Article Text
PDF
Abstract
Objectives To describe recent investigations of potential workplace cancer clusters.
Methods We identified Health Hazard Evaluations (HHEs) of cancer concerns during 2001–2020. We described information about industry, requestors, cancer characteristics, investigative procedures, and determinations about the presence of a cluster (ie, presence of excess cases, unusual case distribution or exposure).
Results Of 5754 HHEs, 174 included cancer concerns, comprising 1%–5% of HHEs per year. In 123 HHEs, the cancer cluster concerns involved different cancer primary sites. Investigation procedures varied but included record review (n=63, 36%) and site visits (n=22, 13%). Of 158 HHEs with a cluster determination by investigator(s), 151 (96%) were not considered cancer clusters. In seven HHEs, investigators found evidence of a cluster, but occupational exposure to a carcinogen was not identified.
Conclusions The proportion of HHEs on workplace cancer cluster concerns remained steady over time; most did not meet the definition of a cluster or uncover an occupational cause. Public health practitioners can use this information to provide updated context when addressing workplace cancer cluster concerns and as motivation to refine investigative approaches. More broadly, this review highlights an opportunity to identify best practices on how to apply community cluster investigation methods to the workplace.
- Public health
- Occupational Health
- Public Health Surveillance
Data availability statement
Data are available on reasonable request.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Suspected workplace cancer clusters are distressing for those affected and can be challenging to investigate. Currently, a consistent framework for applying principles of community cluster investigation guidance to workplace investigations does not exist.
WHAT THIS STUDY ADDS
This study provides a review of recent investigations of potential workplace cancer clusters, investigative methods used and outcomes.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Public health practitioners can use this information to refine investigative approaches and identify best practices on how to apply existing cluster investigation methods to the workplace.
Introduction
Cancer clusters, defined as a greater-than-expected number of the same or aetiologically related cancer cases within a group of people in a geographic area over a period of time, can be distressing for those affected.1–3 Investigations of suspected cancer clusters are important for addressing concerns but rarely identify unusual patterns of cancer or a common cause, prompting need for continuous evaluation of the investigative process and outcomes.3
The National Institute for Occupational Safety and Health’s (NIOSH) Health Hazard Evaluation (HHE) programme was created to comply with a mandate in the Occupational Safety and Health Act of 1970. The programme responds to valid requests for investigation of workplace health hazards from employers, employees and their representatives, and government agencies (https://www.cdc.gov/niosh/hhe/default.html). In response to a request, HHE programme staff perform an evaluation and provide recommendations for controlling occupational health hazards and improving worker safety. Requests concern a range of chemical, physical and biological hazards, and adverse health outcomes. One health outcome the programme is often asked to investigate is suspected cancer clusters in workplaces. Although there is no standardised approach for investigating occupational cancer clusters, NIOSH investigators apply principles used in community cancer cluster investigations and use several lines of epidemiological evidence to make determinations about the presence of a cluster.
A previous review examined 61 cancer cluster HHEs during 1978–1984.4 NIOSH has continued to respond to similar HHE requests. The objective of this analysis was to (1) summarise workplace cancer concerns, investigation methods used and findings from cancer cluster HHEs conducted by NIOSH during 2001–2020 and (2) use this descriptive information to provide context and inform best practices for future investigations. This review focused on recent experience rather than beginning when the previous review ended to provide public health practitioners with current information.
Methods
We analysed HHEs on potential workplace cancer clusters requested, investigated and completed during 1 January 2001–31 December 2020. We queried NIOSH’s HHE database with the keyword cancer and analysed all completed HHEs identified.
We extracted the following information from the database into Microsoft Access: open date, close date, requestor type, North American Industry Classification System code (https://www.census.gov/naics/), workplace state and Occupational Safety and Health Administration region (OSHA Alliance Programme%7COccupational Safety and Health Administration). We abstracted additional information from final documentation for each HHE: number of workers, number and type of cancer(s), investigation procedures (ie, cancer registry, medical, and environmental record review; employee interview or questionnaire administration; industrial hygiene sampling; rate calculations such as proportional mortality ratio, standardised incidence ratio or standardised mortality ratio; site visit), and presence of a suspected exposure.
