Intervening to Reduce the Future Burden of Occupational Cancer in Britain: What Could Work?

Published OnlineFirst September 7, 2012; DOI: 10.1158/1940-6207.CAPR-12-0070

Cancer Prevention Research Article Research

Sally Hutchings1, John W. Cherrie2, Martie Van Tongeren2, and Lesley Rushton1

Abstract In Britain, 14 carcinogenic agents and occupational circumstances currently account for 86% of estimated occupation attributable cancer. The future burden associated with these carcinogens has been forecast, using attributable fractions for forecast scenarios representing patterns of past and predicted future exposure, and exposure levels representing the introduction of new occupational exposure limits, increased levels of compliance with these limits and other reductions in worker exposure. Without intervention, occupational attributable cancers are forecast to remain at more than 10,000 by 2060. With modest intervention over 2,600, or with stricter interventions more than 8,200 cancers could be avoided by 2060 although because of long latency no impact will be seen until at least 10 years after intervention. Effective interventions assessed in this study include reducing workplace exposure limits and improving compliance with these limits. Cancers associated with asbestos, diesel engine exhaust, polycyclic aromatic hydrocarbons, work as a painter, radon, and solar radiation are forecast to continue, with construction remaining the prime industry of concern. Although exposure levels to the established carcinogens are falling, workers are remaining exposed at low levels at which there is still a cancer risk, although the aging population also contributes to rising cancer numbers, These forecasts can be used to assess the relative costs to society of different occupational carcinogenic agents, and the relative merits and savings associated with alternative intervention strategies. The methods are adaptable for different data circumstances, other types of interventions and could be extended to environmental carcinogens and other chronic diseases. Cancer Prev Res; 5(10); 1213–22. Ó2012 AACR.

Introduction scenarios of change for 14 occupational carcinogens and We have estimated that 8% of cancers in men and 2.3% in circumstances in Great Britain that each contribute at least women are caused by work, giving more than 8,000 deaths 100 occupation attributable registrations to current burden and 13,600 cancer registrations in Great Britain (1) for all and account for 86.3% of the total burden (Table 1). occupational carcinogens and occupational circumstances classified by the International Agency for Research on Materials and Methods Cancer (IARC) as Group I (established) or IIA (probable) A full description of the methodology for estimating the carcinogens that had either "strong" or "suggestive" evi- current (4) and future burden (3) of occupational cancer dence of carcinogenicity in humans (2). can be found elsewhere. For our current burden estimation, The methodology has been extended to estimate the Levin’s formula was used to estimate the attributable frac- future burden of occupational cancer and to forecast the tion (AF), that is, the proportion of cases caused by occu- impact of alternative policy decisions affecting future work- pational exposure [(5, AF ¼ p(E) (RR-1)/{1 þ p(E) (RR- place exposure levels (3). This article presents estimates of 1)} in its simplest form]. This requires an estimate of the risk the future burden of occupational cancer under a series of of disease, generally as relative risk (RR) which we obtained from published literature, and the proportion of the population exposed [p(E)], which we derived from national Authors' Affiliations: 1Department of Epidemiology and Biostatistics, data sources, accounting for employment turnover and life Imperial College London, London; and 2Institute of Occupational Medicine, Edinburgh, United Kingdom expectancy, and adjusted for employment trends. To account for cancer latency a risk exposure period (REP) Note: Supplementary data for this article are available at Cancer Prevention Research Online (http://cancerprevres.aacrjournals.org/). was defined for each carcinogen as the exposure period relevant to a cancer appearing in a specific target year Corresponding Author: Sally Hutchings, Imperial College London, Department of Epidemiology and Biostatistics, Faculty of Medicine, St (10–50 years for solid tumors, 0–20 years for lymphohae- Mary's Campus, Norfolk Place, London W23PG, UK. Phone: 44-160-089- matopoetic tumors). As exposure-response risk estimates 0340; Fax: 44-207-594-3196; E-mail: [email protected] and proportions exposed at different levels are not generally doi: 10.1158/1940-6207.CAPR-12-0070 available, risk estimates and proportions exposed were Ó2012 American Association for Cancer Research. obtained wherever possible for "high," "medium," and

www.aacrjournals.org 1213

Downloaded from cancerpreventionresearch.aacrjournals.org on September 28, 2021. © 2012 American Association for Cancer Research. Published OnlineFirst September 7, 2012; DOI: 10.1158/1940-6207.CAPR-12-0070

Hutchings et al.

