Objectives The goal of this study was to leverage data from two environmental regulatory initiatives, Ontario’s Toxics Reduction Act (TRA) and Canada’s National Pollutant Release Inventory (NPRI), to assess their ability to monitor trends in the use and emission of carcinogens by industry sector in Ontario. Methods Data reported to the TRA and NPRI by industrial facilities in Ontario were retrieved from 2011 to 2015. Twenty-six known and suspected carcinogens were identified (IARC) and the trends in the use and emission were evaluated by industry sector. The locations of industrial facilities that used and released carcinogens were mapped by Public Health Unit (PHU). Results Chemical manufacturing and primary metal manufacturing sectors accounted for 84% of all reported industrial use of carcinogens during the period 2011–2015. The largest source of carcinogen emissions came from facilities in the primary metal manufacturing and paper manufacturing sectors. The largest number of industrial facilities that reported the use and release of carcinogens were located in the City of Toronto and Peel Region PHUs. Overall, the use of carcinogens across all sectors appeared to decrease by 8%, while emissions increased by about 2%. Conclusion The results of this study show the need to reduce the use and emission of select carcinogens in priority industry sectors. Environmental reporting programs, such as the TRA and NPRI, can serve as important tools in cancer prevention by tracking potential carcinogen exposures in the environment and in the workplace. Résumé Objectifs Exploiter les données de deux initiatives réglementaires environnementales, la Loi sur la réduction des toxiques (LRT) de l’Ontario et l’Inventaire national des rejets de polluants (INRP) du Canada, pour estimer leur capacité de surveiller les tendances d’utilisation et d’émission de cancérogènes par secteur d’activité en Ontario. Méthode Nous avons récupéré les données déclarées à la LRT et à l’INRP par les installations industrielles de l’Ontario entre 2011 et 2015. Vingt-six cancérogènes connus et soupçonnés (selon le Centre international de recherche sur le cancer) ont été recensés, et les tendances d’utilisation et d’émission de ces substances ont été évaluées par secteur d’activité. Les emplacements des installations industrielles ayant utilisé et rejeté des cancérogènes ont été cartographiés pour chaque bureau de santé publique (BSP). Résultats Le secteur de la fabrication de produits chimiques et celui des métaux de première fusion ont représenté 84% des utilisations industrielles déclarées de cancérogènes pour la période de 2011 à 2015. Les installations du secteur des métaux de première fusion et les papetières ont constitué la plus grande source d’émissions de cancérogènes. Nous avons recensé le plus grand nombre d’installations industrielles disant avoir utilisé et rejeté des cancérogènes dans les BSP de la Ville de Toronto et de la Région de Peel. Dans l’ensemble, l’utilisation de cancérogènes dans tous les secteurs d’activité semble avoir diminué de 8%, mais les émissions ont augmenté d’environ 2%. Conclusion Les résultats de l’étude montrent qu’il faut réduire l’utilisation et les émissions de cancérogènes ciblés dans des secteurs d’activité prioritaires. Les programmes de rapports sur l’environnement comme la LRT et l’INRP peuvent être d’importants outils dans la prévention du cancer en localisant les expositions possibles aux cancérogènes dans l’environnement et sur les lieux de travail. * Catherine E. Slavik School of Geography and Earth Sciences, General Sciences Building, firstname.lastname@example.org McMaster University, 1280 Main Street West, Hamilton, Ontario L8S 4K1, Canada Dalla Lana School of Public Health, University of Toronto, Occupational Cancer Research Centre, Cancer Care Ontario, Toronto, Ontario, Canada Toronto, Ontario, Canada Can J Public Health . . . . Keywords Environmental carcinogens Manufacturing and industrial facilities Industry Environmental exposure Prevention and control . . . Mots-clés Cancérogènes environnementaux Installations industrielles et de fabrication Industrie Exposition environnementale Prévention et contrôle Introduction publicly and freely available and remain a valuable source of information to support policy-making (Pulles 2008). About one out of every two Canadians will be diagnosed with Chemical exposures in the workplace have been identified cancer in their lifetime (Canadian Cancer Society’sAdvisory as a priority research issue in Ontario that could help to Committee on Cancer Statistics 2017). However, according to inform occupational cancer prevention policies in the prov- the Canadian Cancer Society, approximately half of all cancer ince (Hohenadel et al. 2011). deaths can be prevented (Canadian Cancer Society 2014). In Previous studies using emissions data from environmental Ontario, it is estimated that over 3000 cancer cases could be reporting programs have mostly used emissions data to ex- prevented each year if exposures to carcinogens in the work- plore issues of environmental justice relating to the location place were reduced (Cancer Care Ontario, Occupational of facilities in communities with large populations of minori- Cancer Research Centre 2017), and an additional 1970 cases ties (Neumann et al. 1998; Mennis and Jordan 2005), higher if exposures to industrial carcinogens were reduced (Cancer poverty levels (Ringquist 1997), and lower socio-economic Care Ontario, Ontario Agency for Health Protection and status (Mennis 2002). Other research has used emissions data to examine potential health risks associated with exposure to Promotion 2016). These estimates demonstrate the need to reduce the industrial use and emission of carcinogens in industrial emissions, such as cancer and respiratory conditions Ontario. One approach to achieving reductions constitutes (Chakraboty 2004;Wongetal. 