Article Text

Risk of opioid-related harms by occupation within a large cohort of formerly injured workers in Ontario, Canada: findings from the Occupational Disease Surveillance System
  1. Nancy Carnide1,2,
  2. Jeavana Sritharan2,3,
  3. Chaojie Song3,
  4. Fateme Kooshki1,2,
  5. Paul A Demers2,3
  1. 1 Institute for Work & Health, Toronto, Ontario, Canada
  2. 2 Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
  3. 3 Occupational Cancer Research Centre, Ontario Health, Toronto, Ontario, Canada
  1. Correspondence to Dr Nancy Carnide; ncarnide{at}iwh.on.ca

Abstract

Objective Working-age individuals have been disproportionately affected by the opioid crisis, prompting interest in the potential role of occupation as a contributor. This study aimed to estimate the risk of opioid-related poisonings and mental and behavioural disorders by occupation and industry within a cohort of 1.7 million formerly injured workers.

Methods Workers were identified in the Occupational Disease Surveillance System, a system linking workers’ compensation data (1983–2019) to emergency department and hospitalisation records (2006–2020) in Ontario, Canada. Cox proportional hazards models were used to estimate HRs and 95% CIs for hospital encounters for opioid-related poisonings and mental and behavioural disorders by occupation and industry compared with all other workers, adjusted for age, sex and birth year.

Results In total, 13 702 opioid-related poisoning (p) events (n=10 064 workers) and 19 629 opioid-related mental and behavioural (mb) disorder events (n=11 755 workers) were observed. Elevated risks were identified among workers in forestry and logging (HRp=1.45, 95% CI 1.09 to 1.94; HRmb=1.70, 95% CI 1.34 to 2.16); processing (minerals, metals, clay, chemical) (HRp=1.27, 95% CI 1.14 to 1.42; HRmb=1.26, 95% CI 1.14 to 1.39); processing (food, wood, textile) (HRp=1.12, 95% CI 1.01 to 1.24; HRmb=1.19, 95% CI 1.09 to 1.31); machining (HRp=1.13, 95% CI 1.04 to 1.21; HRmb=1.17, 95% CI 1.09 to 1.25); construction trades (HRp=1.57, 95% CI 1.48 to 1.67; HRmb=1.59, 95% CI 1.51 to 1.68); materials handling (HRp=1.32, 95% CI 1.22 to 1.43; HRmb=1.22, 95% CI 1.13 to 1.31); mining and quarrying (HRmb=1.68, 95% CI 1.34 to 2.11); and transport equipment operating occupations (HRp=1.18, 95% CI 1.09 to 1.27). Elevated risks were observed among select workers in service, sales, clerical and health. Findings by industry were similar.

Conclusions Results provide additional evidence that opioid-related harms cluster among certain occupational groups. Findings can be used to strategically target prevention and harm reduction activities in the workplace.

  • Mental Health
  • Public health
  • Public Health Surveillance

Data availability statement

Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • Prior studies demonstrating an occupational pattern in fatal opioid-related poisonings have primarily conducted retrospective case-based assessments of occupation obtained from death data, lacked detailed information on occupation and were mainly conducted in the USA.

WHAT THIS STUDY ADDS

  • This study of 1.7 million formerly injured workers in Ontario, Canada, found workers in construction, forestry, mining and manufacturing-related occupations demonstrated elevated risks of opioid-related harms (poisonings and mental and behavioural disorders).

  • Increased risks were also seen among select workers in other occupational groups, including water transport and trucking, service station attendants, guards and watchmen, janitors and cleaners, nursing aides, orderlies, tellers and cashiers, and occupations in lodging, accommodations and food preparation.

  • Heterogeneity in risk was seen within certain occupational groups, including construction, where workers employed in electrical trades demonstrated reduced risks, while workers in excavating, paving and grading, as well as other trades (eg, roofers, painters, brick and stone masons), had higher risks.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • While more research is needed to identify the specific workplace risk factors for opioid-related harms, findings from this study can be used to strategically target prevention and harm reduction activities within the workplace to mitigate those harms in the workforce.