Table 1. Agents for which future burden has been estimateda

Total current burden Our estimated attributable registrations current (% of total attributable compliance Exposure registrations) Cancer sites WEL/OEL (%) Exposure deﬁned by agent; no appropriate exposure measurements available to use in standard setting ETS 284 (2.1) Lung — PAH as coal tars and 475 (3.5) NMSC — pitches (men only) Radon 209 (1.5) Lung — Solar radiation 1,541 (11.3) NMSC — Occupational circumstance Painters 437 (3.2) Bladder, lung, stomach — Shift work (women only) 1,957 (14.4) Breast — Welders 175 (1.3) Lung — Carcinogenic agents for which exposure standards can be set Arsenic 129 (0.9) Lung WEL ¼ 0.1 mg/m3 99.6 Asbestos 4,216 (31.0) Larynx, lung, Control limit ¼ 0.1 ﬁbers/mL 91.3 mesothelioma, stomach DEE 801 (5.9) Bladder, lung None. Austrian OEL ¼ 99.2 0.1 mg/m3 RCS 907 (6.7) Lung WEL ¼ 0.1 mg/m3 33.0 Strong inorganic 122 (0.9) Larynx, lung No current WEL 96.3 acid mists (was 0.3 mg/m3)b TCDD (dioxins) 316 (2.3) Lung, NHL, STS — Tetrachloroethylene 164 (1.2) Cervix, NHL, esophagus WEL ¼ 345 mg/m3, 100 SCOEL ¼ 138 mg/m3c Total 11,732 (86.3)

aMineral oils which account for a further 12.7% of attributable cancer registrations have been excluded as, because of the changes in the constituents of mineral oils that have occurred in the last few years, it was thought that the future cancer burden would already have been greatly reduced (9). bThis WEL was current in 2007 (10), but is currently under review (11). The estimated compliance indicated is to the WEL current in 2007. cProposed new standard being considered by Scientiﬁc Committee on Occupational Exposure Limits (SCOEL).

"low" exposure levels with a "background" level, where be made, plus Workplace Exposure Limits and current appropriate, assumed to have 0 excess risk [these categories compliance levels to these limits. Full details of the method have been expanded from "high" and "low" only which of adjustment can be found elsewhere (3)]. Where suitable were used to estimate current burden (1)]. Estimated AFs exposure data were not available, RRs could be adjusted to were applied to total British deaths (for 2005) and registra- represent reduced risk scenarios, for example, excess risk tions (for 2004) to give attributable cancer numbers. was reduced successively by 25% per decade for painters, or To estimate future burden AFs were estimated for a series workers could be shifted arbitrarily from higher to lower of forecast target years (FTY), that is, 2010, 2020, ..., 2060 risk categories, for example, shift workers at risk of breast (3). A REP projected forward in time was deﬁned for each cancer were moved from longer to shorter duration of FTY with the contribution of past exposure to future cancer exposure. risk decreasing for each FTY (see Supplementary Fig. S1). For the current burden estimate for mesothelioma Adjustment factors were applied to newly recruited workers uniquely associated with asbestos exposure we used num- (assumed to be aged 15–24 years) in separate 10-year bers directly from the UK register of mesotheliomas for estimation intervals to adjust for changing numbers used 2005, as we believe this captures practically all cases in Great in broad industry sectors, for example, an increase in the Britain and is, therefore, more appropriate for use as a basis service industry sector and a decrease in manufacturing for burden estimation for this disease than our standard industry [based on Labour Force Survey data (6), Supple- methods, which greatly underestimate current mesothelio- mentary Fig. S2]. Where data were available adjustment was ma incidence. Asbestos related lung cancer was also esti- also made for declining exposure levels [see Supplementary mated for current burden from mesothelioma register num- Table S2 for estimated data on exposure levels and future bers using an assumption of a 1:1 ratio (1). We excluded annual declines in levels required for these adjustments to only a small number (30–70 men and women a year in the