2016). NPRI data have also tracking toxic use and emissions (Jacobs et al. 2014). been used to analyze emission trends and the distribution of Chemical releases have been systematically monitored industrial facilities in neighbourhoods by various socio- across various industrial facilities for some time in Canada, economic characteristics in Montreal and in Toronto (Premji with environmental emissions data going as far back as 1993 et al. 2007; Kershaw et al. 2013). (Environment and Climate Change Canada 2017). Facilities A recent study by Setton et al. analyzed NPRI data for report their emissions of contaminants that may impact hu- each province and found that approximately half of the re- man health and the environment to Environment and ported toxic releases to air came from industrial sources in ClimateChangeCanada’s (ECCC) National Pollutant Ontario, making industries a significant contributor of air Release Inventory (NPRI) (Environment and Climate contaminants in the province compared to other sources Change Canada 2016). In Ontario, the Toxics Reduction such as road emissions (Setton et al. 2015). The only study Act of 2009 (TRA) requires industrial facilities in four major that assessed trends in NPRI chemical releases using an manufacturing and mineral processing sectors to track and industry sector approach was carried out two decades ago report their use of toxic substances (i.e., amount entering the by Olewiler and Dawson, who found that nationwide, the facility, amount created and amount contained in product) to four sectors that released the most toxic substances by the Ontario Ministry of Environment and Climate Change weight were the chemical, primary metal, and petroleum (MOECC) (Government of Ontario 2009). Ontario’sTRA and coal manufacturing sectors, and the mining sector was modeled after Massachusetts’s Toxics Use Reduction (Olewiler and Dawson 1998). In Massachusetts, data from Act of 1989, which has led to significant reductions in the their toxic use reduction program was analyzed and facili- use and release of carcinogens by industrial facilities in that ties in the chemical manufacturing sector were found to be state by 32% and 93%, respectively, from 1990 to 2010 responsible for more than half the total use of all chemicals (Jacobs et al. 2014). Previously, we utilized data from in the state, while no other sectors made up more than 10% Ontario’s TRA to examine carcinogen use and release trends of total use (Commonwealth of Massachusetts 2014). No by cancer site (Slavik et al. 2017). study assessing sector trends in the use and release of indus- Data from these environmental reporting programs can trial contaminants has ever been undertaken in Ontario and aid in assessing exposures in the workplace and in the en- very little is known about the presence of specific carcino- vironment by serving as surrogates for potential exposures. gens in particular industry sectors. The results of this study Reported contaminant emissions can serve as an indicator of will fill a key knowledge gap in our understanding of work- an environmental hazard, while use can indicate a potential er exposures in specific industry sectors in Ontario and help occupational hazard. In most cases, these environmental to identify potential sectors where future cancer prevention programs offer the only source of complete data that is efforts ought to be directed. Can J Public Health Since exposures to occupational carcinogens vary by in- exposure within the workplace, which is also a focus of this dustrial sector (Driscoll et al. 2004), the risk of developing study (Lim et al. 2010). In addition, the data for air emis- cancer associated with specific exposures is often linked to sions are more consistently reported by industrial facilities, particular sectors (Mannetje and Kromhout 2003). resulting in a more complete dataset. Analyzing data using industry sector classifications like the North American Industry Classification System Data analysis (NAICS) can therefore serve as a useful tool for assessing potential occupational exposures to carcinogens and for Industrial facilities in Ontario that report their releases to the directing cancer prevention efforts to priority sectors. Data NPRI must report their use of toxic substances to the from the TRA on the use of carcinogens in various industrial MOECC’s TRA program from a list of substances prescribed facilities can make this type of assessment of exposures by by the program. Of the 322 substances prescribed under the industry sector possible. The goal of this study was to