone and concrete; animal dust; flour dust; plant dust; pulp or paper dust; synthetic polymer dust; textile dust; wood dust (hardwood); wood dust (softwood); wood dust (n.s.); leather dust Solvents: gasoline; aliphatic and alicyclic hydrocarbon solvents (n.s.); benzene; styrene and styrene oxide; toluene; aromatic solvents (n.s.); methylene chloride; perchloroethylene; trichloroethylene; chlorinated hydrocarbon solvents (n.s.); formaldehyde Pesticides: fungicides; herbicides; insecticides Metals: arsenic; cadmium; chromium; iron; lead; nickel Other chemicals: carbon monoxide; detergents; diesel engine exhaust; gasoline engine exhaust; isocyanates; synthetic metal processing or drilling oils or fluids; benzo(a)pyrene; bitumen fumes; PAHs (n.s.); sulphur dioxide and trioxide; welding fumes (n.s.)	Expert assessment Direct measurements FIN-JEM	Sweden
SYN-JEMb (Peters et al., 2016)	ISCO 1968	Intensity: quantitative	1960–2010 (1-year intervals)	Dusts and fibres: asbestos; quartz Metals: chromium VI; nickel Other chemicals: benzo(a)pyrene	Direct measurements DOM-JEM	Europe, Canada
US Pesticide JEM (Liew et al., 2014)	IPUM-USA 2000	Intensity: semi-quantitative	n.s.	Pesticides (n.s.)	Self-reported data	North America
Wood dust JEMb (Basinas et al., 2016)	ISCO 1988	Intensity: quantitative	1978–2004 (1-year intervals)	Wood dust (n.s.)	Expert assessment Direct measurements	Europe

ns, not specified; PAH, polycyclic aromatic hydrocarbons.

^aIndustry axis.

^bTime varying.

^cSex specific.

^dNo scientific publications identified.

Meta-data were provided for exposure databases covering the Netherlands (n = 3), France (n = 3), UK (n = 2), Norway (n = 1), and multinational (n = 2) (Supplementary Table 1, available at Annals of Work Exposures and Health online). The earliest data are included in ExpoSYN (1951–2009). For six databases (Colchic, EV@LUTIL, EXPO, HSE-BMDB, NECID, and SCOLA), data collection is still ongoing. Dusts and fibres were also the most common exposures in the databases (7 out of 11), followed by solvents and pesticides (both included in 6 databases).

Information on the occupational coding systems included international (i.e. ISCO), as well as national coding systems from more than 10 countries. Their relation to other coding systems and the availability of crosswalks and/or automated coding systems is shown in Table 2.

Table 2.

Meta-data on the 29 occupational coding systems in OMEGA-NET inventory by August 2021.

System name	Version year	Country/region	Related coding system	Crosswalk available to related system/version	Semi-automated coding
CH-ISCO-19	2019	Switzerland	SSCO 2000	No
CITP-08	2008	France	ISCO-08	Yes
CNO-94	1994	Spain	ISCO-88	Yes
CNO-11	2011	Spain	ISCO-08	Yes
DISCO-88	1996	Denmark	ISCO-88	Yes
DISCO-08	2010	Denmark	ISCO-08	Yes
FOB 80	1980	Sweden	ISCO-58	Yes
ISCO-58	1958	International		Yes
ISCO-68	1968	International		Yes	CAPS
ISCO-88	1988	International		Yes	CAPS
ISCO-08	2008	International		Yes	CAPS
NOC 2006	2006	Canada		Yes
NOC 2011	2011	Canada		Yes	CAPS
NOC 2016	2016	Canada		Yes
NUP06	2006	Italy	ISCO-88	Yes
NYK83	1983	Sweden	FOB80	Yes
PCS	2003	France		No	SICORE
SBC 1992	1992	Netherlands	ISCO-88	Yes
SSCO 2000	2000	Switzerland	CH-ISCO-19	No
SSYK 1996	1996	Sweden	ISCO-88	Yes
SSYK 2012	2014	Sweden		Yes
STYRK-08	2011	Norway	ISCO-08	Yes
UK SOC 1990	1990	United Kingdom		Yes	CASCOT
UK SOC 2000	2000	United Kingdom		Yes	OSCAR
UK SOC 2010	2010	United Kingdom		Yes	CASCOT
UK SOC 2020	2020	United Kingdom		Yes	CASCOT
US SOC 2000	2000	United States		Yes
US SOC 2010	2010	United States		Yes	SOCcer
US SOC 2018	2018	United States		Yes	O*Net