1214 Cancer Prev Res; 5(10) October 2012 Cancer Prevention Research

Forecasts of the Future Burden of Occupational Cancer

UK) of background cases (spontaneous or from naturally excess risk, so an estimate of the threshold level for back- occurring asbestos) from the register numbers. For future ground exposure was obtained from independent data. For burden, we have allocated the current attributable meso- DEE, 0.001 mg/m3 as elemental carbon was chosen to theliomas to 3 exposure levels, high and medium for reflect exposure levels in daily life, based on background occupational exposure and a low category for domestic urban and suburban exposure measurements in Britain exposure and environmental exposure believed to have (12). For asbestos an upper boundary of 0.00001 f/mL was been experienced by the post 1940s birth cohort and assumed for background exposure, based on urban expo- additional to the background rate (see footnotes to Sup- sure levels from which mesothelioma cases considered to be plementary Table S1 for the details of the methodology). caused by "background" exposure may arise (13). Total AFs for cancer sites associated with multiple expo- For occupational circumstances such as painters and sures have been estimated using the product equation welders, where no specified carcinogen has been identified, [AFSi ¼ 1 Pi(1AFi)] for independent multiplicative only a single RR was available. A decline in exposure level estimates (7). To estimate attributable cancer numbers the has therefore been assumed to translate linearly to a fall in forecast AFs were applied to an estimate of total cancers for excess risk. This approach was also adopted for dermal that site based on current age-specific rates applied to Great exposure to polycyclic aromatic hydrocarbon (PAH) in coal Britain population projections (Supplementary Table S3). tars and pitches for which no exposure level RRs were For mesothelioma however associated only with asbestos available. For shift work, limits on the total time spent on exposure recent projections based on past mortality rates night shifts over a lifetime were used as the intervention. were used (8). Where levels of exposure were not amenable to WEL setting, proportions of the exposed workers were moved Carcinogens and occupational circumstances included to lower exposure categories (e.g., solar radiation) or a in the estimates reduction in total numbers exposed (e.g., radon) was used Table 1 gives the carcinogens and occupational circum- in the forecasting. For environmental tobacco smoke (ETS), stances from the current burden estimation for which future the effects of different levels of compliance to the current cancer burden was estimated, and the cancer sites that indoor smoking bans were tested. were affected. The carcinogens have been categorized as: (i) those for which no appropriate exposure measurements were available to use in standard setting; (ii) exposures Results defined by occupational circumstance; and (iii) carcino- Table 2 gives for all the carcinogens and occupational genic agents for which standards exist or can be set. circumstances tested the attributable numbers of cancer registrations for 2010 and 2060 together with the numbers Choosing scenarios of cancer registrations avoided in 2060 for each of the Supplementary Table S4 gives the scenarios tested for scenarios in Supplementary Table S4. Attributable cancer each carcinogen and occupational circumstance. Unless registrations are shown per year for each target year. Sup- otherwise stated 2 baseline scenarios have been evaluated: plementary Table S5 also gives AFs and combined results by baseline scenario 1 historic employment and exposure level cancer site across the 14 carcinogens or occupational cir- trends until 2010, no change thereafter, and baseline trend cumstances. Full results can be found elsewhere (14). scenario 2 historic and predicted employment and exposure As historic and forecast exposure levels decline, the AFs trends included up to 2030, constant thereafter. Interven- generally decline for baseline scenarios 1 and 2 to about tion scenarios have been compared with baseline 1. 1.5% of all cancer by 2060 (Table 2, and for example Figs. For those agents where standards can be set the scenarios 1A(i) and A(ii) for DEE and lung cancer). However, for test the introduction of or reductions in current occupa- breast cancer associated with shift work, rising employment tional exposure limits (OEL) and improved compliance to in service sector industries (Supplementary Fig. S2) leads to these standards. For arsenic and tetrachloroethylene the rising occupational AFs [Fig. 1B(i)]. existing workplace exposure limits (WEL), and for strong Predictions of total Great Britain cancers taking account inorganic acid mists an earlier WEL, were much greater than of only demographic changes leads to increasing attribut- estimated current average exposure levels (Supplementary able occupational cancer numbers because of the aging and Table S2). For these and also for TCDD, the estimated increasing population [Figs. 1A(ii) and B(ii)]. Between boundary level between the 2 lowest exposure categories 9,900 (scenario 2) and 10,500 (scenario 1) occupational was used as a starting point for a possible exposure standard cancers can be expected per year by 2060 from the baseline (Supplementary Table S1). For TCDD the low/background scenarios, not much lower than the numbers attributed to boundary was used, representing a threshold below which occupation in 2010. Breast cancer and nonmelanoma skin excess risk for the agent was 0 (background exposed). No cancer (NMSC) from sun exposure account for 60%–70% such threshold is generally recognized for genotoxic carci- of these (Table 2). nogens, so the high/low- or medium/low-boundary level Without new intervention strategies, the numbers of was chosen for the other 3 substances. cancers from low-level exposure will continue to increase For asbestos and diesel engine exhaust (DEE), no indus- even though exposure levels are forecast to decline (scenario tries were categorized as background exposed with zero 2). Figs. 2A and B illustrate this for lung cancer and DEE,

www.aacrjournals.org Cancer Prev Res; 5(10) October 2012 1215

Hutchings et al.

Table 2. Total forecast cancers attributable to leading occupational carcinogens, 2060

Attributable numbers of cancer registrations Attributable numbers of cancer registrations avoided a by 2060

Scenariob: All (1) (2) (3) (4) (5) (6) (3) (4) (5) (6) Exposure Cancer site 2010 2060 2060 Exposure deﬁned by agent; no appropriate exposure measurements available to use in standard setting ETS Lung 1,470 0 0 68 158 68 158 PAHs—coal tars NMSC 489 805 883 606 479 437 405 200 327 368 401 Radon Lung 218 379 411 341 318 310 190 38 61 69 189 Solar radiation NMSC 1,751 3,096 3,307 2,574 2,048 1517 165 522 1,048 1,580 2,931 Occupational circumstance Painters Bladder, lung, stomach 455 639 645 480 380 347 321 159 259 292 318 Shift work Breast 1,660 3,111 3,904 2,169 1,198 197 0 942 1,913 2,914 3,111 Welders Lung 190 141 64 106 84 77 71 35 57 64 71 Carcinogenic agents for which exposure standards can be set Arsenic Lung 128 93 47 91 89 88 88 1 4 5 5 Asbestos Larynx, lung, mesothelioma, 3,841 268 271 264 264 262 243 4 4 6 24 stomach DEEc Bladder, lung 380 409 402 454 415 377 35 (410) 0 0 32 374 (0) Silica Lung 837 799 446 102 50 22 10 697 750 778 789 Strong acids Larynx, lung 122 39 7 18 12 10 10 21 27 29 29 TCDD Lung, NHL, STS 283 8 0 4 4 4 4 4 4 4 4 Tetrachloroethylene Cervix, NHL, oesophagus 139 136 120 124 120 118 118 12 16 18 18 Total 11,663 10,472 9,880 7,379 5,601 3,755 1,657 2,616 4,446 6,125 8,223

