Background The International Agency for Research on Cancer (IARC) has determined there is sufficient evidence that working in the rubber manufacturing industry increases the risk of cancers of the stomach, lung, bladder and leukaemia and lymphoma.
Objectives To examine mortality patterns of a prospective cohort of men from the rubber and cable manufacturing industries in Great Britain.
Methods SMRs were calculated for males aged 35+ years at start of follow-up in 1967–2015 using the population of England and Wales as the external comparator. Tests for homogeneity and trends in SMRs were also completed.
Results For all causes, all malignant neoplasms, non-malignant respiratory diseases and circulatory diseases, SMRs were significantly elevated, and also particularly for cancers of the stomach (SMR=1.26,95% CI 1.18 to 1.36), lung (1.25,95% CI 1.21 to 1.29) and bladder (1.16,95% CI 1.05 to 1.28). However, the observed deaths for leukaemia, non-Hodgkin’s lymphoma (NHL) and multiple myeloma were as expected. Bladder cancer risks were elevated only in workers exposed to antioxidants containing 1-naphthylamine and 2-naphthylamine.
Conclusions This study provides evidence of excess risks in the rubber industry for some non-cancer diseases and supports IARC’s conclusions in relation to risks for cancers of the bladder, lung and stomach, but not for leukaemia, NHL or multiple myeloma.
- rubber and cable industry
- industrial cohort study
Statistics from Altmetric.com
What is already known about this subject?
According to the International Agency for Research on Cancer (IARC), working in the rubber industry increases the risk of cancer of the urinary bladder, lung, stomach and of leukaemia, non-Hodgkin’s lymphoma (NHL) and multiple myeloma.
What are the new findings?
This paper confirms earlier findings from this cohort that removal of antioxidants containing 1-naphthylamine and 2-naphthylamine has reduced the bladder cancer risk in the cohort.
It confirms IARC’s views of an increased risk for cancers of the urinary bladder, lung and stomach, but does not provide support for increased risks of leukaemia, NHL or multiple myeloma and provides evidence that work in the rubber industry is associated with an increased risk in circulatory diseases (ischaemic heart diseases and cerebrovascular disease) and non-malignant respiratory diseases (bronchitis).
How might this impact on policy or clinical practice in the foreseeable future?
Occupational physicians and health and safety managers within the industry should remain aware of the associated hazardous exposures and ensure they are kept to a level that is as low as reasonably practical.
The International Agency for Research on Cancer (IARC) working group assessment of the rubber manufacturing industry in 1982 concluded that workers with documented exposure to 2-naphthylamine were at an increased risk of cancer of the urinary bladder.1 In its most recent evaluation, in 2012, the IARC working group identified an increased risk for leukaemia, possibly due to exposure to solvents, such as benzene, as well as sufficient evidence of increased risks of non-Hodgkin’s lymphoma (NHL), multiple myeloma and lung and stomach cancers. The evidence for laryngeal cancer and prostate cancer was deemed inconsistent, and that for oesophageal cancer was considered weak. For other cancer sites, the evidence was regarded as insufficient.2 In addition to these increased cancer risks, several other studies of rubber industry workers suggest potentially higher levels of certain respiratory3 4 and circulatory4 5 diseases.
The previous analysis of the current study cohort examined the mortality of 40 867 men in the British industry over a 10-year follow-up period. The main finding was that an excess risk of bladder cancer was shown in men with exposure to antioxidants (containing trace amounts of 1-naphthylamine and 2-naphthylamine) that may have occurred before the compounds were withdrawn in 1949, with no such risk in men who had joined after 1949 or who had worked in factories where those antioxidants had never been used. Other significant findings included an overall excess of lung cancer in the industry, associated with rubber fume, and an excess of stomach cancer, probably associated with exposure to rubber dust in the tyre manufacturing sector.6
The current study extends the previous 10-year follow-up to 49 years and compares the mortality experience of this cohort of rubber industry workers with the general population of England and Wales, examining SMRs by decade of employment start, industrial sector and department. A priori important disease groups were those cancer sites examined previously by IARC, namely leukaemia, NHL, multiple myeloma and cancers of the urinary bladder, lung, stomach, prostate, oesophagus and larynx, as well as non-malignant respiratory diseases (NMRDs) and circulatory diseases (CDs).
