Designing exposure registries for improved tracking of occupational exposure and disease.

Arrandale, Victoria H. ; Bornstein, Stephen ; King, Andrew 等

Effective surveillance systems can provide information on population-level trends in occupational exposure and disease. Surveillance is defined as the "systematic ongoing collection, collation, and analysis of data and the timely dissemination of information to those who need to know so that action can be taken". (1) Registries are one of various approaches to collecting information on occupational exposure and disease. Others include population-based surveys that tend to be cross-sectional in nature (rather than ongoing), providing a 'snapshot' of exposure or disease in a population. Sentinel event notification systems are designed to detect individual cases of disease (or exposure); these systems may lead to the development of research studies or additional surveillance activities. Screening can also be used, but is most often focused on detecting disease among at-risk individuals rather than in the population as a whole.

Exposure registries involve the registration of individuals based on their exposure to a particular agent or agents and the collection of various types of information, including demography, employment, exposure and, sometimes, health information. (2) An exposure registry generally seeks to include all exposed individuals within a specified population. (3) The information collected as part of an exposure registry may be used for several purposes, including: disease screening initiatives; compensation claims adjudication; primary prevention strategies (e.g., exposure reduction); and health surveillance within populations. It may also serve as a basis for additional surveillance activities.

Several industries and jurisdictions use exposure registries for tracking occupational exposures, and in some cases these registries are used as a basis for monitoring occupational disease or health status (e.g., allergic sensitization). In Canada, the systematic collection of data on occupational exposure and disease data is rare and exposure registries are no exception. However, leaders in the Canadian occupational health and safety community have recently shown interest in the use of registries for monitoring occupational exposure and disease, in part driven by the continuing epidemic of asbestos-related disease. (4) In particular, the recent development of the WorkSafe registry in British Columbia and the completion of the Baie Verte registry for asbestos miners in the province of Newfoundland and Labrador have drawn attention to the potential utility of registries for employees, employers, compensation boards and policy makers. The primary goal of this study was to review a series of existing exposure registries in order to identify their strengths and weaknesses as tools for tracking occupational exposure and disease in Canada. A further goal was to identify the features of an exposure registry that can contribute to prevention (primary, secondary or tertiary) and to identifying new exposure-disease relationships. A full report on this study is available online; (5) this paper focuses on results from a review of five Canadian exposure registries.

METHODS

Five registries operated by provincial or federal agencies were identified as the only exposure registries operating in Canada (Table 1). These five Canadian exposure registries were reviewed: Canadian National Dose Registry (NDR), Ontario Asbestos Workers Registry (Ontario Asbestos), Workplace Safety and Insurance Board (WSIB) Program for Exposure Incident Reporting (PEIR), WorkSafeBC Exposure Registry Program (WorkSafe), Baie Verte Miners' Registry (Baie Verte). (6-11)

Data on each registry were collected using documentary analysis of peer-reviewed literature, websites and reports. Supplementary information was collected through telephone conversations and e-mail exchanges with officials involved with each of the registries. Basic descriptive information was collected on all registries: the included exposure(s), active years of registration, exposure period captured, target population, number of registrants enrolled, methods used for data collection, data collected, data management and data access.

This study sought to examine the exposure registries with respect to a set of nine a priori features:

1. Overall registry goals

2. Legislated (L) or voluntary (V) registration--Did the organizations operating the registry, and the registrants themselves, have a choice in whether they participated in the registry?

3. Single (S) or multiple (M) occupational exposure(s) of interest --Did the registry monitor a single exposure, or multiple?

4. Inclusion of health surveillance (Y/N)--Did the registry also include health surveillance of any kind?

5. Prospective (P) or retrospective (R) exposure assessment--Did the registry collect exposure data prospectively or retrospectively?

6. Active (A) or passive (P) recruitment--Did the registry actively recruit registrants, or was enrolment left up to the workers?

7. Open (O) or closed (C) registration--Is the registry currently collecting data (open) or has data collection ceased (closed)?