lever- TRA, those classified as known human carcinogens (group 1) age data from two environmental regulatory initiatives, and probable human carcinogens (group 2A) by the Ontario’s TRA and Canada’s NPRI, to assess their ability International Agency for Research on Cancer (IARC) were to monitor trends in the use and emission of carcinogens by selected for analysis in this study (n = 26) (International industry sector in Ontario. In addition, we use the data to Agency for Research on Cancer 2017). The extracted data examine the geographic location of industrial facilities from the 2011–2015 NPRI and TRA datasets were restricted reporting carcinogen use and release in the province. to the 26 identified known and suspected carcinogens accord- ing to their CAS numbers. Although the NPRI program re- quires facilities to report on releases of the particulate matter Methods less than 2.5 μmin diameter (PM ), particulate matter less 2.5 than 10 μm(PM ), and total particulate matter, only releases Databases of PM were selected for analysis in this study as they were reported more frequently. IARC has classified all particulate Data on the use of toxic substances from industrial facilities matter as a known human carcinogen (The International reporting to the Ontario TRAwere retrieved online from the Agency for Research on Cancer (IARC) 2016). MOECC’s website for the years 2011 to 2015 (Government A postal code conversion file program was used to convert of Ontario 2017). Data reported for the year 2010 (the first the postal code of each industrial facility reporting to the TRA year of data collection after the program’s implementation) and to the NPRI into its respective Public Health Unit (PHU), were omitted as the reporting requirements were different in order to geographically analyze the data for industrial facil- compared to subsequent years. The following variables ities in both datasets. There are 36 PHUs in all of Ontario were used: substance name and Chemical Abstract Service (Ontario Ministry of Health and Long-Term Care 2014). The (CAS) number, facility NAICS code, facility postal code, number of industrial facilities that reported the use and the number of employees, and the amount of toxic substances release of carcinogens in 2015, the most recent dataset avail- entering the facility referred to as Bestimated use^ (in tonnes able at the time of this study, was identified for each PHU and [t]). Facilities reported Bestimated use^ ranges (i.e., > 0 to mapped using ArcGIS 10.5.1 (Environmental Systems 1t,> 1to10t,etc...), andthemid-pointofthe rangewas Research Institute 2017). Data was classified using the used for statistical purposes. Bnatural breaks (jenks)^ method to maximize the difference Emissions data from industrial facilities reporting to in numbers of facilities between PHUs on the maps. Canada’s NPRI were also retrieved online from ECCC’s The top 10 industry sectors were ranked by the amount of website as a bulk dataset for data from all facilities reporting known and suspected carcinogens used and released which since 1993 (Environment and Climate Change Canada were summed for all years from 2011 to 2015. The top three 2017). Data was limited to Ontario (2011–2015) and the substances used and released, as well as the mean number of following variables were used: year, substance name and employees, were also identified for each of the top 10 industry CAS number, facility NAICS industry code (31-33, 212), sectors. Additionally, the amount used and released in each facility postal code, number of employees, and the amount sector, and the number of facilities in each sector, was identi- of toxic substances released to air that was calculated as the fied for each year from 2011 to 2015. Carcinogen use was sum of releases to air across multiple variable categories plotted by year for each sector and trends were calculated by (i.e., stack or point releases, fugitive releases, spills, storage dividing the slope of the trend line generated (for each sector’s and handling, and other releases, in tonnes). We restricted use by year) by its intercept and multiplying by 100%. The our analysis to carcinogens released in air as releases to air same method was used to calculate trends for releases. All have been identified as likely having a larger impact on data analyses were performed using the statistical software human health, and inhalation is the primary route of SAS 9.4 (SAS Institute Inc. 2013, Cary, NC). Can J Public Health Results out of any sector (Table 1). Facilities in this sector released approximately 30% of the total carcinogenic emissions from The total estimated use and release of carcinogens by indus- all sectors. Other significant carcinogen emissions were re- trial sector summed for all years from 2011 to 2015, the mean ported by the paper manufacturing and mining sectors, with number of employees for each sector and the top three known 12,130 and 9875 t, respectively (Table 1). and suspected carcinogens by amount used and released are The largest number of workers was employed on average shown in Table 1. in the primary metal manufacturing sector (n = 132,401) Facilities in the chemical manufacturing sector ranked first and the transportation equipment manufacturing