Cancer Risks among Emergency Medical Services Workers in Ontario, Canada

Abstract

Objectives

Emergency medical services workers, such as paramedics, provide important emergency care and may be exposed to potential carcinogens while working. Few studies have examined the risk of cancer among paramedics demonstrating an important knowledge gap in existing literature. This study aimed to investigate cancer risks among paramedics in a large cohort of Ontario workers.

Methods

Paramedics were identified in the Occupational Disease Surveillance System (ODSS) from 1996 to 2019. The ODSS was established by linking lost-time worker’s compensation claims to administrative health data, including the Ontario Cancer Registry to identify incident cases of cancer. Cox-proportional hazard models were used to calculate age and sex-adjusted hazard ratios and 95% confidence intervals to estimate the risk of cancer among paramedics compared to all other workers in the ODSS.

Results

A total of 7240 paramedics were identified, with just over half of the paramedics identifying as male similar to the overall ODSS cohort. Paramedics had a statistically significant elevated risk of any cancer (HR 1.19, 95% CI 1.06–1.34), and elevated risks for melanoma (HR 2.18, 95% CI 1.46–3.26) and prostate cancer (HR 1.73, 95% CI 1.34–2.22). Paramedics had a statistically significant reduced risk for lung cancer (HR 0.48, 95% CI 0.28–0.83). Findings were similar to cancer risks identified in firefighters and police in the same cohort.

Conclusions

This study contributes valuable findings to understanding cancer risks among paramedics and further supports the existing evidence on the increased risk of cancer among emergency medical services workers. We have observed some similar results for firefighters and police, which may be explained by similar exposures, including vehicle exhaust, shiftwork, and intermittent solar radiation. This can lead to a better understanding of carcinogens and other exposures among paramedics and inform cancer prevention strategies.

Introduction

Emergency medical services (EMS) workers, specifically paramedics, provide medical care and transport to patients, in non-hospital settings as part of emergency services. Paramedics respond to a variety of calls at private homes, public spaces, and at emergency incidents such as fires, building collapses, motor vehicle accidents, and natural disasters (1). The unpredictable and dynamic nature of paramedic work can potentially expose them to carcinogens as part of their daily work (e.g., shiftwork, diesel and gas vehicle exhaust, solar radiation) or when at emergency incidents (e.g., agents in fire smoke including polycyclic aromatic hydrocarbon) (2–4). Exposure to these carcinogens may put paramedics at an increased risk of cancer.

Very few studies have focused on cancer risks among paramedics. In a recent national US cross-sectional study of EMS workers, skin (52.4%), prostate (6.4%), and thyroid (4.7%) cancer were the most commonly self-reported diagnosed cancers during their EMS career. Overall cancer rates were slightly higher among ever a firefighter versus never a firefighter (6.7 vs. 5.1%) (5). A recent World Trade Center cohort also reported an increased risk of thyroid cancer overall and among asymptomatic (OR: 2.3 vs. 3.1) emergency medical service workers compared to the general population (6). A study by McClure et al. investigated the impact of using different occupational record linkage approaches to identify cancer risks among Florida firefighters. They evaluated two approaches, the first used a linked database of cancer registries and state firefighter data and the second used occupation data only identified in cancer registries. The second approach included emergency medical technicians and paramedics as Florida firefighters are often certified for these occupations. When comparing the two approaches, differences in cancer risk were observed for endocrine system cancer (OR: 2.08 vs 2.36); respiratory system cancer (OR: 0.73 vs. 0.99); leukemia (OR: 0.66 vs 0.92); and lymphoma (OR: 0.86 vs 0.90), respectively. However, the risk of skin cancer decreased (OR: 1.54 vs 1.06) (4). This study demonstrated the importance of accurate cancer surveillance among emergency service workers through linked administrative data.

Paramedics share some similar carcinogenic exposures with firefighters and police that may put them at risk for cancer; however, less attention has been given to paramedics in occupational cancer research. Due to limited evidence and understanding of cancer risks specifically among paramedics, there is a need to explore this further. This study aimed to examine the risk of cancer among paramedics in a large cohort of Ontario workers.

Methods

Study Population & Linkage Process

The study population and linkage process has been described elsewhere (7–8). Workers were identified from the Occupational Disease Surveillance System (ODSS), a cohort established to examine occupational diseases among Ontario workers. The ODSS was developed by linking work information identified in accepted lost-time workers’ compensation claims from the Workplace Safety and Insurance Board (WSIB) to the Ontario Cancer Registry (OCR), as well as hospital, emergency department, and out-patient visit databases (7–8) (Figure 1).

Figure 1. This flow diagram provides a description of the overall ODSS linkage process. A total of 7240 paramedics were identified in the Workplace Insurance & Safety Board (WSIB) claims data from 1996 to 2019 and incidence of cancer among paramedics were identified through the Ontario Cancer Registry (OCR).