8. Occupational (O) or environmental (E) exposure--Did the registry collect information on occupational, environmental or both types of exposure?

9. Individual (I) or population (P) focus--Did the registry focus on collecting information about individuals or about populations?

Of particular interest was whether enrolment in the registry was mandatory or voluntary and what types of data, especially involving exposure, were collected. If the registry used its data for research purposes, issues of informed consent were investigated, including ethical approval and how personal identifying information was handled. Any formal evaluation of the registry and its success in achieving its goals was reviewed, including whether each registry had contributed to prevention activities in the relevant jurisdiction.

RESULTS

The exposure registries examined varied considerably; they represent a variety of approaches to occupational exposure and disease tracking. In all cases, the target population was defined by exposure rather than disease. Table 1 presents key features of the registries, including the exposure being monitored, the population of interest, the date on which the registry activity began, and the period for which exposure was recorded. The NDR, Ontario Asbestos, WSIB and WorkSafe registries collected only exposure data, while the Baie Verte registry collected both exposure and health data. Table 2 presents the principal goals that each registry set for itself (the first of our nine a priori themes). Table 3 provides a summary of the five registries on the remaining eight a priori themes.

The goals of the registries varied. The WorkSafe and WSIB registries focused on reporting individual exposure incidents, in the hope that this information might be helpful in the future. The Ontario Asbestos registry aimed to identify asbestos-exposed workers and to notify them, when a set level of exposure is reached, about possible testing and treatment options. The Baie Verte registry was slightly different, aiming to collect a diverse range of information from former workers in order to assist in compensation claims and undertake epidemiological analysis. The NDR had the most complex goals, aiming to assist in the monitoring and control of radiation exposure in Canada as well as to provide data for compensation and research purposes. Evaluation of the exposure registries' success in achieving their stated goals was seldom undertaken; only the NDR was found to have undergone any evaluation.

Several strengths and weaknesses of these exposure registries were identified during the documentary review (Table 4). One of the main strengths of exposure registries generally is that workers are usually enrolled before disease occurs, which allows for primary prevention activities (e.g., removal from, or limitation of, exposure). The corollary to this is that exposure information is collected prospectively, before disease occurs, and is thus less subject to the challenge of recalling historical exposures. This issue is of particular concern when data are used in epidemiological studies of occupational disease (recall bias) but it can also be an important issue in the adjudication of workers' compensation claims.

Not all the exposure registries reviewed were prospective in nature. The Baie Verte registry was a retrospective exposure registry that enrolled workers many years (even decades) after exposure ceased; information on both exposure and disease was collected retroactively. All other reviewed exposure registries were designed to collect exposure information prospectively, though both WSIB and WorkSafe permitted retrospective reporting of exposure incidents. Only the NDR and Baie Verte registries included quantitative exposure measurements. The remainder of the exposure registries relied on a variety of reported exposure information, which could include the amount of a toxicant used, hours of work, or a simple yes/no report of exposure.

Exposure registries can be useful either when the exposure of interest is linked clearly to one or more health outcomes, or in situations where an exposure-response relationship is not yet known. The Ontario Asbestos and Baie Verte registries focused on asbestos-exposed individuals. The NDR focused specifically on radiation. The remaining exposure registries (WSIB and WorkSafe) set no restrictions on the types of exposure that could be reported.

Data collected as part of an exposure registry can also be used in the adjudication process for compensation claims. In the case of PEIR and WorkSafe, where any exposure can be reported by any workplace party, the hope is that this information will be archived for use in any future claims process. Workers electing to participate in these two exposure registries provide information on individual exposure incidents with no requirement for the provision of complete job or exposure history information.

Many of the registries reviewed were voluntary rather than mandatory. Only the NDR and the Ontario Asbestos registry were mandatory and both are prescribed as part of legislation. Data collected as part of a voluntary registry may not be suitable for population surveillance or epidemiology as information on the population as a whole is likely to be lacking. Where possible, comparison of participants and nonparticipants can assist in assessing generalizability of the registry.