aRelative to baseline scenario (1). Except for ETS, negative results, where the intervention has increased forecast cancer numbers [e.g. for DEE scenario (3) 90% compliance is lower than the current estimated compliance to the proposed standard], have been set to zero. All negative results are excluded from the total estimates. bScenarios are as described in Supplementary Table S4. cResults in brackets for DEE are for the additional intervention scenario 6a.

where, although exposure levels are falling by an estimat- other asbestos-related cancers as, in the absence of a suitable ed 7.4% a year, substantial proportions of the population risk estimate, zero excess risk has been assumed other than remain exposed at low levels of exposure which still carry for mesothelioma at this low (nonoccupational) level. asmallexcessrisk(RR¼ 1.1). Introducing an exposure [Supplementary Figs. S3A–F and G–L show results for lung standard of 0.1 mg/m3 and assuming even 99% compli- cancer and mesothelioma]. Even if exposures could be ance does not improve on this (Figs. 2C and D, scenario reduced to the levels indicated by the strictest scenario 6a in Supplementary Table S4). In contrast if an exposure (6) tested, that is, to below 1/100th of the existing standard, standard could be introduced for DEE at the estimated some mesotheliomas remain in 2060 (61 men and 182 level below which excess risk was 0, that is, 0.001 mg/m3 women), because of continued exposure at the level of as elemental carbon [scenario 6, Figs. 2E and F], lung additional background risk estimated for the 1940s birth cancers induced by DEE would nearly disappear by 2060 cohort, above the low threshold at which it is believed [Fig. 1A(ii)]. For this level of reduction, however, tech- excess risk will be zero (13). However, these forecasts do nology driven intervention may be the only realistic not take account of any change to the background risk solution. among later birth cohorts. Our forecast mesothelioma Low-level exposures will also continue to give high fore- estimates for women are higher than for men because of cast numbers of asbestos related mesotheliomas for both the higher RRs for medium and low exposure used in their baseline scenarios (63 in men and 208 in women for calculation (Supplementary Table S1); they are given sep- baseline trend 2) because of the increasing proportion arately as we have less conﬁdence in the results for women exposed at low levels, even though a 13% annual reduction because of the small number of cases on which the risk in average exposure levels has been assumed, and the estimates were based. mesothelioma projections used to estimate our forecast For solid tumor cancers for which long latencies are numbers have also taken falling exposure levels into assumed, no difference is seen between any of the inter- account (8). However, this is not the case for lung and the ventions and the baseline scenarios before 2030. The

1216 Cancer Prev Res; 5(10) October 2012 Cancer Prevention Research

Forecasts of the Future Burden of Occupational Cancer

Figure 1. Results for baseline and intervention scenarios for (A) lung cancer attributable to DEE exposure (men plus women) and (B) breast cancer attributable to night shift work (women only), in terms of (i) attributable fractions and (ii) cancer registrations.

www.aacrjournals.org Cancer Prev Res; 5(10) October 2012 1217

Hutchings et al.

Figure 2. Proportions exposed to DEE and occupation attributable registrations for lung cancer assuming (A and B) linear employment trends and 7.4% annual exposure level decline to 2021–30, (C and D) introducing an exposure standard of 0.1 mg/m3 from 2010 with 99% compliance, and (E and F) introducing a much stricter exposure standard of 0.001 mg/m3 from 2010 with 90% compliance, by achieved exposure level in the forecast target year, men and women together.

intervention scenarios were developed to be progressively as standards are tightened or exposure is progressively more effective leading to progressive reductions in AFs and reduced (scenarios 4 and 5, Table 2). attributable cancers. Together the minimum interventions The scenarios with the most extreme intervention (sce- proposed in scenario 3 (the ﬁrst and least restrictive scenario 5 for most of the chemical agents; scenario 6 for nario) would avoid over 2,600 cancers a year by 2060, asbestos and DEE, shift work and solar radiation, and for although only about 38 of these are for the chemicals radon, painters, welders, and coal tars and pitches) show (arsenic, acid mists, TCDD, and tetrachloroethylene). The how close to zero the attributable cancer numbers might current (0.1 f/mL) standard for asbestos or a proposed (0.1 realistically be expected to fall (Table 2). Numbers fall to 0 mg/m3) standard for DEE do not result in any "avoided" only if an intervention results in all workers moving to cancers by 2060 (current compliance to these standards, exposure categories with zero excess risk, for example, shown in Table 1, exceeds 90%). For respirable crystalline achieving full compliance to no smoking in workplaces. silica (RCS) an improvement in compliance to the 0.1 mg/ Comparing the results for scenario 5, where excess risk has m3 8 hour time weighted average OEL from 33% to 90% been reduced by 25% in successive decades for painters, could lead to nearly 700 fewer lung cancers annually by welders, and coal tars and pitches, and (6) where a halving 2060 (Table 2). The number of cancers avoided increases of risk is achieved in the ﬁrst decade, indicates the