The WSIB covers approximately 70–75% of workers in the province of Ontario (7). A total of 2,387,756 workers in the WSIB data from 1983 to 2019 were eligible for linkage to the Registered Persons Database (RPDB) to obtain participants’ sex, birthdate, death date, emigration out of Ontario, and residence. Any worker missing sex, birthdate, claim date, occupation, and industry information or below the age of 15 years was excluded from the analysis (n = 17,557). A total of 1,964,519 claimants were linked to the RPDB by use of health insurance number (HIN), which is available for all registered Ontarians. Those without a HIN were probabilistically linked to the RPDB by using the name, sex, birthdate, and death date (n = 405,680). A small number of claimants could not be linked due to missing information (n = 1979), resulting in an analytic cohort of 2,368,220 workers.

Occupation was coded by the WSIB using standardized 1992 National Occupational Classification (NOC-1992) codes at the time of the accepted lost-time compensation claim. NOC-1992 codes were available in the WSIB database from 1996 to 2019. Paramedics were defined as any worker classified with a NOC-1992 code of 3234 “Ambulance Attendants and Other Paramedical Occupations” (9). Workers in this group may administer prehospital emergency care to patients, transport patients to/between medical facilities, assist other medical professionals, complete documentation, maintain equipment, and potentially supervise and train other workers. In Ontario, a college diploma, apprenticeship training (2 or more years) or experience in a supervisory occupation are required to be employed as a paramedic. The analytical cohort was further restricted to 1996 to 2019 to represent the study period with available paramedic NOC-1992 codes, resulting in a total of 1,009,783 workers.

The cohort was linked to the Ontario Cancer Registry (OCR) to identify incident cases of cancer that were diagnosed between 1964 and 2020 using deterministic and probabilistic linkage methods (Figure 1). All prevalent cases of cancer, defined as those with a diagnosis date prior to cohort entry on January 1, 1996 (n = 206,514) were excluded from this analysis. Incident cancer cases were coded using the International Classification of Diseases 10th Revision (ICD-10).

Statistical Analysis

Participants were followed for cancer until the date of first diagnosis, emigration out of Ontario, death, age 85 years, or end of study (December 31, 2020). Cox proportional hazard regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for specific cancer diagnoses among paramedics. The hazard ratio is a measure used to compare the risk of an event (cancer) in one group compared to another group (10). If the risk of an event is higher in one group than the other, the hazard ratio will be >1; if the risk is lower in one group than the other, the hazard ratio will be <1; and if the risk is approximately equal in both groups the hazard ratio =1 (10). Models compared paramedics to all other workers in the ODSS. Adjustments were made for age at the start of follow-up, birth year, and sex. Groups with observed cell counts smaller than six are not presented as per organizational and provincial reporting guidelines. Statistical analyses were performed using SAS V9.4.

Results

The cohort included 7240 paramedics and the characteristics of these workers are shown in Table 1. Paramedics share a similar makeup to the overall cohort, with over half identified as males (63.5%).

Table 1. Descriptive characteristics of paramedics and the overall ODSS cohort (1996–2019).

	Paramedics	Overall ODSS Cohort
Total number	7240	1,009,783
Male; %	4597; 63.5%	576,684; 57.1%
Year of birth, median (IQR)	1974 (1964–1982)	1970 (1959–1980)
Years of follow-up, median (IQR)	12.0 (6.0, 18.0)	14.0 (7.0, 19.0)
Age at start of follow-up, median (IQR), years	33.0 (28.0, 40.0)	36.0 (26.0, 47.0)
Age at end of follow-up, median (IQR), years	45.0 (37.0, 55.0)	50.0 (40.0, 60.0)

NOC: National Occupational Classification; IQR: Interquartile Range; n: number of workers; ODSS: Occupational Disease Surveillance System.

Table 2 presents the number of cancer cases and hazard ratios among paramedics. An elevated risk of at least 5% was observed for eleven individual cancer sites (i.e., stomach, colon, rectum, pancreas, melanoma, female breast, prostate, testis, kidney, thyroid, and leukemia); however, statistically significant elevated risks were only observed for risk of any cancer, prostate cancer, and melanoma. A statistically significant reduced risk was observed for lung cancer.

Table 2. Hazard ratios and corresponding 95% confidence intervals for risk of cancer among paramedics compared to all other workers in the ODSS cohort.