The review of the five Canadian exposure registries led to the development of eight key factors to consider when planning an exposure registry: overall goal(s), definition of exposure, intended uses of registry data, whether registration will be voluntary or mandatory, logistics, privacy, management of expectations and equity. The registry goals are arguably the most important determinant of what information needs to be collected and how (Table 5). It is important to note that it is possible for a registry to fulfill multiple goals (i.e., both to facilitate compensation and to conduct exposure surveillance in a population), though this is likely to increase the resources and costs involved.

The definition of the exposure of interest must be clear and this definition must be communicated clearly to stakeholders and potential participants. A decision will be required as to whether a self-reported exposure will suffice or whether reported exposures will have to be independently confirmed. When a registry is using previously collected data, this point may be moot. If there is interest in using the collected data beyond the basic registry activities (for research purposes, for example), this should be considered early on and should involve stakeholders, both those whose data will be collected and those who might wish to make use of the data. It is critical to understand the limitations of voluntary registries. Voluntary registries can be sufficient for collecting information to assist with individual compensation claims or for providing individual-level screening. However, if a voluntary registry is also intended to contribute to population-level surveillance, it will need to capture a very large (and representative) portion of the target population.

Registries that operate prospectively and want to monitor workers' exposure to a particular hazard or hazards will need to be able to link registrants' exposure information collected over time and potentially across workplaces. This will require a system for identifying the individual within the registry and care should be taken to protect the privacy of registrants, preferably through the replacement of personally identifying information with encrypted identification codes.

Logistical issues will also need to be considered, specifically who will be responsible for managing the registry, how data will be collected, how errors will be corrected and entries updated, who will have access to the data, and how access will be monitored. The expectations of registrants also need to be addressed. Equitable access to exposure registries must be ensured, particularly voluntary registries that may confer benefits at a later date in terms of compensation or medical care.

DISCUSSION

The design and implementation of exposure registries (and other surveillance approaches) are complex tasks and are likely to involve trade-offs in terms of cost and prospective benefits for both individuals and policy-makers. Many of these decisions hinge on the goals of the registry, the exposure of interest, and the reasons for initiating a registry. The five Canadian exposure registries varied considerably. Our review led to the identification of eight key factors to consider when planning an exposure registry: overall goal(s), definition of exposure, whether registration will be voluntary or mandatory, intended uses of registry data, management of the registry and logistical arrangements, privacy, management of expectations, and ensuring equitable access to registry activities.

Some of the Canadian registries, for example the NDR, focus on a single exposure, collecting detailed exposure information on workers and playing a significant role in the prospective monitoring of exposure levels at the individual and population levels, as well as triggering removal from exposure when set thresholds are reached. The NDR has also been an important source of data for several epidemiological investigations of exposure and disease among different occupational groups in Canada. (12-14) The Ontario Asbestos Workers Registry, also a mandatory registry, has similar goals in terms of prospectively monitoring exposure and notifying workers who reach a particular duration of exposure, but has not to date been used for any surveillance or epidemiological investigations, though these activities could be supported. (15) Though none of the Canadian registries did so, exposure registries can also be used as a basis for medical screening in individuals. Examples of this exist in the US and include the Beryllium Associated Workers Registry at the U.S. Department of Energy and the Tremolite Asbestos Registry in Libby, Montana. (16,17)

With the exception of the NDR, the data in Canadian exposure registries have not been extensively used for research or surveillance purposes. The WorkSafe and the Baie Verte registries are relatively new; data from the latter have been used in epidemiological analyses and publications are expected. (18) The Ontario Asbestos Workers Registry data have been described by Kone Pefoyo et al. (2014) but never used for any type of tracking, surveillance or research. (15) The WSIB PIER data seem only to be used internally for evaluating claims.