1218 Cancer Prev Res; 5(10) October 2012 Cancer Prevention Research

Forecasts of the Future Burden of Occupational Cancer

importance of early versus delayed intervention. Halving scenario 6a) only avoids an additional 109 cancers by the proportions exposed in workplaces to radon in 2010 2060 (including 92 lung cancers from RCS exposure). (scenario 6), for example by introducing appropriate tech- Forecasts for industry sectors with a current estimate nology, is far more effective than the gradual reduction (2004; ref. 15) of more than 80 attributable cancers are shown in the other interventions. Similarly, achieving a given in Table 3 and Supplementary Table S6. The ranking reduction in risk from solar radiation to that associated with of the predicted cancers by industry sector in 2060 (scenario mixed indoor and outdoor exposure (RR ¼ 1.01, Supple- 2) remains similar to that in 2010 with construction mentary Table S1), for example, using appropriate skin remaining the most important industry sector for potential protection measures, removes most of the large numbers risk reduction targeting (21% in 2060), followed by 3 of predicted NMSCs. Restricting women to a maximum of 5 service industry sectors, and with breast cancer associated years on night shift work, for which the epidemiological with night-shift work across all industry sectors also a evidence suggests excess risk is 0, would eliminate breast leading contributor. By 2060, large numbers of workers are cancers attributable to this exposure [Fig. 1B(ii)]. still projected to be exposed at low levels to the relevant Our testing scenarios assume that compliance to expo- carcinogens (DEE and asbestos, plus tetrachloroethylene in sure standards is less than 100%. Our results indicate that a dry cleaners) in personal and household services and land large reduction in number of cancers can be achieved with transport. In land transport more than two-thirds, and in 90% compliance to a current or proposed standard (sce- defense (armed forces) nearly all attributable cancers are nario 3), and that 99% compliance (arsenic, RCS, strong forecast to be NMSCs because of high level (outdoor) sun acids, TCDD, tetrachloroethylene, scenario 6, and DEE exposure (Supplementary Table S6).

Table 3. Forecast cancersa attributable to leading occupational carcinogens, by industry currently estimated with over 80 attributable registrations, ordered by baseline scenario (2) forecasts for 2060

Attributable numbers of cancer registrations

Scenariob: All (1) (2) (3) (4) (5) (6) Industry/occupation 2010 2060 Construction 4,680 2,450 2,200 1,550 1,260 940 130 Painters and decorators (construction) 340 560 610 420 330 300 280 Roofers, road surfacers, Roadmen, Paviors (Construction) 480 800 880 600 480 430 400 Shift work (across all industries/occupations)c 1,650 3,090 3,880 2,160 1,190 200 0 Land transport 420 560 660 520 440 370 30 Public administration and defence 340 580 660 490 400 300 30 Personal and household services 400 230 240 230 230 200 140 Sanitary and similar services 90 160 200 140 110 90 10 Recreational and cultural servicesd 100 130 160 110 100 70 10 Wholesale and retail trade and restaurants and hotels 660 130 150 150 180 110 70 Farming 320 170 120 120 70 10 10 Financing, insurance, real estate and business servicesd 190 70 80 60 60 60 40 Welders 190 140 60 110 80 80 70 Painters (not construction) 110 80 30 60 50 50 40 Mining 130 40 20 40 30 20 10 Non-ferrous metal basic industries 80 40 10 30 30 30 30 Manufacture of transport equipment 190 20 0 10 10 10 10 Manufacture of industrial chemicals 100 10 0 10 10 10 0 Manufacture of other chemical products 100 10 0 10 0 0 0

aTotals may differ from main tables as agents are summed (product sums) separately by industry and other subgroups. bScenarios are as described in Supplementary Table S4. cShift workers may be employed in all industries/occupations, but the cancer site involved (breast) does not overlap with cancer sites. associated with other carcinogenic agents or occupations to which these workers may also have been exposed. dThese industry sectors had less than 80 attributable registrations in the current burden estimates (2004), but are included as at least 80 are forecast for 2010. Iron and steel basic industries, manufacture of instruments, photographic and optical goods and of non- electrical machinery, metal workers and printing, publishing and allied industries had at least 80 current burden attributable registrations, but are excluded as these were predominantly due to exposure to mineral oils.

www.aacrjournals.org Cancer Prev Res; 5(10) October 2012 1219

Hutchings et al.