sector among all sectors for reported carcinogen use, using more (n = 42,223) (Table 1). Some of the carcinogens reported than 10 million t in the 5-year period analyzed (Table 1). in the greatest quantities among all industry sectors were The primary metal manufacturing and petroleum and coal lead, nickel, and benzene (Table 1). Particulate matter was products manufacturing sectors ranked second and third for reportedly released by facilities in every sector and in the total reported carcinogen use with approximately 4.7 million t largest amounts (Table 1). and 2 million t, respectively (Table 1). The chemical Trends for facilities reporting estimated carcinogen use manufacturing and primary metal manufacturing sectors to- and release, by industry sector and year, are shown in gether accounted for 84% of all carcinogen use across all Tables 2 and 3, respectively. The largest reported increase sectors. in the use of carcinogens from 2011 to 2015 occurred in the Facilities in the primary metal manufacturing sector report- petroleum and coal product manufacturing sector, which ed the release of 23,166 t of carcinogens into the air, the most increased by 14% (Table 2). All other industrial sectors, Table 1 Total estimated use and release of carcinogens and top carcinogens used and released by industrial sector, ranked by use and releases, TRA program 2011–2015, NPRI 2011–2015 Carcinogen use Carcinogen releases Sectors using carcinogens Mean Total estimated Top carcinogens used Sectors releasing Total Top carcinogens employees use carcinogens releases released Chemical manufacturing 12,819 10,468,540 Benzene; vinyl chloride; Primary metal 23,165.75 Particulate matter; 1,3-butadiene manufacturing benzene; nickel Primary metal 132,401 4,749,630 Nickel; benzene; lead Paper manufacturing 12,129.99 Particulate matter; manufacturing formaldehyde; benzene Petroleum and coal 14,891 1,977,480 Benzene; 1,3-butadiene; Mining (except oil 9874.80 Particulate matter; product manufacturing nickel and gas) nickel; lead Mining (except oil 28,461 658,310 Nickel; lead; arsenic Non-metallic mineral 9648.90 Particulate matter; and gas) product benzene; manufacturing formaldehyde Transportation equipment 42,223 205,020 Nickel; hexavalent Petroleum and coal 5718.41 Particulate matter; manufacturing chromium; lead product benzene; nickel manufacturing Paper manufacturing 18,307 28,530 Formaldehyde; lead; Wood product 5571.53 Particulate matter; arsenic manufacturing formaldehyde; benzene Fabricated metal product 10,676 25,140 Nickel; hexavalent Chemical 4416.50 Particulate matter; manufacturing chromium; lead manufacturing benzene; dichloromethane Wood product 4102 9770 Formaldehyde; arsenic; Food manufacturing 3112.15 Particulate matter; manufacturing benzene formaldehyde Machinery 1523 7650 Nickel; lead Transportation 1360.71 Particulate matter; manufacturing equipment tetrachloroethylene; manufacturing trichloroethylene Plastics and rubber 3891 6370 Lead; hexavalent Plastic and rubber 731.92 Particulate matter; product manufacturing chromium; product tetrachloroethylene; 1,3-butadiene manufacturing trichloroethylene Values for estimates of use were rounded to the nearest 10th. Top carcinogens used and released represent the three carcinogens reportedly used and released in the largest quantities, listed in decreasing order Can J Public Health out of the top 10 analyzed, appeared to experience a de- crease in carcinogen use during the study period (Table 2). The largest decreases in carcinogen use were reported by the transportation equipment manufacturing sector at 24%, as well as the primary metal manufacturing and plastic and rubber product manufacturing sector, each at 13% (Table 2). Overall, there was an observed decrease in the use of car- cinogens across all industrial sectors by 8% (Table 2). The paper manufacturing sector reported the largest in- crease in the release of carcinogens from 2011 to 2015, increasing by 24% (Table 3). Smaller increases in releases of less than 5% were observed for the primary metal and food manufacturing sectors (Table 3). For all other indus- trial sectors, out of the top 10 analyzed, an overall decrease in carcinogen release was observed (Table 3). The largest decrease in carcinogen release was reported by the transpor- tation equipment manufacturing sector by 8% (Table 3). Overall, there was an observed increase in carcinogen emis- sions across all industrial sectors by about 2% (Table 3). The releases from the paper manufacturing sector appeared to drive the overall upward trend observed. Of the 326 industrial facilities reporting the use of carcin- ogens in 2015, the City of Toronto PHU and Peel Region PHU contained the largest number of industrial facilities that reported the use of carcinogens, with each containing 34 facilities (Fig. 1). Six hundred and three industrial facilities reported the release of carcinogens in 2015, with the largest number (n = 56) located in the City of Toronto PHU, follow- ed by Peel Region PHU (n = 51) (Fig. 2). However, carcin- ogen use and release in 2015 by volume were highest in Lambton Health Unit (n = 1,343,790 t) and Sudbury and District Health Unit (n = 2650 t), respectively (data not shown). The chemical manufacturing