One challenge for three of the Canadian registries (WSIB, WorkSafe, Baie Verte) was their voluntary nature. In these cases, information collected will likely be incomplete at both the individual and the population level, making it challenging to reconstruct individual exposure histories at a later date, and rendering the data less useful for surveillance or epidemiology in the population. The key piece of information when undertaking surveillance is the population size, which serves as the denominator when calculating rates of exposure or disease; without this, it is challenging to estimate the rate of exposure or disease occurrence in the population accurately.

Voluntary registries also offer fewer options for undertaking rigorous evaluation. A mandatory registry can be linked to enforcement efforts within the health and safety system. An example of this is the mandatory Finnish ASA registry* which collects information on carcinogen use (both type and amount) in Finnish workplaces. (19,20) Kauppinen et al. (2007) undertook an evaluation of the ASA registry by focussing on workplaces that had ceased reporting, to determine whether carcinogens were no longer in use, or whether the cessation of reporting was in error. (21) Making the exposure registry reporting available to front-line inspectors provides a mechanism for reminding workplaces about the need to report (where mandatory) or the benefits of reporting (where voluntary). Linkages with regulatory bodies that collect exposure measurement can also provide the opportunity for evaluating whether a registry has any effect on exposure prevalence or exposure levels in workplaces, or industries more broadly. The only Canadian registry where some form of evaluation had been undertaken is the NDR; exposure data in this registry have been compared to self-reported exposure among a cohort of nurses. (22) Results highlighted the gaps in the NDR data in this occupational group. Evaluating an exposure registry against its initial goals is challenging but is important for ensuring effectiveness and also for allowing ongoing adjustment and improvement of registry design and activities. In a voluntary registry, participation rates are not particularly informative, but measures of data completeness and evaluation of the reliability and/or validity of exposure reports can provide alternative ways of assessing the success of a registry. Though superior in many ways, mandatory registries are likely to require legislative or regulatory action. This can make them more difficult to initiate as well as to modify later on.

Three of the Canadian registries relied on self-reported exposures; two (NDR and the Baie Verte registry) included measurement data. The inclusion of measurement data comes at a significant cost. In the US, the Beryllium Associated Workers Registry collects exposure measurements from all sites; this program has an annual budget in excess of US $1 million (personal communication, B. Richter, September 18, 2013). Exposure registries that include quantitative exposure information allow for the investigation of dose-response relationships within an exposed population, a necessary condition for demonstrating a causal relationship. If a registry is relying on self-reported exposures, it may be better to set the bar low and recruit any worker who may have been exposed even at a low level. Challenges with exposure assessment and verification of self-reported exposure have previously been discussed in the context of environmental exposure registries. (23) In an occupational exposure registry, it is possible to assign exposure estimates at a later date based on job histories, historically collected data from similar workplaces, or existing job-exposure matrices. These activities will be best supported if job history data are collected systematically, using standardized job and industry codes. Where possible, information on exposure duration, frequency, job tasks, processes, chemical names, personal protective equipment, ventilation used, and the specific time periods involved should be collected. (24)

There are many factors to consider when planning an exposure registry; eight were highlighted in the results section. The goals of a registry will determine what information should be collected and from whom. It is clear that planning and consultation are necessary. (2,25) It is advisable to engage with the stakeholders at an early stage so that concerns, comments, ideas and expectations can be discussed before decisions about these issues are made. (25) This can also serve to ensure that the goals, data to be collected, method of collection, storage methods and format are amenable to the intended secondary uses. If the data required for secondary uses are not collected upfront, or consent for their future collection is not obtained, future health surveillance and epidemiology may require making contact with each individual registrant to obtain consent for these activities. This is likely to be prohibitively expensive and logistically challenging and may run up against privacy issues.