CAPS: https://ssl3.isped.u-bordeaux2.fr/CAPS-CA/Langue.aspx; CASCOT: https://warwick.ac.uk/fac/soc/ier/software/cascot/; O*Net: https://www.onetonline.org/; OSCAR: https://pubmed.ncbi.nlm.nih.gov/27973677/; SICORE: https://www.census.gov/prod/2/gen/96arc/ixbschuh.pdf; SOCcer: https://doi.org/10.1136/oemed-2015-103152.

Discussion

Existing occupational exposure assessment tools, including JEMs, exposure databases, coding systems, and crosswalks have been collated in an inventory. Although many different types of exposures have been covered by the 36 JEMs, the most common exposure group was dusts and fibres, while biological factors and employment conditions were much less frequent. This distribution may partly represent the major research focus in occupational epidemiology over the last decades. On the other hand, not all exposure types are equally appropriate for assessment by JEMs (Peters, 2020), which may also be reflected by our inventory. The availability of multiple JEMs on the same exposure allows for studying method uncertainty and to study if associations are method dependent (e.g. Offermans et al., 2012). The geographical coverage showed that most JEMs were developed in Western and Northern Europe. Based on the current inventory, it would appear that JEMs developed for Eastern and Southern Europe, in particular, could be a major improvement on the current toolbox for occupational exposure assessment in large cohorts.

While the OMEGA-NET team actively made contacts and sought contributions from researchers, our inventory was largely dependent on the person(s) responsible for each tool to enter meta-data in the online system. This approach ensured the relevant information was collected as accurately as possible. Particularly for older exposure assessment tools, institutional knowledge may be lost if the responsible persons are no longer active in the research area. We, therefore, focussed on more recent and currently cited exposure tools, which we also considered to be most relevant. The downside of this dependency on individual researchers was that not all identified exposure tools have been included. For example, meta-data on a major national exposure database [i.e. MEGA from Germany (Gabriel et al., 2010)] were not entered by August 2021.

We further focussed our efforts on collecting information on JEMs and databases that are active and could potentially be used in the exposure assessment of general population cohorts. Hence, tools that were highly specific for one type of occupation or one study population were not our main priority. For example, although we know there are many exposure measurements collected in specific industries (Peters et al., 2012), we did not actively approach their database custodians, as many such databases are not available for use outside their intended scope [e.g. the Dust Monitoring Program of the European Industrial Minerals Association (Zilaout et al., 2017)]. There were also JEMs developed for one specific population, e.g. the Matex-JEM that was specifically developed for one company, using its internal job classification, and as such is not applicable to other settings (Imbernon et al., 1991).

Furthermore, we initially focussed on European tools, fitting with the initial objectives of OMEGA-NET. However, the inventory and its website remain open for new entries and a more global coverage would certainly be preferable to support the broader objectives to promote collaborative and harmonized research in the area of occupational epidemiology.

To have an easy entry point into finding these important exposure tools was one of the goals of OMEGA-NET. Therefore, all collected meta-data on exposure assessment tools have been made publicly available via a searchable web-based database (https://occupationalexposuretools.net/inventory/). With this effort we have brought together a wealth of information on available exposure assessment tools, that will aid the exposure assessment process in many occupational cohorts.

Funding

This publication is based upon work from COST Action CA16216 (OMEGA-NET), supported by COST (European Cooperation in Science and Technology). MCT is funded by a Ramón y Cajal fellowship (RYC-2017-01892) from the Spanish Ministry of Science, Innovation and Universities and co-funded by the European Social Fund. ISGlobal acknowledges support from the Spanish Ministry of Science and Innovation through the ‘Centro de Excelencia Severo Ochoa 2019-2023’ Program (CEX2018-000806-S), and support from the Generalitat de Catalunya through the CERCA Program.

Conflict of interest

None declared.

Data availability

The data were derived from sources in the public domain: https://occupationalexposuretools.net/inventory/.