Because there is no standard method for determining the presence of an occupational cancer cluster, investigators considered individual cancer cluster criteria (ie, presence of an excess of cancer cases, an unusual distribution of cancer or carcinogens in the workplace), and latency to make a final cluster determination about whether a cancer cluster existed, all of which we abstracted. An unusual distribution of cancer is defined as the occurrence of cancer in a demographic group that is not typical (eg, breast cancer in males). Latency is the duration between the start of potential exposure or employment at the work site and cancer diagnosis. Investigators typically considered latency only when one or more of the other individual cancer cluster criteria was met. We calculated the frequency of variables listed above and presented counts and proportions for all categorical variables using R (V.4.0.3; R Foundation for Statistical Computing). If an investigator omitted their determination on individual cancer cluster criteria or their final cluster determination in the abstracted documentation for each HHE, we considered this information missing. In some HHEs, there was information available on an investigators’ final cluster determination but not on individual cancer cluster criteria determinations. We calculated the frequency of variables listed above and presented counts and proportions for all categorical variables using R (V.4.0.3; R Foundation for Statistical Computing). This activity was reviewed by the Centers for Disease Control and Prevention (CDC) and conducted in accordance with applicable federal law and CDC policy (See for example, 45 C.F.R. part 46; 21 C.F.R. part 56; 42 U.S.C. §241(d), 5 U.S.C. §552 a, 44 U.S.C. §3501 et seq.).
Results
Of 5754 HHEs identified, 237 were completed and had ‘cancer’ as a keyword; we excluded duplicates (n=9) and HHEs where cancer was not a main concern (n=54), leaving 174 HHEs for analysis. HHEs focused on cancer cluster concerns comprised 1%–5% of HHEs per year. The most common industry sector was public administration (32%) (table 1). Most requests came from employees (40%), management (35%) and union representatives (19%).
Characteristics and findings of completed NIOSH HHEs of workplace cancer cluster concerns—the USA, 2001–2020
HHEs included 1–117 cancer cases, involving <1%–50% of the workforce (data not shown). Concerns usually involved multiple cancer sites (n=123/174, 71%). The most frequently reported cancer sites were breast (n=93), lung/bronchus (n=53), and colon/rectum (n=50). Investigators reviewed records for 36% (n=63) of HHEs; records reviewed included cancer registry data (n=10), environmental records (n=53) and medical records (n=19). Investigators conducted interviews or administered questionnaires for 14% (n=25) of HHEs, made a site visit for 13% (n=22), calculated cancer rates in 8% (n=14) and conducted environmental sampling in 7% (n=13) of HHEs.
Seventy-four per cent (128/174) of HHEs included investigator determinations regarding individual cancer cluster criteria (table 1). Most (n=110, 63%) found no evidence of an excess or unusual distribution of cases and no presence of a known exposure. Eighteen HHEs (10%) met ≥1 criterion; in 8/18 (44%), investigators found a sufficient latency period (data not shown).
In 158 of 174 HHEs (91%), investigators made a final determination regarding evidence of a cluster (table 1); 7/158 (4%) were considered a cluster, the remaining 151 (96%) were not. Of the seven HHEs where evidence of a cluster was found, investigators found an unusual distribution of cancer in six (86%) (online supplemental table 1). Three (43%) involved brain cancers. Three (43%) were from the healthcare and social assistance sector. Requestors reported exposures of concern in two of the seven clusters (29%); exposures of concern were radon and ionising radiation. However, investigators determined that occupational exposure to a carcinogen was unknown or unclear in all seven HHEs.
Supplemental material
Discussion
The proportion of HHEs conducted to investigate cancer cluster concerns has remained constant during the last two decades. As with community cancer concerns, workplace cancer concerns are distressing to all parties involved and there is often an expectation that an investigation will document an unusual pattern of cancer linked to a workplace exposure.5 However, similar to Schulte et al, we found that most concerns submitted to the HHE programme involved cancer sites that are prevalent in the US population and lacked evidence of a common workplace carcinogenic exposure.4 6
Nonetheless, government agencies have a responsibility to respond to concerns affecting populations they serve even when the situation is not ideal for conducting a high-quality, epidemiological study.4 In these situations, investigators must rely on available data, resources and guidance to evaluate the concerns. Investigation steps taken as part of the HHEs in this review varied widely. For example, only a third of cancer cluster HHEs involved record review. Variability in the approaches taken reflects the characteristics of the concern (eg, presence of factors indicating the need for further investigation) and what types of data were available. While HHE programme investigators rely on principles provided in community investigation guidance, no roadmap for applying such guidance to occupational settings exists.