With regards to privacy, the inclusion of personal health information at the outset or as a secondary health surveillance activity will be subject to the laws in the relevant jurisdictions. Privacy and legal concerns can present challenges to the development of an exposure registry, particularly in cases where the collection of, or linkage with, additional health information is the goal. In some cases, a third party can be contracted to carry out the data collection and registry management, as in the case of the Finnish ASA Registry. If so, the contracted party should ideally be an organization with expertise in the area of exposure assessment, data management and analysis--areas in which employers, worker representatives, governments, regulatory agencies and compensation boards often lack sufficient expertise. The potential downside is that this third party then becomes the data steward for registrants' personal information; some parties may not support this arrangement.

In terms of communicating with workers and potential registrants, an exposure registry must not create or support unrealistic expectations among registrants, especially concerning the relationship between exposure and disease or the impact of registration on future compensation decisions. Not all exposed workers will become ill and, among workers who do become ill, not all of these cases will be the result of occupational exposures. Clear and careful communication with potential registrants may help limit confusion and frustration. Barriers to registration should also be minimized. Inequity could arise from differences in company size, unionization status of the workplace, type of employment contract (e.g., temporary/permanent, domestic/ foreign), demographics, or workplace culture across occupational or industry groups. Communication to potential registrants should be broad so that all exposed workers are made aware of the registry and of the opportunity to enrol.

CONCLUSION

Existing Canadian exposure registries vary considerably. They offer several options on which to model new exposure registries, as do various international registries. The exposure registry approach has numerous strengths for collecting data on targeted populations but also a number of actual and potential limitations. Exposure registries provide the opportunity to intervene early in the exposure-disease pathway and can facilitate primary prevention through elimination or reduction of exposure. Registries can also facilitate secondary prevention through screening in high exposure groups and population surveillance activities. However, what can be achieved with data from an exposure registry depends on several factors, most notably the proportion of the target population who enrol and what information is collected from participants.

The development of new exposure registries, regardless of their area of focus (industry, occupation or specific exposures), requires careful consideration. When planning new registries, the overall goals should be clearly defined and the registry designed to support the stated goals. Canada faces unique challenges because of the size of the country, our federal system that assigns primary responsibility for occupational health and safety to the provinces and territories, and the diversity of industry within and between provinces. New registries will ideally have mandatory registration and be linked to enforcement activities, while ensuring that appropriate privacy protections are in place and enrolment is accessible for all eligible workers.

REFERENCES

(1.) Last JM (Ed.). A Dictionary of Epidemiology, 4th ed. New York: Oxford University Press, 2001. doi: 10.1093/acref/9780195314496.001.0001.

(2.) Schulte PA, Kaye WE. Exposure registries. Arch Environ Health 1988; 43(2):155-61. PMID: 3288130. doi: 10.1080/00039896.1988.9935845.

(3.) Goldberg J, Gelfand HM, Levy PS. Registry evaluation methods: A review and case study. Epidemiol Rev 1980; 2:210-20. PMID: 7000537.

(4.) Del Bianco A, Demers PA. Trends in compensation for deaths from occupational cancer in Canada: A descriptive study. CMAJ Open 2013; 1(3): E91-96. PMID: 25077112. doi: 10.9778/cmajo.20130015.

(5.) Bornstein S, Demers PA, Arrandale VH. Tracking Occupational Exposure and Disease: An Analysis of Approaches for the Canadian Context. Report to WorkSafeBC. 2014. Available at: http://www.worksafebc.com/contact_us/ research/research_results/res_60_10_1440.asp (Accessed October 27, 2015).

(6.) Ashmore J, Grogan D. The National Dose Registry for radiation workers in Canada. Radiat Prot Dosimet 1985; 11(2):95-100.

(7.) Health Canada. National Dose Registry. Available at: http://www.hc-sc.gc.ca/ ewh-semt/occup-travail/radiation/regist/index-eng.php (Accessed April 2, 2015).

(8.) Ontario Ministry of Labour. A Guide to the Regulation Respecting Asbestos on Construction Projects and in Buildings and Repair Operations. 2011. Available at: http://www.labour.gov.on.ca/english/hs/pdf/asbestos_guide.pdf (Accessed October 27, 2015).