Introduction

North America continues to face a public health crisis characterised by unprecedented opioid-related deaths. From 2016 to 2022, the rate of opioid-related deaths increased from 7.8 to 19.4 per 100 000 in Canada, resulting in almost 37 000 deaths, and from 13.1 to 25.0 per 100 000 in the USA, leading to over 417 000 fatalities.1 2 In both countries, substance-related fatalities have become the leading cause of accidental death.1 3 Life expectancy at birth has also stagnated in North America, attributable in part to opioid-related deaths occurring among younger age groups.3 4 Indeed, individuals of working age, particularly men, have been disproportionately affected by the opioid crisis,1 2 prompting interest in understanding the potential role of occupation as a contributing factor.

Several US studies have demonstrated an occupational pattern in fatal opioid-related poisonings (often referred to as overdoses). Across these death certificate studies, some of the highest rates of death have occurred among individuals working in construction; natural resources (including extraction, agriculture, fishing, forestry and hunting); transportation, warehousing and material moving; installation, maintenance and repair; production; accommodation and food services; and healthcare support.5–9 By contrast, there exists a notable gap in information on opioid-related harms among the working population in Canada. In British Columbia, from 2014 to 2016, the most common industries of employment in the year before an opioid-related death for individuals with known employment status were construction; administrative and support, waste management and remediation services; accommodation and food services; retail trade; manufacturing; transportation and warehousing; and healthcare and social assistance.10 In Ontario, almost a third of all opioid-related deaths among employed decedents from 2019 to 2020 occurred among construction workers.11

These data suggest there may be certain groups of workers at a higher risk of experiencing opioid-related harms. However, these prior studies took a case-based approach, examining occupational information from death certificates or documents pertaining to coroner investigations, rather than an epidemiological risk assessment approach. Further, they frequently lacked data on detailed occupational groups.

The Occupational Disease Surveillance System (ODSS) in Ontario, Canada, was created by linking provincial health data with job information from workers’ compensation and has been instrumental in filling gaps in knowledge regarding occupational patterns in various cancers and non-malignant diseases among a large cohort of workers.12–14 Leveraging the ODSS, the objectives of this study were to estimate the risk of hospital encounters for two opioid-related harms (poisonings and mental and behavioural (MB) disorders) by occupation and industry occurring among 1.7 million formerly injured workers between 2006 and 2020.

Methods

Study cohort and data sources

The ODSS was created using accepted lost-time workers’ compensation claims records provided by the Ontario Workplace Safety and Insurance Board (WSIB) from 1983 to 2019 (n=2 368 218). The WSIB is the provincial workers’ compensation insurer in Ontario, Canada, providing coverage to approximately 70–75% of the Ontario workforce.

Workers were matched to records in the Ontario Health Insurance Plan’s Registered Persons Database (RPDB) through deterministic linkage methods, matching on exact criteria and specific personal identifiers (ie, full name, date of birth, sex), and probabilistic linkage methods, matching based on the likelihood that records correspond to the same worker (ie, partial names or incomplete information). The RPDB includes basic demographic information and a unique Health Insurance Number (HIN) for all Ontarians registered for provincial health insurance. A total of 1 973 312 workers were successfully matched to an RPDB record with a corresponding HIN. Those without a HIN were excluded from analysis (n=394 906). Using the HIN, workers were then deterministically linked to hospitalisation records in the Discharge Abstract Database (DAD) and emergency department records in the National Ambulatory Care Reporting System (NACRS) (2006–2020). A total of 283 677 workers were excluded if they were outside of working age (15–65) or had died or emigrated out of Ontario during follow-up, leaving 1 689 635 workers for this analysis. Details of the ODSS linkage and methods have been previously described.14 15

Study variables

Independent variables

Occupation and industry data for each worker were obtained from WSIB records, representing their occupational information at the time they experienced their work-related injury or illness. Occupation was coded using the 1971 Canadian Classification Dictionary of Occupation,16 and industry was coded using the 1970 and 1980 Canadian Standard Industrial Classification.17 18 Using these coding systems, occupations and industries were classified into three levels: divisions, major groups and minor groups. Divisions represent the broadest level classification of an occupation (22 divisions) or industry (10 divisions). Major and minor groups represent a more detailed classification of occupation or industry.

Outcomes

Two opioid-related harms were examined: poisonings and MB disorders. Hospitalisations and emergency department visits for each outcome were identified in the DAD and NACRS, respectively, using diagnoses coded according to the International Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA) (opioid-related poisonings: T40.0–T40.4, T40.6; opioid-related MB disorders: F11.0–F11.9; see online supplemental table 1 for a detailed description). These case definitions have been used previously.19 Note these codes do not distinguish between fatal and non-fatal outcomes.