		Among all other workers (non-paramedics)	Paramedics only
ICD-10 code	Cancer site	Number of cases	Number of cases	HR (95% CI)
C00–C97	Any Cancer	48,349	282	1.19 (1.06–1.34)
C01–C14	Oral	1129	7	1.02 (0.48–2.14)
C16	Stomach	1027	8	1.46 (0.73–2.94)
C18	Colon	3019	18	1.21 (0.76–1.92)
C20	Rectum	1594	10	1.18 (0.63–2.19)
C25	Pancreas	1195	8	1.43 (0.71–2.87)
C34	Lung	6333	13	0.48 (0.28–0.83)
C43	Melanoma	1947	24	2.18 (1.46–3.26)
C50	Breast (female)	7020	26	1.28 (0.87–1.89)
C61	Prostate	6365	62	1.73 (1.34–2.22)
C62	Testis	623	6	1.50 (0.67–3.35)
C64	Kidney	1906	15	1.38 (0.83–2.30)
C67	Bladder	2339	15	1.22 (0.73–2.02)
C71	Brain	1092	7	1.10 (0.53–2.32)
C73	Thyroid	2746	17	1.15 (0.72–1.86)
C82	Non-Hodgkin’s Lymphoma	2615	15	1.04 (0.63–1.73)
C91	Leukemia	1373	9	1.21 (0.63–2.32)

Adjusted for birth year (linear, continuous), sex, and age at start of follow-up.

- Bone, esophageal, eye, hodgkin’s lymphoma, larynx, lip, liver, myeloma, ovarian, nasal, and male breast cancers are not shown here due to case counts less than 6 which are unreportable.

- Bolding identifies statistically significant results.

- Number of cases are shown for all workers in the ODSS excluding paramedics and for paramedics only.

Sex-specific cancer risks showed consistent findings for male paramedics with the highest risks observed for melanoma (HR 2.71, 95% CI 1.78–4.13), prostate cancer (HR 1.73, 95% CI 1.34–2.22), any cancer (HR 1.22, 95% CI 1.06–1.40), kidney cancer (HR 1.53, 95% CI 0.90–2.59), and bladder cancer (HR 1.33, 95% CI 0.80–2.21) (data not shown in table). A reduced risk was also observed for lung cancer (HR 0.28, 95% CI 0.13–0.63). No associations were observed for any cancer, breast, lung, or thyroid cancer among female paramedics (data not shown in table). Other cancer sites could not be assessed among female paramedics due to low case counts.

Figure 2 reports the current ODSS paramedic hazard ratios alongside previously reported hazard ratios for firefighters and police (analyzed separately from paramedics) for eleven cancer sites and any cancer overall (8). The figure illustrates similar patterns of risk for paramedics as police and firefighters.

Figure 2. Risk estimates (HRs) and corresponding confidence intervals (95% CI) for selected cancers among paramedics in the Occupational Disease Surveillance System cohort are shown, in comparison to police and firefighters in the same ODSS cohort reported in a previous publication (8). The ODSS was established by linking accepted lost-time claimants from the Workplace Insurance & Safety Board (WSIB) to various databases such as the Ontario Cancer Registry (OCR). Occupation was reported at the time of the claim in the WSIB data and was coded using various coding systems. Paramedics were coded using the National Occupational Classification (NOC) system and police and firefighters were coded using the Canadian Classification Dictionary of Occupation (CCDO) system in WSIB data. Incident cancer cases were identified through the Ontario Cancer Registry (OCR) and coded using the International Classification of Diseases 10th Revision (ICD-10).

Discussion

This study aimed to identify cancer risks among paramedics within a large cohort of Ontario workers. Paramedics exhibited statistically significant elevated risks for any cancer, prostate cancer, and melanoma compared to all other workers. These findings align with a previous publication on firefighters and police using the same cohort (8). Although various excesses unique to these groups were observed, there were some striking similarities especially for melanoma, prostate, and lung cancer. Despite the differences in job tasks among these occupations, similar elevated risks may indicate shared exposures, including shiftwork, vehicle exhaust, intermittent solar radiation, and potentially fire smoke contaminants when at fire events (11–17). In comparison to police and firefighters, there were fewer paramedics in the ODSS which represents the shorter follow-up among paramedics.