(9.) Workplace Safety and Insurance Board (WSIB). Occupational Disease. Available at: http://www.wsib.on.ca/WSIBPortal/faces/WSIBArticlePage?fGUID= 835502100635000357&_afrLoop=593501850478000&_afrWindowMode= 0&_afrWindowId=tpjdeox1q_1#%40%3F_afrWindowId%3Dtpjdeox1q_1% 26_afrLoop%3D593501850478000%26_afrWindowMode%3D0%26fGUID% 3D835502100635000357%26_adf.ctrl-state%3Dtpjdeox1q_25 (Accessed April 2, 2015).

(10.) WorkSafeBC. Occupational Diseases. Available at: http://www.worksafebc. com/claims/occupational_diseases/default.asp (Accessed April 2, 2015).

(11.) Baie Verte Miners' Registry. Available at: http://www.mun.ca/bvminers (Accessed October 27, 2015).

(12.) Newcombe HB. Design and future uses of national dose registers for regulatory control and epidemiology. Health Phys 1980; 39(5):783-96. PMID: 7204029. doi: 10.1097/00004032-198011000-00008.

(13.) Zablotska LB, Lane RS, Thompson PA. A reanalysis of cancer mortality in Canadian nuclear workers (1956-1994) based on revised exposure and cohort data. Br J Cancer 2014; 110(1):214-23. PMID: 24231946. doi: 10. 1038/bjc.2013.592.

(14.) Zielinski JM, Ashmore PJ, Band PR, Jiang H, Shilnikova NS, Tait VK, et al. Low dose ionizing radiation exposure and cardiovascular disease mortality: Cohort study based on Canadian National Dose Registry of radiation workers. Int J Occup Med Environ Health 2009; 22(1):27-33. PMID: 19329385. doi: 10.2478/ v10001-009-0001-z.

(15.) Kone Pefoyo AJ, Genesove L, Moore K, Del Bianco A, Kramer D. Exploring the usefulness of occupational exposure registries for surveillance: The case of the Ontario Asbestos Workers Registry (1986-2012). J Occup Environ Med 2014; 56(10):1100-10. PMID: 25162835. doi: 10.1097/JOM.0000000000000235.

(16.) U.S. Department of Energy Office of Health and Human Security. Beryllium Associated Worker Registry Summary-2012. Available at: http://energy.gov/ ehss/beryllium-associated-worker-registry (Accessed April 2, 2015).

(17.) Bateson T. Surveillance activities related to asbestos-contaminated vermiculite ore from Libby, Montana. In: Canadian Partnership Against Cancer: Asbestos Surveillance and Compensation Meeting-March 4, 2010; Toronto, ON, 2010.

(18.) Giles Murphy TL. Asbestos exposure and incidence of disease among a group of former chrysotile asbestos miners and millers from Baie Verte, NL, Canada, PhD thesis, St. John's, NL: Memorial University of Newfoundland, 2015.

(19.) Alho J, Kauppinen T, Sundquist E. Use of exposure registration in the prevention of occupational cancer in Finland. Am J Ind Med 1988; 13(5):581-92. PMID: 3376947. doi: 10.1002/(ISSN)1097-0274.

(20.) Kauppinen T. Finnish occupational exposure databases. Appl Occup Environ Hyg 2001; 16(2):154-58. PMID: 11217703. doi: 10.1080/104732201460253.

(21.) Kauppinen T, Saalo A, Pukkala E, Virtanen S, Karjalainen A, Vuorela R. Evaluation of a national register on occupational exposure to carcinogens: Effectiveness in the prevention of occupational cancer, and cancer risks among the exposed workers. Ann Occup Hyg 2007; 51(5):463-70. PMID: 17625219. doi: 10.1093/annhyg/mem030.