and petroleum and coal product manufacturing sectors are prominent industries in Lambton Health Unit and were likely responsible for the carcinogen use observed in that PHU. In Sudbury, nickel mining remains an important industry contributing to the large quantity of carcinogenic emissions. Discussion Exposures to industrial carcinogens can occur both in the workplace and in communities in the proximity of indus- trial facilities (Hohenadel et al. 2011). This study identi- fied some key industry sectors in Ontario, which have also been identified in similar studies using emissions data as well as in epidemiological studies. For example, studies from the US utilizing data from the Toxics Release Inventory Program have also identified the chemical manufacturing sector as a key industry for its large release of toxic substances (Zhou and Schoenung 2009;US Environmental Protection Agency 2017). The findings by Table 2 Total estimated use and percent change, by industrial sector, ranked by use, TRA program 2011–2015 Sector 2011 2012 2013 2014 2015 Percent change Number of Estimated Number of Estimated Number of Estimated Number of Estimated Number of Estimated facilities use facilities use facilities use facilities use facilities use Chemical manufacturing 47 2,624,120 56 1,630,370 52 2,554,800 53 2,075,200 50 1,584,060 − 6 Primary metal manufacturing 54 1,741,390 57 752,800 53 751,350 49 749,690 51 754,400 − 13 Petroleum and coal product 6 297,750 6 401,850 6 357,850 6 457,290 6 462,740 14 manufacturing Mining (except oil and gas) 17 129,800 18 132,660 18 137,890 17 136,020 16 121,940 − 1 Transportation equipment 52 177,590 47 9690 48 4870 54 6110 54 6760 − 24 manufacturing Paper manufacturing 14 5710 15 5700 15 5710 14 5710 12 5700 < − 1 Fabricated metal product 69 5990 61 5320 63 5060 61 4450 55 4330 − 7 manufacturing Wood product manufacturing 11 2520 9 2530 11 1650 8 1650 10 1420 − 11 Machinery manufacturing 6 1230 6 2310 7 1510 7 1340 6 1260 − 5 Plastics and rubber product 20 1750 20 1870 18 1130 15 530 17 1100 − 13 manufacturing Other 58 2780 49 3420 50 2100 48 1650 49 1690 − 11 Total (all) 354 4,990,630 344 2,948,520 341 3,823,920 332 3,439,640 326 2,945,400 − 8 Values for estimates of use were rounded to the nearest 5th. Percent change was calculated by dividing the slope of the trend line generated (for each sector’s use by year) by its intercept and multiplying by 100% Can J Public Health Table 3 Total releases and percent change, by industrial sector, ranked by releases, NPRI 2011–2015 Sector 2011 2012 2013 2014 2015 Percent change Number of Number of Number of Number of Number of facilities Releases facilities Releases facilities Releases facilities Releases facilities Releases Primary metal 70 4159.00 54 5109.97 66 4217.76 63 4628.18 63 5050.84 3 manufacturing Paper manufacturing 19 1968.17 16 1421.79 17 2918.45 16 2937.59 15 2883.99 24 Mining (except oil 115 1944.85 97 1979.83 98 2172.16 95 2034.64 99 1743.32 − 2 and gas) Non-metallic mineral 106 2078.67 57 1931.92 72 1869.86 71 1891.24 68 1877.21 − 2 product manufacturing Petroleum and coal 27 1328.34 18 1118.28 16 1263.40 15 928.16 17 1080.23 − 5 product manufacturing Wood product 25 1233.26 18 1054.24 22 1038.15 22 1002.29 24 1243.59 < − 1 manufacturing Chemical manufacturing 75 877.04 52 817.66 69 1066.92 68 928.40 61 726.48 − 2 Food manufacturing 80 562.01 59 639.89 76 674.81 71 611.91 68 623.53 2 Transportation equipment 76 371.08 50 233.53 74 293.94 69 210.77 71 251.39 − 8 manufacturing Plastic and rubber 34 160.23 14 172.51 22 130.36 18 132.79 19 136.03 − 6 product manufacturing Other 106 217.35 59 167.16 106 181.80 108 147.77 98 118.10 − 9 Total (all) 733 14,900.00 494 14,646.78 638 15,827.61 616 15,453.74 603 15,734.71 2 Percent change was calculated by dividing the slope of the trend line generated (for each sector’s releases by year) by its intercept and multiplying by 100% Fig. 1 The number of industrial facilities reporting the use of carcinogens in Ontario, by Public Health Unit. Toxics Reductions Act, 2015 Can J Public Health Fig. 2 The number of industrial facilities reporting the emission of carcinogens in Ontario, by Public Health Unit. National Pollutant Release Inventory, 2015 Olewiler and Dawson from Canada indicate that current for some of these differences and allow for a more detailed trends in carcinogen releases in Ontario are similar to na- picture of the life cycle of carcinogen use from entering a tionwide trends in toxic releases from over 20 years ago facility to emission. (Olewiler and Dawson 1998), with the primary metal Other potential discrepancies in the data were due to manufacturing, petroleum and coal product manufacturing changes in the number of facilities reporting to the TRA sectors, and mining sectors still making up the top 5 in- and NPRI. Facilities that do not meet the reporting thresh- dustrial sectors for releases. In a study from the UK, the olds for a given substance no longer need to report, which manufacturing and mining sectors have been previously could reflect some of the annual changes in carcinogen use identified as having a large burden of occupational cancers and release. To accommodate these changes, we used a compared to others (Hutchings et al. 