Discussion It is also possible that the asbestos-related lung cancer to Our results have shown the potential for considerable mesothelioma ratio is higher than the 1:1 we have assumed. eventual reduction in future occupationally related cancers By estimating AFs on proportions exposed from the UK through a range of interventions, although the long legacy study and lung cancer RRs, Supplementary Table S1, there of past exposures will continue for up to 50 years. Even with would be 5,194 attributable lung cancers rather than 1,768 the most stringent scenario tested, cancers are forecast to in 2010 falling to 13 (not 5) in 2060 for baseline scenario 1. continue because of exposure to asbestos, PAHs as coal tars This would represent a current lung:mesothelioma ratio of and pitches, work as a painter, and exposure to radon and about 3:1 in men and 6:1 in women. Similar ratios have solar radiation, with construction remaining the prime been observed in asbestos exposed cohorts elsewhere (17). industry of concern. Expected increases in cancer in general However, using an alternative, lower estimate of the pro- as the population ages contribute to the continuing high portions exposed to asbestos (still higher however than the levels of some occupational cancers, and predicted increases CAREX estimates), based on the mesothelioma attributable in numbers working particularly in service sector industries fractions derived from CAREX data and the RRs in Supple- also makes a contribution, for example to forecasts for shift mentary Table S1 but that have then been uprated to work breast cancers, exposure to solar radiation, DEE and mesothelioma register numbers, gives more modest esti- asbestos. In estimating the future burden of occupational mates, of 1984 attributable lung cancers in 2010 falling cancer, we have included the top 14 carcinogenic agents and to 5 in 2060 for baseline scenario 1. This would represent occupational circumstances, which account for 86% of the a current lung:mesothelioma ratio of 1.2:1 in men and estimated current burden of occupational cancer in Britain. 0.3:1 in women. Forecasts for agents currently contributing a further 1,800 Pragmatic approaches have been developed to take cancer registrations, including mineral oils, chromium VI, account of limitations in the data available for Britain. cobalt, aromatic amines and inorganic lead, nonarsenical Alternative exposure levels or employment trends and dif- insecticides, work as a hairdresser or barber, soots and wood ferent risk estimates could be used for other countries and dust exposure, and other agents currently responsible for situations and variations in existing standards can readily be fewer cancers in Great Britain but classiﬁed by IARC as explored. Group I carcinogens [including benzene, benzo(a)pyrene Only limited intervention options were tested in this (PAH), beryllium, cadmium, formaldehyde, occupational study, for example, reducing workplace limits and improv- exposure during iron and steel founding, leather dust, ing compliance with these limits. Many other potentially nickel compounds, and rubber manufacturing] have not effective interventions have not been assessed, such as been included in the projection, but are equally important improving technology, increasing awareness, and changing for cancer prevention. If these had been included, 14% attitudes and behaviors that are important in exposure more occupational attributable cancers (an additional control and risk reduction. Translating these interventions 1,600 a year) might be forecast (proportionately) by into testable scenarios is problematic; our solution has been 2060 without intervention, with about 500 of these avoided to show the impact of the results that might be achieved, with some minimum intervention as described for the such as reduction in excess risk or shifts to lower exposure estimated agents. categories. We have shown that intervening to reduce expo- The contribution to the future total burden of large sure to workplace carcinogens could lead to the avoidance numbers of workers exposed at low levels within several of more than 8,200 cancers per year by 2060. In compar- service industries is highlighted, rather than the current ison, with nearly 20% of all cancers (excluding NMSC) more highly exposed manufacturing industry sectors, where currently attributed to smoking (18), about an 8% imme- interventions appear to be more effective in transferring diate reduction in smoking levels would be required to workers from high to low exposed groups. For asbestos and avoid the same number of tobacco-related cancers. The DEE in particular, although exposure levels have been removal of a workplace carcinogen entirely is of course the declining cancers still occur because of the low thresholds most effective possible intervention, by replacement with a below which it is thought that excess risk disappears. less toxic or nonchemical means of fulﬁlling the same If numbers exposed from CAREX had been used to function, for example, for tetrachloroethylene. In this case, estimate cancer caused by asbestos exposure, lung cancer only the legacy of past exposures will remain. would have been underestimated possibly 12-fold and The level of compliance to future OELs, that is the pro- mesothelioma 2-fold for men and women compared with portion of worker-exposures remaining above these limits, observed UK mesotheliomas(14).Thissuggeststhat has also been tested. Full (100%) compliance cannot in numbers exposed to asbestos are underestimated by practice be used with the lognormal distribution assumption CAREX; our forecasts based on observed mesothelioma and testing compliances approaching 100% gives unrealis- cases take this into account. CAREX-based estimates of tically low results; as compliance approaches 100%, even numbers occupationally exposed, 2.6% of men and 1.5% though the standard may be well above the zero risk of women, contrast with estimates of 65% of men and threshold, the distribution mean and proportions exposed 23% of women based on the population controls in the above any zero risk threshold approach zero. In general, UK study from which we have drawn risk estimates for 90% compliance has been assumed to represent a realisti- mesothelioma (16). cally achievable target. Testing the timings of the