(22.) Teschke K, Chow Y, Chung J, Ratner P, Spinelli J, Le N, Ward H. Estimating nurses' exposures to ionizing radiation: The elusive gold standard. J Occup Environ Hyg 2007; 5(2):75-84. PMID: 18075880. doi: 10.1080/ 15459620701793050.

(23.) Schultz MG, Sapp JH 2nd, Cusack CD, Fink JM. The National Exposure Registry: History and lessons learned. J Environ Health 2010; 72(7):20-25. PMID: 20235405.

(24.) Froines J, Wegman D, Eisen E. Hazard surveillance in occupational disease. Am J Public Health 1989; 79(Suppl):26-31. PMID: 2817208. doi: 10.2105/AJPH. 79.Suppl.26.

(25.) Trout DB, Schulte PA. Medical surveillance, exposure registries, and epidemiologic research for workers exposed to nanomaterials. Toxicology 2010; 269(2-3):128-35. PMID: 20006668. doi: 10.1016/j.tox.2009.12.006.

Received: April 7, 2015

Accepted: October 27, 2015

* Finnish Register of Employees Exposed to Carcinogens.

Victoria H. Arrandale, PhD, [1,2] Stephen Bornstein, PhD, [3] Andrew King, LLB, LLM, [4] Timothy K. Takaro, MD, MPH, [5] Paul A. Demers, PhD [1,2]

Author Affiliations

[1.] Occupational Cancer Research Centre, Cancer Care Ontario, Toronto, ON

[2.] Dalla Lana School of Public Health, University of Toronto, Toronto, ON

[3.] Centre of Applied Health Research, Memorial University, St. John's, NL

[4.] School of Labour Studies, McMaster University, Hamilton, ON

[5.] Faculty of Health Sciences, Simon Fraser University, Burnaby, BC

Correspondence: Victoria H. Arrandale, PhD, Occupational Cancer Research Centre, Cancer Care Ontario, 525 University Avenue (3-010), Toronto, ON M5G 1X3, Tel: 416-971-9800, E-mail: victoria.arrandale@cancercare.on.ca

Acknowledgements: This study was funded by WorkSafeBC through the Focus on Tomorrow--Innovation at Work Program in partnership with the Workers' Compensation Board of Nova Scotia and Alberta Human Services. VA was supported by a Career Development Award in Prevention from the Canadian Cancer Society Research Institute. The authors thank Dr. Mieke Koehoorn for providing valuable comments on the manuscript. Institution where work was completed: Memorial University, Newfoundland & Labrador, Canada