2012). slope to estimate trends in use and emission rather than We observed some notable differences between the re- calculating the differences between the first and last sults for carcinogen use and release, where most sectors reporting years covered by this study. analyzed in this study reported large quantities of carcino- Certain sectors appeared to report more substantial in- gens used and comparatively smaller quantities of carcino- creases in releases, such as the paper manufacturing sector. gens released. Substance use and release for any given Some substances may have driven the use and emission trends sector may not be correlated. Substances that enter an in- observed. For example, releases in particulate matter appeared dustrial facility do not result in the direct release of those to drive the trends in emissions for many sectors, whereas same substances by that facility as some of the substance benzene appeared to drive trends in use (data not shown). A will be contained in the facility’s product, be stored on-site, more detailed analysis of use and emission trends by carcino- or leave the facility as wastes to landfills or releases to gen were previously discussed (Slavik et al. 2017.). These water, etc. Supplementing the dataset of air releases with results illustrate the benefit of taking a sector-by-sector ap- data from facilities’ storage on-site or off-site, recycling proach in gaining a more nuanced understanding of industrial use and emission trends in Ontario. practices, releases to water and other media could account Can J Public Health In addition to analyzing carcinogen use and emission by program staff and errors can be identified and corrected industry sector, this study examined the distribution of before publication of the datasets. Due to the nature of industrial facilities by PHU to explore potential geograph- self-reported data, there may be cases where reported car- ic patterns in industrial sites. The largest number of indus- cinogen use or emissions do not reflect true use or releases trial facilities reporting the use and release of carcinogens by the facilities. were observed in the city of Toronto, Ontario’smostpop- In addition, there are limitations in the reporting re- ulous city. This finding is consistent with previous re- quirements of the TRA program for toxic substance use. search that identified significant sources of industrial pol- The program allows for the reporting of use quantities by lution in Toronto, where neighbourhoods with higher industrial facilities as ranges as opposed to absolute quan- emissions of toxic contaminants were also characterized tities, which limits the analysis to estimates. Another lim- by populations who were racialized or socio- itation of most environmental reporting programs is the economically disadvantaged (Kershaw et al. 2013). In re- fact that only larger industrial facilities meeting specific gions where industrial facilities from specific sectors tend use and release thresholds are required to report (Dolinoy to be concentrated as sector hotspots, certain carcinogens and Miranda 2004), potentially leaving out a significant that are commonly used and released in those industries source of emissions from smaller facilities. Therefore, it are likely to be present. In Sudbury, Porcupine and other is likely that both the use and the emission of industrial health units in northern Ontario where mining industries carcinogens are actually much higher than what is indicat- are concentrated, the industrial use of nickel and other ed by such databases (Simmons 2013). carcinogenic metals is likely. Such regions could benefit There are also limitations in applying carcinogen use from exposure reduction strategies targeted towards partic- and release as indicators of potential exposure since nu- ular industries and carcinogens. merous other factors impact a worker’s exposure to Lung carcinogens are among the most used in industrial chemicals. These other factors, including exposure con- facilities in Ontario (Slavik et al. 2017) and are responsible trols such as safe work practices and personal protective for at least 1395 occupational cancer diagnoses every year equipment, were not examined in this study. It should also (Cancer Care Ontario, Occupational Cancer Research be noted that high sector emissions do not necessarily Centre 2017). Sectors such as the primary metal equate to high cancer risk since contaminants are dispersed manufacturing and transportation equipment manufactur- and are often transported across large distances in the en- ing sectors, which use particularly large quantities of car- vironment (Li and Ma 2016). There are also issues with cinogens associated with these cancers, e.g., nickel, identifying the timing of exposures and the long latency hexavalent chromium, and arsenic, ought to be prioritized period of cancer that make it difficult to link particular cancers to particular carcinogenic exposures (Brody and for exposure reductions. These two sectors also employed the largest numbers of employees, indicating that workers Rudel 2003). in these sectors may be at risk to increased exposures. In addition, since particulate matter has been identified as a major air pollutant