1220 Cancer Prev Res; 5(10) October 2012 Cancer Prevention Research

Forecasts of the Future Burden of Occupational Cancer

introduction of standards and also the effect of different estimates may be considered to be inflated either (i) as the compliance levels in different industry sectors or sizes of population is aging and numbers are increasing when industry has been explored. For RCS in Britain, it has been employment levels and, therefore, exposed numbers are shown that improvement in compliance in small companies declining or (ii) as other causal factors decline so that and among the self-employed is more effective at reducing occupational agents operating synergistically with an envi- lung cancer than reducing the current standard (3). ronmental or lifestyle factor (e.g., asbestos and smoking for Where the current standard was found to exceed the mean lung cancer) produce fewer cancers. The effect is illustrated of current exposures by up to 2 orders of magnitude, testing in Fig. 3 for forecast lung cancers attributable to the expo- values at a half or even a quarter of the standard does not sures contributing at least 100 cancers (Table 1), estimated really inform risk reduction strategies, as the estimated using cancer projections based on no change from 2005, proportions exposed at high levels under the test scenario demographic change only and increase to 2030 based on an will unrealistically exceed the proportions exposed in the age-period-cohort modeling approach (19). mid 1970s at those levels, leading to increased AFs and All results presented here are subject to the biases to negative estimates of cancers "avoided." Although this can which our estimates of the current burden of occupation- be addressed by assuming compliance levels stricter than al cancer are subject, described elsewhere (4). The most currently achieved estimates, we have tested standards that important of these are data-based, particularly relating to are less than the current estimated mean levels of exposure the matching of RRs to our allocation of industries to and therefore of more interest, although these may be exposure level categories, and the reliability of the data difficult to achieve in practice. contributing to estimates of numbers ever exposed. Some If there are several risk factors contributing to the burden reallocation of industries between exposure categories has of a disease, a change in attribution for one factor will result occurred between the estimation of current and future in a change in the attribution of the others. For example, if burden where additional categories have been intro- future smoking–related lung cancer falls giving a reduced duced. In particular moving large numbers in construc- nonoccupational AF, the relative importance of occupation tionandlandtransportfrom"high"toanew"medium" as a risk factor could increase leading to a rise in the exposed category for DEE has resulted in much lower occupational AF, although this AF would now be applied numbers of lung cancers attributable to DEE than esti- to lower projected lung cancer numbers. Attributable num- mated for current burden, as a reduced RR has been used bers rather than AFs, therefore, represent a more useful for the medium exposed. For bladder cancer, the current estimate of the future cancer burden caused by occupation. burden "high" exposed RR was retained for this large In addition, it is for this reason that estimated future group and a more specifically targeted higher RR has been occupational AFs have been applied to estimates of future used for the new and smaller high exposed group of cancer numbers based on current cancer rates applied to miners and services allied to transport. Also, introducing projected population estimates, ignoring future changes in a low "nonoccupational" category for asbestos exposure other lifestyle or environmental risk factors. Cancer num- does reduce the forecast estimates for the asbestos-related bers attributable to occupation are then comparable by rank cancers other than mesothelioma; as no risk estimates order between agents and industries. However the actual were available for this group, zero excess risk was assumed

Figure 3. Lung cancers in men caused by occupational exposures, various cancer projections.

www.aacrjournals.org Cancer Prev Res; 5(10) October 2012 1221

Hutchings et al.

for the workers moving out of the higher occupational Authors' Contributions risk categories with our estimated annual fall in work- Conception and design: S.J. Hutchings, J.W. Cherrie, M.V. Tongeren, L. Rushton place exposure levels. Development of methodology: S.J. Hutchings, J.W. Cherrie, M.V. In summary, comparison of a range of interventions for Tongeren, the most important current occupational carcinogens has Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.W. Cherrie, M.V. Tongeren, L. Rushton shown the potential for future reduction of occupationally Analysis and interpretation of data (e.g., statistical analysis, biosta- related cancer. Interventions to reduce exposure to carcino- tistics, computational analysis): S.J. Hutchings, M.V. Tongeren, L. Rushton gens may often also lead to reductions in other health Writing, review, and/or revision of the manuscript: S.J. Hutchings, M.V. related conditions in the working and living environment, Tongeren, L. Rushton e.g. reduction of silica exposure will not only reduce lung Study supervision: L. Rushton cancer but will affect respiratory function and other non- Acknowledgments malignant respiratory diseases. Our methods can be The authors thank the participants of the future burden methodology adapted for different data circumstances, to investigate workshop, particularly Drs. John Hodgson, David Kriebel, Hans Kromhout, other types of interventions and could be extended to Damien McElvenny, Kyle Steenland, Kurt Straif, and organizer Gareth Evans. The contributions and advice from the Health and Safety Executive and the environmental carcinogens and other chronic diseases. rest of the project team is gratefully acknowledged, in particular Andy Although the forecasts presented here are for Britain, the Darnton for his advice on the issues surrounding asbestos. methods have been used to test the impact of introducing alternative exposure standards in the countries of the Euro- Grant Support The work was supported by the UK Health and Safety Executive (grant pean Union for 25 chemical occupational carcinogens (20), number JN 3117). and are readily transferable to other national and occupa- The costs of publication of this article were defrayed in part by the pay- ment of page charges. This article must therefore be hereby marked adver- tional settings. tisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Disclosure of Potential Conﬂicts of Interest Received May 8, 2012; revised July 11, 2012; accepted July 31, 2012; No potential conﬂicts of interest were disclosed. published OnlineFirst September 7, 2012.