Conflict of Interest: None to declare. Table 1. Key features of the five exposure registries Exposure Year Exposure Target started period population Canadian Ionizing 1951 1951-ongoing All radiation- National Dose radiation exposed workers Registry in Canada Ontario Asbestos 1986 1986-ongoing All asbestos- Asbestos exposed workers Workers in Ontario, Registry Canada Ontario WSIB Various 2002 2002-ongoing All workers in Program for Ontario, Canada Exposure Incident Various 2012 2012-ongoing All workers in Reporting British WorkSafeBC Columbia, Exposure Canada Registry Program Baie Asbestos 2008 1959-1994 All workers Verte Miners' from the former Registry Baie Verte, Newfoundland asbestos mine Approx. # registrants (year) Canadian 500,000 National Dose (2012) Registry Ontario 24,000 Asbestos (2004) Workers Registry Ontario WSIB 16,000 Program for (2011) Exposure Incident N/A Reporting WorkSafeBC Exposure Registry Program Baie 1003 Verte Miners' (2012) Registry N/A = not available. Table 2. The stated goal(s) of each exposure registry Goal(s) Canadian National To assist regulatory authorities in the Dose Registry monitoring and control of occupational radiation exposures, provide exposure information for legal/compensation purposes and to serve as a source of exposure data for epidemiological analyses. Ontario Asbestos To identify asbestos-exposed workers and notify Workers Registry workers and their physicians of the potential need for appropriate diagnostic testing and, potentially, therapeutic treatment. Ontario WSIB Program To provide a method for workers and employers for Exposure to report individual unplanned exposure Incident Reporting incidents in the workplace that do not result in lost time and do not result in an immediate injury or illness. WorkSafeBC Exposure To provide a system for workers, employers and Registry Program others to report an exposure to a harmful substance in the workplace as a means to assist the adjudication of any future occupational disease claim as a result of the reported exposure. Baie Verte Miners' To establish contact with as many former Registry employees as possible from one former asbestos mine and to collect (with full informed consent) as much information as possible on work histories, asbestos exposure, personal health history, and current health status. Table 3. Thematic analysis of the five registries (themes 2-9) Legislated Single Health (L) or (S) or surveillance? voluntary multiple (M) (Y/N) (V)? occupational exposure(s) of interest? Canadian National L S N Dose Registry Ontario Asbestos L S N Workers Registry Ontario WSIB Program V M N for Exposure Incident Reporting WorkSafeBC Exposure V M N Registry Program Baie Verte Miners' V S Y Registry Prospective Active (A) Open (O) or (P) or or passive (P) closed (C)? retrospective recruitment? (R)? Canadian National P A O Dose Registry Ontario Asbestos P A O Workers Registry Ontario WSIB Program P P O for Exposure Incident Reporting WorkSafeBC Exposure P P O Registry Program Baie Verte Miners' R A C Registry Occupational (O) Individual or environmental (I) or (E) exposures? population (p) focus? Canadian National O P Dose Registry Ontario Asbestos O I Workers Registry Ontario WSIB Program O I for Exposure Incident Reporting WorkSafeBC Exposure O I Registry Program Baie Verte Miners' O P Registry Table 4. Strengths and limitations of exposure registries as a tool for tracking occupational exposure and disease in Canada Strengths Limitations * Can provide the opportunity to * Financial costs for design, intervene before the onset of implementation and operations. disease (primary prevention). * Utility of registry data for * Can collect exposure population surveillance is information prospectively on limited if participation is not workers. mandatory. * If participation is high (more * A mandatory registry may likely in a mandatory registry), require legislation. a registry can be used as a basis for population-level surveillance * Development of a registry of occupational exposure and/or presents many challenges, disease. including participation of workers, cooperation between * Allow for the investigation workplace parties, operational of new exposure-response costs, ethics requirements. relationships. * If detailed personal exposure * National registries are information is collected, this difficult to organize in Canada can assist individuals in the because occupational health and assessment of workers' safety (for most workers) is a compensation claims, regardless provincial and not federal of the population participation responsibility. rates. Table 5. Exposure registry goals, requirements and limitations Goal Information required Limitations to achieve this goal 1. To conduct exposure * exposure * best undertaken surveillance in order information for the with a mandatory to prevent or reduce entire population of registry hazardous workplace interest (or of a exposure in representative populations of workers sample) [right arrow] Primary * if not available, Prevention the registry may still contain data that are useful for individuals in compensation (see Goal #4) 2. To conduct health * a population with * will not surveillance in order known exposure necessarily include to prevent or reduce detailed exposure disease occurrence in * health information information at the populations of workers for the entire individual level population of [right arrow] interest (or a * may not be useful Secondary Prevention representative for epidemiology sample) 3. To protect * a population with * will not individual workers who known exposure necessarily include are known to have had detailed exposure hazardous exposures * provision of health information at the from the progression screening at the individual level of disease (screening) individual level * may not be useful [right arrow] for epidemiology Secondary Prevention 4. To assist in the * detailed personal * limited uses for filing and exposure information the data beyond the adjudication of future individual compensation for * if a large individual workers proportion of the population enrols, [right arrow] Tertiary the data may also be Prevention useful for population-level * data likely not to health surveillance be useful for exposure (or health) surveillance purposes 5. To identify new * detailed exposure exposure-disease and health * expensive relationships through information for a epidemiological representative analyses in population, including populations of workers both exposed/ unexposed and healthy/diseases (cases/controls)