contributing to the environmental bur- Conclusions den of cancer in Ontario (Cancer Care Ontario, Ontario Agency for Health Protection and Promotion 2016), ef- Databases like the TRA and NPRI can be used for surveil- forts should be made to reduce these emissions from all lance to provide estimates of industrial carcinogen use and sectors. By shifting towards greener chemistry alterna- release when detailed exposure assessments and routine en- tives, sectors could take steps to reduce, substitute, or vironmental monitoring are not feasible. A major strength eliminate the use and release of hazardous industrial pol- of the approach used in this study is the combination of lutants by altering production processes or redesigning descriptive and quantitative approaches to assess industrial products and systems (Thorpe and Rossi 2007). trends in the province by both characterizing potential ex- posures in sectors of concern and quantifying changes in Study limitations carcinogen use and release. The methods are also highly transferable to other jurisdictions where similar environ- One limitation of this study is that the amount of toxic mental reporting databases exist and concerns over toxics substances used and emitted by facilities is self-reported use and emission are present. in both the NPRI and the TRA, though some consistency There remain opportunities to reduce the use and release between volumes of self-reported pollutant releases by in- of carcinogens in many Ontario industries. The TRA pro- dustrial facilities and pollution monitoring data have pre- gram is still in its infancy and future analyses drawing from viously been found (De Marchi and Hamilton 2006). more datasets will be able to better indicate longer-term Datasets from the NPRI and TRA are validated by trends in carcinogen use and release. The findings from this Can J Public Health carcinogens: assessing the environmental burden of disease at na- study may inform future efforts to quantify levels of expo- tional and local levels. Geneva: World Health Organization. sures in particular industry sectors or geographic regions Environment and Climate Change Canada. (2016). National Pollutant where those industries are present. While this study does Release Inventory. Available at: https://www.ec.gc.ca/inrp-npri/ not attempt to draw conclusions on the risk of developing (Accessed 15 Sep 2017). Environment and Climate Change Canada. (2017). National Pollutant cancer in workers and populations residing near certain in- Release Inventory (NPRI)—Bulk Data. Available at: http://open. dustries, it demonstrates the need to prioritize exposure pre- canada.ca/data/en/dataset/40e01423-7728-429c-ac9d- vention strategies in particular sectors where the most car- 2954385ccdfb (Accessed 15 Sep 2017). cinogens are used and emitted. Government of Ontario. Toxics Reduction Act, 2009, S.O. 2009, c. 19, 2017. Available at: https://www.ontario.ca/laws/statute/09t19 (Accessed 15 Sep 2017). Funding information This research was supported by funds from the Government of Ontario. (2017). Toxics Reduction Act – Reporting. Canadian Cancer Society and by Cancer Care Ontario. Available at: https://www.ontario.ca/data/toxics-reduction (Accessed September 15, 2017). Compliance with ethical standards Hohenadel, K., Marrett, L., Bukvic, D., Brown, J., Harris, S. A., Demers, P. A., et al. (2011). Priority issues in occupational cancer research: Ontario stakeholder perspectives. Chronic Diseases and Injuries in Conflict of interest The authors declare that they have no conflict of Canada, 31(4), 147–151. interest. Hutchings, S. J., Rushton, L., & with the British Occupational Cancer Open Access This article is distributed under the terms of the Creative Burden Study G. (2012). Occupational cancer in Britain: industry Commons Attribution 4.0 International License (http:// sector results. Br J Cancer, 107(Suppl 1), S92–S103. creativecommons.org/licenses/by/4.0/), which permits unrestricted use, International Agency for Research on Cancer. (2017). Monographs- distribution, and reproduction in any medium, provided you give appro- Classifications. Available at: http://monographs.iarc.fr/ENG/ priate credit to the original author(s) and the source, provide a link to the Classification/ (Accessed 1 Sep 2017). Creative Commons license, and indicate if changes were made. Jacobs, M. M., Massey, R. I., Tenney, H., & Harriman, E. (2014). Redu cing the use of carcinogens: the Massachusetts experience. Reviews on Environmental Health, 29(4), 319–340. Kershaw, S., Rinner, C., Gower, S., & Campbell, M. (2013). Identifying inequitable exposure to toxic air pollution in racialized and low- income neighbourhoods to support pollution prevention. References Geospatial Health, 7(2), 265–278. Li, P.-C., & Ma, H.-w. (2016). Using risk elasticity to prioritize risk reduction strategies for geographical areas and industry sectors. J Brody, J. G., & Rudel, R. A. (2003). Environmental pollutants and breast Hazard Mater, 302,208–216. cancer. Environ Health Perspect, 111(8), 1007–1019. Lim, S.