The strongest association observed among paramedics was an increase in malignant melanoma, which was similar to previous findings among firefighters and police. This is supported by a recent US study where melanoma was the most prevalent self-reported cancer among emergency medical service workers (5). Paramedics may be exposed to solar radiation through open windows or while outside during emergency calls like police/firefighters, which may explain this increased risk (11). To a lesser extent paramedics may be exposed to smoke that may contain polychlorinated bisphenyls (PCBs) or polycyclic aromatic hydrocarbons (PAHs) when at fires providing medical care, which have been associated with melanoma (12–13). Numerous studies have demonstrated a connection between prostate cancer risk and firefighters and police (18–20). The etiology of prostate cancer remains unclear (21); however shiftwork has been suggested as a potential cause, which may explain the increased risk among paramedics. The International Agency for Research on Cancer classified shiftwork as a probable carcinogen, primarily based on increased breast cancer risk among females (14–15). Mechanistic pathways shared between shiftwork and breast cancer risk (nighttime light exposure, potentially lowering melatonin levels and disrupting the circadian rhythm, impacting hormonal balance and DNA damage detection and repair) may also explain the relationship with prostate cancer (22). However, previous studies examining the relationship between shiftwork and prostate cancer have mixed findings (15, 22, 23). Possible exposures at fires may also expose paramedics to heavy metals (i.e., cadmium and arsenic) associated with prostate cancer (16). Paramedics had a reduced risk of lung cancer compared to all other workers in the cohort, despite potential exposure to agents that increase respiratory illnesses, e.g., vehicle exhaust (17). This reduced risk may be attributed to potentially lower cigarette smoking habits compared to other workers, however information on smoking habits is not available in the ODSS. It is possible that paramedics have reduced smoking habits due to the physically demanding nature of their work (24) and the various evaluations they are required to undergo prior to/during employment. Reduced smoking rates have been reported among certain workers, such as firefighters in the United States, when compared to other adult males or similar occupational groups (e.g., military personnel) (25).

Other similarities in cancer risks between paramedics and other emergency workers were observed. Male paramedics demonstrated a slight elevated risk for kidney cancer, similar to firefighters (8). Fire smoke and foams have been linked to kidney cancer (11), but beyond this, it is unclear what exposures among paramedics may increase their risk. Paramedics also had an elevated risk of testicular cancer similar to police (but lower than firefighters). Exposure to firefighting foams, and occupational exposure as a firefighter have been linked to testicular cancer (16). Additional research on potential sources of exposure to explain these excesses in paramedics is needed. Findings underscore the need for cancer surveillance among paramedics. This can be accomplished through integrated systems that link occupational information to health records. Continuous monitoring of cancer risks over time can provide a better understanding of the association between cancer and paramedical occupations, including the potential hazardous exposures involved.

Limitations & Strengths

Job information is only available at a single point in time, limiting the ability to evaluate employment duration or changes in employment over time. The cohort also lacks exposure information, limiting the ability to investigate specific exposure-response relationships. Information on potential non-occupational confounders (e.g., smoking, ethnicity, physical activity, and cancer screening behaviours) is not available in this cohort. Although we have information on sex, female-specific risks could not be examined due to low case counts (<6) which may not be a reliable representation of the true risk among females and this can make it challenging to detect associations. The large cohort size, however, allows for the exploration of occupational groups (that are often understudied) and the examination of multiple cancers. Additionally, the comparability to existing firefighter and police studies supports the possibility of shared exposures among emergency service workers.

Conclusions

This study provides valuable findings regarding cancer risks among paramedics in a large cohort of Ontario workers and adds to the growing evidence among emergency medical services workers. Paramedics constitute a distinct and relatively underexplored group with limited available data on cancer risks and potential occupational exposures. While only a small number of cancers were found to be significantly elevated among paramedics, the results align with other emergency service workers’ findings, emphasizing the importance of investigating potential shared exposures among these workers. Additionally, better resources and education efforts could be provided to increase awareness about the importance of precautions and appropriate controls to reduce carcinogen exposures. This can lead to better understanding of carcinogens and other exposures among paramedics and inform cancer prevention strategies.

Author Contributions

JS provided substantial contributions to the conception of the research, analysis of data, interpretation of the data, drafting of the manuscript and further revisions, gave final approval of the version to be published, and agreed to act as a guarantor of the work. PD contributed substantially to the design of the research, interpretation of the data, revised the manuscript critically, gave final approval of the version to be published, and agreed to act as a guarantor of the work. FE contributed to the acquisition of the data, analysis of the data, revised the manuscript critically, gave final approval of the version to be published, and agreed to act as a guarantor of the work. CB contributed to the acquisition and interpretation of the data, analysis of the data, revised the manuscript critically, gave final approval of the version to be published, and agreed to act as a guarantor of the work. MD contributed to the acquisition of the data, analysis of the data, revised the manuscript critically, gave final approval of the version to be published, and agreed to act as a guarantor of the work. TK contributed substantially to the design of the research, interpretation of the data, revised the manuscript critically, gave final approval of the version to be published and agreed to act as a guarantor of the work.

Ethical Approval

This study was approved by the University of Toronto Health Sciences Research Ethics Board (#39013).

Acknowledgments

We would like to acknowledge Nelson Chong for his support with the ODSS data linkage.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Data Availability Statement

Supporting data is available upon request and can be accessed if conditions are met that comply with organization guidelines.

Additional information

Funding

This work is supported by the Ontario Ministry of Labor, Immigration, Training, and Skills Development (14-R-029). The Occupational Cancer Research Center is supported by the Ontario Ministry of Labor, Immigration, Training, and Skills Development and the Ontario Health agency.