Supplemental material

Opioid poisonings were also differentiated by intent using ICD-10-CA diagnostic codes: X42, accidental poisonings; X62, intentional poisonings; and Y12, unknown poisonings.

Analysis

Workers were followed from 2006 or their first WSIB accepted claim to the earliest of the date of first diagnosis, age 65 years, death, emigration out of Ontario or end of study follow-up (31 December 2020). Information on death status and emigration was obtained from the RPDB. Follow-up for each opioid-related harm was conducted separately.

Cox proportional hazards models were used to estimate HRs and 95% CIs of the risk of each of the two opioid-related harms for each occupation and industry, compared with all other workers in the ODSS. All models were adjusted for birth year (continuous), sex and age at start of follow-up (continuous). The pattern of findings from crude models was similar (available on request). In a set of sensitivity analyses, all models were run among a subset of the cohort with claim years paralleling the start of the follow-up period (2006 onward, n=645 111 for poisonings, n=644 987 for MB disorders). All analyses were performed using SAS V.9.4 (SAS Institute).

Results

The cohort was 66.3% male, with an average age of 44.0 years (SD 11.5). From 1 January 2006 to 31 December 2020, a total of 13 702 opioid-related poisoning events were observed among 10 064 workers (18 738 876.4 person-years), 72.1% of which were male. Just over half (53.8%) of the poisoning events were accidental, 25.1% were intentional and 21.1% were either coded as being of unknown intent or were missing information on intent (online supplemental table 2). The most common types of opioids involved in the poisoning event were other opioids (54.4%; eg, codeine, morphine, hydromorphone, oxycodone), other synthetic narcotics (19.5%; eg, fentanyl, tramadol) and opiates not elsewhere classified (15.3%) (online supplemental table 2).

Similarly, 19 629 opioid-related MB disorder events were observed among 11 755 workers (18 741 612.2 person-years), with 78.2% occurring among men. The most common diagnoses included withdrawal state (41.9%), dependence syndrome (28.4%) and harmful use (25.0%) (online supplemental table 3).

Risk of opioid-related harms by occupation

The HRs and corresponding 95% CIs for each opioid-related harm by division-level occupation are presented in table 1. Workers employed in forestry and logging; processing (minerals, metals, clay, chemical); processing (food, wood, textile); machining and related; construction trades; and materials handling and related occupations were observed to have elevated risks of both types of opioid-related harms, while elevated risks of poisonings only were found among workers in transport equipment operating, and increased risk of MB disorders was found among workers in mining. Decreased risks of one or both types of harm were observed among workers in managerial, administrative and related; natural sciences, engineering and mathematics; teaching and related; product fabricating, assembling and repairing; artistic, literary, recreational and related; and clerical and related occupations.

Table 1

HRs and 95% CIs for opioid-related poisonings and mental and behavioural disorders by division-level occupation group in the ODSS worker cohort (2006–2020)

Table 2 presents the results for select occupations at the major and minor group levels, where the pattern of findings within a particular division group was mixed or conflicted with the division-level results. Findings for all major-level and minor-level occupation groups can be found in online supplemental table 4.

Table 2

HRs and 95% CIs for opioid-related poisonings and mental and behavioural disorders for select major-level and minor-level occupation groups in the ODSS worker cohort (2006–2020)

Although occupations in medicine and health demonstrated null findings at the division level, results among nursing-related occupations at the minor level were mixed. An elevated risk of both harms was seen among nursing aides and orderlies and nursing assistants (non-significant), while reduced risks were found among nurses (registered, graduate and nurses in training), as well as nursing, therapy and related assisting occupations, not elsewhere classified.

Within clerical and related occupations, most major-level and minor-level groups had null findings, with a few exceptions at the minor level: tellers and cashiers had an increased risk of poisonings, while mail carriers and messengers were at a decreased risk of MB disorders. Similarly, while most sales occupations demonstrated null or protective effects, service station attendants had a significantly increased risk of opioid-related poisonings.