During 2001–2020, several versions of community cancer investigation guidance were in place, which included slightly different definitions of ‘cancer cluster.’1 3 The definition from CDC’s 2013 cancer cluster investigation guidelines1 encompassed some of the time frame of HHE investigations reviewed. In 2022, CDC published updated guidance including a revision to the definition of a cancer cluster to recognise that different cancers may be similar aetiologically, which can be used moving forward.7 While investigation principles discussed in community investigation guidance are applicable to the occupational setting, available data sources and thus methods differ. Cancer registry and denominator data are based on residence whereas a workplace may draw workers from multiple geographic areas, even crossing state boundaries at times. Administrative data available from workplaces may not always include demographic or work history information to define the population at risk. Finally, because of the healthy worker effect and long latency for some cancers, identifying an appropriate comparison population can be challenging. As a result, calculating cancer rates or comparing incidence to a referent population can require additional considerations. One element that may be more straightforward in the occupational setting is identification of carcinogens through review of records and sampling, when appropriate. Recognising these distinctions and unique challenges illustrates the importance of a tailored approach to occupational cluster investigations that builds on best practices identified in community investigations.
A tailored approach to occupational cancer cluster investigations can benefit the HHE programme and other entities, including other health agencies that perform these investigations or refer concerned parties to the HHE programme or other entities. A common framework established in collaboration with organisations that perform occupational cluster investigations such as state or local health agencies and cancer registries should incorporate best practices from existing guidelines from within the United States and other countries, address how to apply existing guidance to workplace settings, and encourage interdisciplinary collaboration. Based on this review and differences between community and occupational cancer cluster investigations, we propose three areas of focus to aid public health practitioners in developing best practices for occupational cancer cluster investigations and inform what should be considered in a framework. First, similar to community investigation guidance, having a stepwise process for determining the need for further investigation is helpful in providing an efficient response to concerned parties. Second, identifying administrative data sources to define and understand the population of workers who should be included in the evaluation is necessary. Third, stronger collaboration between occupational health practitioners and cancer registries can help facilitate case identification and cluster determination.
Considering our results and previous findings, it is reasonable to ask whether the benefits of these investigations outweigh the costs. However, there are many benefits. Such investigations can provide surveillance for workplace exposures to unknown or suspected carcinogens. Approximately one-third of human carcinogens were first documented in worksite studies.8 Further, these investigations provide opportunities to engage with workers, to encourage prevention and screening as appropriate.9 How investigations address heightened emotions surrounding a suspected workplace cancer cluster is hard to quantify.2 The role of active listening and validation of worker concerns cannot be undervalued and is a critical public health responsibility.5 Workplace cancer cluster investigations can be opportunities to empower and engage workers, management and unions to speak up about and address workplace health concerns, establish trust and provide education.5 10 11 These benefits should be considered when developing more tailored guidance for occupational cancer cluster investigations.
The findings presented here are subject to several limitations. Data were abstracted from existing investigation documents; some were incomplete. Methods for investigating cancer clusters were not standardised. The biological plausibility of carcinogenesis related to reported exposures, an investigator’s training, and requestors’ perception of workplace health hazards affect investigation design, limiting consistency across HHEs. Finally, our findings might not be generalisable as our programme’s investigations might not represent all workplace cancer cluster investigations.
This evaluation indicates that workplace cancer cluster concerns continue to be reported. Although investigations often find no evidence of a cancer cluster or common cause, it remains important to address these concerns using a clear approach understandable to all parties. Public health practitioners can use this information to provide context for future cancer cluster investigations, refine investigative approaches and more broadly, to develop a national framework to apply cluster investigation methods to the workplace.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
CDC did not consider this activity research as defined in 45 CFR 46.102(l) because the data were collected through HHEs, which are a public health practice activity conducted by NIOSH under statutory authority (42 CFR 85). Therefore, IRB review was not required.
Acknowledgments
We acknowledge Sheldon Walker, Barbara Jenkins, Kelsey Sumner and Kaitlin Kelly-Reif for their contributions to this project.
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.
Footnotes
Contributors SKC, GRG and DSS planned and performed data abstraction for the study. DSS performed the data analysis. JLR and SKC supervised the analysis and assisted with data interpretation. All authors contributed to the final version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data belong to the Health Hazard Evaluation (HHE) programme and are available on reasonable request.
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.