-R., Lam, C. W., & Schoenung, J. M. (2010). Quantity-based and Canadian Cancer Society. (2014). About half of cancers can be prevented. toxicity-based evaluation of the US Toxics Release Inventory. Available at: http://www.cancer.ca/en/about-us/for-media/media- Journal of Hazardous Materials, 178(1), 49–56. releases/national/2014/world-cancer-day-2014/?region=on Mannetje, A., & Kromhout, H. (2003). The use of occupation and indus- (Accessed October 1, 2017). try classifications in general population studies. International Canadian Cancer Society’s Advisory Committee on Cancer Statistics. Journal of Epidemiology, 32(3), 419–428. (2017). Canadian Cancer Statistics 2017. Toronto: Canadian Mennis, J. (2002). Using geographic information systems to create and Cancer Society. ISSN 0835-2976. analyze statistical surfaces of population and risk for environmental Cancer Care Ontario, Occupational Cancer Research Centre. Burden of justice analysis. Social Science Quarterly, 83(1), 281–297. occupational cancer in Ontario: Major workplace carcinogens. Mennis, J. L., & Jordan, L. (2005). The distribution of environmental Toronto, ON: Queen’s Printer for Ontario, 2017. ISBN 978-1- equity: exploring spatial nonstationarity in multivariate models of 4868-0415-3. air toxic releases. Annals of the Association of American Cancer Care Ontario, Ontario Agency for Health Protection and Promotion. Geographers, 95(2), 249–268. (2016). Environmental Burden of Cancer in Ontario. Toronto, ON: Neumann, C. M., Forman, D. L., & Rothlein, J. E. (1998). Hazard screen- Queen’s Printer for Ontario. ISBN 978-1-4606-8367-5. ing of chemical releases and environmental equity analysis of pop- Chakraboty, J. (2004). The geographic distribution of potential risks ulations proximate to toxic release inventory facilities in Oregon. posed by industrial toxic emissions in the U.S. Journal Of Environmental Health Perspectives, 106(4), 217–226. Environmental Science & Health, Part A: Toxic/Hazardous Olewiler, N. D., & Dawson, K. (1998). Analysis of national pollutant Substances & Environmental Engineering, 39(3), 559–575. release inventory data on toxic emissions by industry: technical Commonwealth of Massachusetts. (2014). Reporting year 2012: toxics committee on business. Taxation. use reduction information release. Boston: Massachusetts Ontario Ministry of Health and Long-Term Care. (2014). Health services Department of Environmental Protection. in your community. Available at: http://www.health.gov.on.ca/en/ De Marchi, S., & Hamilton, J. T. (2006). Assessing the accuracy of self- common/system/services/phu/ (Accessed 15 Sep 2017). reported data: an evaluation of the toxics release inventory. Journal Premji, S., Bertrand, F., Smargiassi, A., & Daniel, M. (2007). Socio- of Risk and Uncertainty, 32(1), 57–76. economic correlates of municipal-level pollution emissions on Dolinoy, D. C., & Miranda, M. L. (2004). GIS modeling of air toxics Montreal Island. Canadian Journal Of Public Health, 98(2), 138– releases from TRI-reporting and non-TRI-reporting facilities: im- 142. pacts for environmental justice. Environ Health Perspect, 112(17), Pulles, T. (2008). Quality of emission data: community right to know and 1717–1724. national reporting. Environmental Sciences, 5(3), 151–160. Ringquist, E. J. (1997). Equity and the distribution of environmental risk: Driscoll, T., Steenland, K., Nelson, D. I., Prüss-Üstün, A., Campbell- Lendrum, D. H., Corvalán, C. F., et al. (2004). Occupational the case of TRI facilities. Social Science Quarterly,811–829. Can J Public Health Setton, E., Veerman, B., Erickson, A., Deschenes, S., Cheasley, R., Thorpe, B., & Rossi, M. (2007). Require safer substitutes and solutions: making the substitution principle the cornerstone of sustainable Keller, C., et al. (2015). Identifying potential exposure reduction priorities using regional rankings based on emissions of known chemical policies. New Solutions: a Journal of Environmental and and suspected carcinogens to outdoor air in Canada. Occupational Health Policy, 17(3), 177–192. Environmental Health, 14,69–84. US Environmental Protection Agency. (2017). Comparing industry sec- Simmons, G. (2013). Clearing the air? Information disclosure, systems of tors in the 2015 TRI National Analysis. Available at: https://www. power, and the National Pollution Release Inventory. McGill Law epa.gov/trinationalanalysis/comparing-industry-sectors-2015-tri- Journal, 59(1), 9–48. national-analysis (Accessed 15 Sept 2017). Slavik, C., Kalenge, S., & Demers, P. (2017). Recent trends in the indus- Wong, S., Coates, A., & To, T. (2016). Exposure to industrial air pollutant trial use and emission of known and suspected carcinogens in emissions and lung function in children: Canadian Health Measures Ontario, Canada. Reviews on Environmental Health. Survey, 2007 to 2011. Health Reports, 27(2), 3–9. The International Agency for Research on Cancer (IARC). (2016). IARC Zhou, X., & Schoenung, J. M. (2009). Combining US-based prioritiza- monographs: outdoor air pollution Volume 109. Geneva: World tion tools to improve screening level accountability for environmen- Health Organization. Available at: http://monographs.iarc.fr/ENG/ tal impact: the case of the chemical manufacturing industry. Journal Monographs/vol109/mono109.pdf (Accessed 1 Oct 2017). of Hazardous Materials, 172(1), 423–431.
Canadian Journal of Public Health
– Springer Journals
Published: May 29, 2018