Variability in findings at the major and minor levels was observed among service occupations. Protective service occupations (namely firefighting and policemen and detectives) were at a reduced risk of both types of harm. However, guards and watchmen within this group had an elevated risk of opioid-related poisoning. Most food and beverage preparation and related service occupations had null findings with some exceptions: reduced risks for both harms were seen among supervisors, while chefs and cooks and waiters, hostesses and stewards had positive associations with opioid-related poisonings. Occupations in lodging and other accommodation, personal service, apparel and furnishing service, and other service (namely janitors, charworkers and cleaners) were all associated with elevated risks of one or both types of harm.

Within farming, horticulture and animal husbandry, farm workers had a reduced risk of opioid-related MB disorders, while nursery and related workers were at an elevated risk of both harms. The increased risks seen at the division level for machining and related occupations were driven by findings among workers in metal shaping and forming occupations, including sheet metal workers, metalworking machine operators, and welding and flame cutting occupations. On the other hand, metal machining occupations, namely tool and die making and machinist and machine tool setting-up occupations, showed decreased risks of both poisonings and MB disorders.

Product fabricating, assembling and repairing occupations also demonstrated mixed findings at the major and minor levels. Fabricating and assembling occupations of metal products, wood products not elsewhere classified and other products were associated with greater risk of one or both harms. Reduced risks (some non-statistically significant) were seen among fabricating, assembling and repairing occupations of electrical, electronic and related equipment and of textile, fur and leather products, as well as among mechanics and repairers (except electrical) and cabinet and wood furniture makers.

Most construction trade occupations in excavating, grading, paving and related, as well as other construction trades (eg, carpenters; brick and stone masons and tile setters; roofing, waterproofing and related), demonstrated elevated risks of both harms, with some exceptions (eg, pipefitting, plumbing and related occupations). On the other hand, electrical power, lighting, and wire communications equipment, erecting, installing and repairing occupations were at a reduced risk of both harms. Similarly, most transport equipment operating occupations had increased risks for both types of harms, including water transport and motor transport occupations (mainly truck drivers), while air transport occupations had reduced risks of both poisonings and MB disorders.

Risk of opioid-related harms by industry

Adjusted HRs and corresponding 95% CIs for each opioid-related harm by division-level industry group are shown in table 3. Similar to the occupation findings, elevated risks were observed for workers in forestry, fishing and trapping; mines, quarries and oil wells; community, business and personal service; construction; and manufacturing. All industry results at the major and minor group levels are shown in online supplemental table 5. The pattern of results was similar to that observed by occupation.

Table 3

HRs and 95% CIs for opioid-related poisonings and mental and behavioural disorders by division-level industry group in the ODSS worker cohort (2006–2020)

Sensitivity analyses

Findings from analyses repeated using a subcohort of more recent claims (2006 onward) were similar to the main analysis, although due to the smaller sample size and number of available cases in this smaller subset (n=3172 poisonings, n=3567 MB disorders), some estimates were no longer statistically significant. Details are available on request.

Discussion

Little is known about the role of work in the opioid crisis, particularly in Canada. We aimed to fill this knowledge gap by estimating the risks of opioid-related poisonings and MB disorders by occupation and industry in a large cohort of formerly injured workers in Ontario, Canada. Our findings confirm that distinct occupational patterns exist among this group of workers experiencing opioid-related harms.

In line with findings from US studies,5–9 the most substantial elevated risks at the division level were observed among formerly injured workers in construction, forestry and mining. Workers in manufacturing-related occupations, namely processing, machining, materials handling and some fabricating and assembling occupations, also demonstrated statistically significant increases in risk. Although the reasons for these patterns are not well understood, researchers have previously identified the physical demands required in these occupations to be a common thread. As a result of the heavy lifting, repetitive movements and awkward postures involved in these jobs, workers experience high rates of injury and chronic pain.6 20–22 Studies have also shown work-related injuries often lead to opioid prescribing,23 24 and occupational patterns in opioid prescribing consistent with occupations at higher risk of opioid-related harms in this and other studies.25 Although various policy interventions have led to declines in overall opioid prescribing among injured workers,26 27 recent data suggest a substantial proportion of injured workers may still be receiving prescriptions in some jurisdictions.28