References 1. RushtonL,BaggaS,BevanR,BrownTP,CherrieJW,HolmesP, 12. Adams H, Nieuwenhuijsen MJ, Colville RN, Older MJ, Kendall M. et al. Occupation and cancer in Britain. Br J Cancer 2010;102: Assessment of road users' elemental carbon personal exposure levels, 1428–37. London, UK. Atmospheric Environment, 2002;36:5335–42. 2. Siemiatycki J, Richardson L, Straif K, Latreille B, Lakhani R, Campbell 13. Health and Safety Executive. Amendment to the control of asbestos at S, et al. Listing occupational carcinogens. Environ Health Perspec work regulations 1987 and ACOP, Regulatory Impact Assessment, 2004;112:1447–60. 2002. Available from: http://www.hse.gov.uk/ria/chemical/ria-176.pdf. 3. Hutchings S, Rushton L. Towards risk reduction: predicting the Accessed 7/03/2006. future burden of occupational cancer. Am J Epidemiol 2011;173: 14. Hutchings S. The Burden of Occupational Cancer in Great Britain. 1069–77. Technical Report: Predicting Future Burden–Results 2012. Available 4. Hutchings S, Rushton L. The burden of occupational cancer in from http://www.hse.gov.uk. Forthcoming 2012. Britain: statistical methodology. Br J Cancer 2012;107 Supp 1: 15. Hutchings S, Rushton L. The Burden of Occupational Cancer in Britain, S8–S17. industry sector results. Br J Cancer 2012;107 Supp 1:S92–S103. 5. Levin M. The occurrence of lung cancer in man. Acta Unio Internatio- 16. Rake C, Gilham C, Hatch J, Darnton A, Hodgson J, Peto J. Occupa- nalis Contra Cancrum 1953;9:531–41. tional, domestic and environmental mesothelioma risks in the British 6. Ofﬁce for National Statistics, Labour Force Survey. Available from: population: a case–control study. Br J Cancer 2006;100:1175–83. http://www.statistics.gov.uk/ (Accessed June 2010). 17. McCormack V, Peto J, Byrnes G, Straif K, Boffetta P. Estimating the 7. Steenland K, Armstrong B. An overview of methods for calculating the asbestos-related lung cancer burden from mesothelioma mortality. Br burden of disease due to speciﬁc risk factors. Epidemiology 2006;17: J Cancer 2012;106:575–84. 512–9. 18. Parkin DM, Boyd L, Walker LC. The fraction of cancer attributable to 8. Tan E, Warren N. HSE Research Report RR876: Mesothelioma mor- lifestyle and environment factors in the UK in 2010: summary and tality in Great Britain, the revised risk and two-stage clonal expansion conclusions. Br J Cancer 2011;105 Supp 2:S77–S81. models 2011. Available from: http://www.hse.gov.uk/research/rrpdf/ 19. Mistry M, Parkin DM, Ahmad AS, Sasieni P. Cancer incidence in the rr876.pdf. Accessed 26/03/2012. United Kingdom: projections to the year 2030. Br J Cancer 2011; 9. Cherrie J. Reducing occupational exposure to chemical carcinogens. 105:1795–803. Occup Med 2009;59:96–100. 20. Cherrie JW, Gorman Ng M, Shafrir A, van Tongeren M, Searl A, San- 10. Pritchard J. HPA Compendium of chemical hazards: sulphuric acid. UK chez_Jimenez A, et al. Health, socio-economic and environmental Health Protection Agency CHAPD HQ, 2007 version 2. aspects of possible amendments to the EU Directive on the protection 11. Health and Safety Executive Board. 3rd European Commission direc- of workers from the risks related to exposure to carcinogens and muta- tive on Indicative Occupational Exposure Limit Values (IOELVs)–draft gens at work. Results from the SHEcan project funded by the European consultative document. 26th January 2011, HSE/11/05. Commission. IOM Research Project P937. Forthcoming, 2012.

1222 Cancer Prev Res; 5(10) October 2012 Cancer Prevention Research

Intervening to Reduce the Future Burden of Occupational Cancer in Britain: What Could Work?

Sally Hutchings, John W. Cherrie, Martie Van Tongeren, et al.

Cancer Prev Res 2012;5:1213-1222. Published OnlineFirst September 7, 2012.

Updated version Access the most recent version of this article at: doi:10.1158/1940-6207.CAPR-12-0070

Supplementary Access the most recent supplemental material at: Material http://cancerpreventionresearch.aacrjournals.org/content/suppl/2012/09/10/1940-6207.CAPR-12-0070.DC 1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerpreventionresearch.aacrjournals.org/content/5/10/1213. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.