The inherent nature of many of these occupations may also increase the likelihood of workers falling into problematic opioid use patterns after an injury. Pressure to return to and stay at work arising from job insecurity, non-standard employment arrangements and limited sick leave may compel workers to work despite residual pain.6 20 22 Suitable work accommodations that reduce exposure to ergonomic hazards may also be limited in these occupations.20 Furthermore, normative expectations of masculinity frequently documented in these occupational groups may lead to a normalisation of injuries and promote a culture of working through pain and stigmatise help seeking for both men and women.29 Indeed, the emerging evidence supports a role for work-related pain and injuries in the development of opioid-related harms, including in our own analysis in which we found workers in the ODSS had a greater risk of opioid-related harms compared with the general working-age population.15 30–32 In addition to injuries, workplace psychosocial risk factors have been hypothesised to contribute to opioid-related harms in these occupations.20 Although data are limited, effort–reward imbalance, low skill discretion, job strain and psychological demands have been shown to be associated with opioid misuse.33 34

Workers in construction and trades have received considerable attention in this crisis. Results of the current analysis, however, suggest that heterogeneity in risk may exist within this group. Formerly injured workers in excavating, paving and grading, as well as other trades (eg, roofers, painters, brick and stone masons), had higher risks of both harms. In contrast, workers in electrical power, lighting and wire communications occupations demonstrated reduced risks. The reasons behind these divergent results are unclear but may reflect differences in the extent of physical demands and job precarity across these occupations. It may also suggest a broader role for social inequities, with workers in the latter group being more highly skilled due to greater training requirements, a potential proxy for higher socioeconomic status. This may also explain the results seen for occupations within machining, whereby metal shaping and forming occupations (eg, sheet metal workers, welding and flame cutting occupations) had elevated risks, while tool and die makers and machinists were found to have a reduced risk.

Similar to other studies,5–9 elevated risks were observed among select formerly injured workers in other occupations, including transportation, service, health, clerical and sales. Within transportation, air transport workers demonstrated reduced risks of both harms, which may reflect the extensive medical and fitness-for-duty evaluations required of this group. On the other hand, workers in water transport and trucking demonstrated elevated risks. The ergonomic as well as psychosocial risk factors of these jobs, including long work hours, work away from home, time demands and social isolation, may be contributors. The higher risks seen among service station attendants, guards and watchmen, janitors and cleaners, nursing aides, orderlies, tellers and cashiers, and occupations in lodging, accommodations and food preparation likely share some of these risk factors, in addition to non-standard work arrangements, job insecurity, limited sick leave and lack of health insurance benefits.

Only one other study, to our knowledge, has examined substance-related harms by occupation and industry at a similar detailed level, although data were not specific to opioid involvement in deaths.9 While findings were similar, notable differences existed. In contrast to our findings, elevations in risk were not seen among healthcare support workers, security guards, janitors or among workers in lodging and accommodations, while increased risks among electricians and mechanics were evident. The reasons for these discrepant findings are unclear, though may reflect regional differences.

More research is needed to better understand the workplace factors contributing to greater risks of opioid-related harms. Nonetheless, the data provide an important signal for which subgroups of the workforce are at highest risk. Findings underscore the opportunity to leverage workplaces as a unique setting to mitigate opioid-related morbidity and mortality across the spectrum of substance use.6 20 35–37 Primary prevention of opioid use disorder in the workplace may focus on increasing education and awareness of opioid use and harms among workers, reducing the physical and psychosocial risk factors for pain, injury and poor mental health in the workplace, and making supports available to assist workers who are in pain, including workplace accommodations, sufficient sick leave and health insurance options that include non-pharmaceutical pain management alternatives.6 20 35–37 Workplaces can support workers with substance use disorders by facilitating access to treatment supports, as well as by providing them with workplace accommodations to assist them in returning to and remaining at work.20 36 37 Workplaces can also cultivate a supportive culture that allows workers the freedom to disclose their experiences with pain or substance use without fear of reprisal or stigma.20 35–37 Recent evaluations of opioid awareness and prevention training initiatives have shown positive results.38 39 There may also be a role for harm reduction in the workplace. Although most poisonings occur at home, US data suggest workplace poisonings are on the rise.40 Naloxone availability and overdose prevention training in workplaces may help reduce fatal outcomes from poisoning events that occur in the workplace. Ontario recently became the first jurisdiction in North America to introduce legislation mandating naloxone in workplaces at a higher risk of poisonings.

A key strength of this analysis was the use of data from a large cohort of 1.7 million workers, providing the statistical power to examine risks of opioid-related harm by detailed occupational and industry groups. This analysis is also a novel addition to the extant literature on occupational patterns in opioid-related harms, from a Canadian perspective, due to the longitudinal methodology that goes beyond retrospective assessments of case-based coroner data, and the consideration of cases in emergency department and hospitalisation records. Our study also examines risk of opioid-related MB disorders, an outcome that is comparatively less often studied in the context of occupation.

There are also limitations. The ODSS cohort used in this analysis includes only workers who had an accepted lost-time workers’ compensation claim for a work-related injury or illness. As such, workers in occupations and industries with a high risk of injury are over-represented, and results may not be as generalisable to those in management, business and administrative occupations. The selection of workers into the cohort based on having an injury, which is likely a risk factor for opioid use and related harms, also limits generalisability, such that the results for specific occupations or occupational groups may not be representative of the risk to all people employed in these groups. Restricting the sample only to workers with work-related injuries may also have introduced a selection bias due to collider stratification bias. Given injury is influenced by the exposure (occupation/industry) and there may be unmeasured characteristics associated with both injury and opioid-related harms (eg, risk-taking behaviour), there may have been spurious and/or overestimated associations between occupation/industry and opioid-related harms. Unfortunately, we are not able to determine the risk of opioid-related harms among workers who had not experienced an injury. However, it is important to recognise that some of the risk for opioid-related harms may be due to unmeasured factors beyond workplace injuries. Further, occupations at high risk of injury, which the ODSS does capture, may also be more likely to use opioids for pain, and an in-depth understanding of these higher risk groups may be needed to formulate effective prevention and harm reduction strategies. Data on occupation and industry were recorded only at the time of the workers’ compensation claim, and non-differential misclassification in these exposure measures was possible if workers changed jobs. Further, we lacked data on employment status at the time of harm. Nevertheless, it is important to note that results were similar when limiting the cohort to more recent claims to coincide with the follow-up period.

The administrative health data used to identify cases of opioid-related harms do not identify whether the outcome was fatal and only include individuals who presented to hospital, thus underestimating the extent of opioid-related harm in this cohort. Our team is currently working to incorporate provincial mortality data into the ODSS in future. Residual confounding may also be present, as only a limited set of variables were controlled for in the analysis.

Results of this study of formerly injured workers provide additional evidence that opioid-related harms cluster among certain occupational groups. While more research is needed to elucidate the specific workplace risk factors for opioid-related harms, findings from the current analysis can be used to strategically target prevention and harm reduction activities within the workplace to mitigate those harms and enhance current public health efforts to address the opioid crisis.

Data availability statement

Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the University of Toronto Health Sciences Research Ethics Board (reference 39013). The study is based on the use of administrative data.

Acknowledgments

We thank Nelson Chong for his work with conducting the data linkage for the Occupational Disease Surveillance System (ODSS).

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • X @nancycarnide

  • Contributors PAD and NC conceived the study. PAD, NC and JS contributed to the study design, methods and/or acquisition of funding. CS analysed the data. All authors contributed to interpretation of data. NC and JS wrote the initial draft of the manuscript. All authors critically reviewed and contributed to revising the manuscript for important intellectual content. All authors approved this final version. NC agreed to act as the guarantor of the work.

  • Funding This project has been made possible through funding from the Public Health Agency of Canada (2021-HQ-000092). The ODSS was initially funded by the Ontario Ministry of Labour, Immigration, Training and Skills Development (14-R-029) and the Public Health Agency of Canada (1516-HQ-000066). Both the Institute for Work & Health and the Occupational Cancer Research Centre receive support from Ontario’s Ministry of Labour, Immigration, Training and Skills Development. The Occupational Cancer Research Centre also receives support from Ontario Health.

  • Disclaimer The views expressed in this report do not necessarily represent the views of the Public Health Agency of Canada. The funders had no role in the conduct of this study, in the writing of the report, or in the decision to submit the article for publication. All inferences, opinions and conclusions drawn in this report are those of the authors and do not reflect those of the Province of Ontario.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Author note X Institute for Work & Health @iwhresearch; Occupational Cancer Research Centre @OCRC_CA

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.