Assembly of study cohort
The original census to recruit workers into the cohort was initiated in 1967 by a predecessor to the UK Health and Safety Executive. In the previous analysis,6 40 867 workers were included once women, men younger than 35 years of age or those employed in the rubber industry for <1 year, together with those who could not be traced, were excluded. As well as the start year for working in the industry, the occupational data ascertained during the original survey included job position, factory of employment, industry sector (eg, tyres, cables, hoses) and department, all at the time of study recruitment only, with the information for each participant recorded on individual study cards. In addition, exposure to antioxidants was recorded based on the findings from a 1965 survey by the HM Factory Inspectorate. The categories were as follows: group A—started work before 1 January 1950 in a factory which used the suspect antioxidants; group B—started work on or after 1 January 1950 in a factory which had used the suspect antioxidants and group C—worked in factories which had never used the suspect antioxidants.6–9 For this analysis, the cohort had to be retraced from original study cards, which could only be located for 37 626 (92%) workers. Individuals were censored at the time of cancellation of registration with a GP or emigration (n=136). Although the tracing exercise extended beyond 2015, the number of deaths reported in 2016 was incomplete compared with 2014 and 2015 (125 deaths vs 306 and 297, respectively). Thus, follow-up was extended until 31 December, 2015, a total of 49 years. After exclusion of unmatched or non-uniquely matched records, those not adhering to the initial study inclusion criteria (eg, <35 years of age at study start), and checks for internal consistency (eg, year of first employment≥birth year+14), the final cohort for available analysis consisted of 36 442 workers (89.2% of the original cohort). A total of 1939 deceased workers included a death date, but no cause of death; these deaths were included only in the all-cause mortality analysis. Fifteen job groups, based on the British Rubber Manufacturing Association classification, were originally included in the survey, and were allocated into four departments for analysis purposes (table 1).
Population estimates and number of deaths by cause for males in England and Wales from 1967 to 2015 were obtained from the Office for National Statistics. Annual 5-year male reference rates were calculated for causes of death for each year in the follow-up period (1967–2015) (see online supplementary table 1 for a list of International Classification of Diseases codes). Person-years, SMRs, 95% CIs and Breslow and Day tests for trends and homogeneity8 9 were carried out using Stata’s ‘strate’ and ‘smrby’ commands.10 SMRs were reported where there were at least five observed deaths.
Supplementary file 1
For a priori important causes of death (based on the IARC monograph conclusions and some early studies for non-malignant diseases), SMRs were also compared by decade of start in the industry, tyre and non-tyre (eg, moulding, footwear) industries and department. Sensitivity analyses were undertaken to examine any differences in SMR results for rubber industry-related health outcomes if individuals aged 90 years or older were censored, as well as for all individual job groups (see online supplementary table 3).
Table 1 presents descriptive characteristics of the study cohort. The workforce was distributed fairly evenly across the age groups 35–44, 45–54, and 55–64 years at start of follow-up, with 2.1% of the cohort aged 65 years or over. Almost 5% of the cohort began work in the industry before 1930, nearly 10% during the 1930s and just over 20% during the 1940s. Slightly over 40% of workers were employed in tyre manufacturing. Around 10% worked with crude materials, and around 20% were involved in preprocessing operations, with over 10% in curing/vulcanising and the remainder in finishing. Over 90% of the cohort were deceased at the end of follow-up with just under 95% of the deaths including a coded cause of death.
For the a priori selected causes of death, statistically significantly elevated SMRs were obtained for all causes, all MNs and for cancers of the stomach, lung, bladder and NMRDs and CDs (table 2). SMRs were not different to those expected for multiple myeloma, leukaemia, NHL and cancers of the larynx, oesophagus and prostate. Undertaking the SMR analysis with the exclusion of those aged 90 years or older had a negligible impact on study findings (data not shown).
Assessing SMRs across decade of employment start (table 3) showed statistically significant increasing trends for all causes (p<0.001), all neoplasms (p=0.017), all causes excluding neoplasms (p<0.001), lung cancer (p<0.001), CDs (p=0.041) and NMRDs (p<0.001). The exception to the direction of these trends was bladder cancer, which exhibited a strong decreasing SMR trend across decade (p=0.015) and only demonstrated a significantly increased SMR for those starting employment before 1930 (n=36) (1.89, 95% CI 1.37 to 2.63). Bladder cancer was the only a priori selected end point that was increased in workers starting before 1949 (n=170) (1.33, 95% CI 1.15 to 1.55), but not during or after 1949 (n=247) (1.06, 95% CI 0.94 to 1.20) (see online supplementary table 2). SMRs were significantly increased for bladder cancer only among those who worked in factories that used 1-naphthylamine and 2-naphthylamine prior to 1949 (n=156) (1.32, 95% CI 1.13 to 1.54). No increase was evident for those who started work in such factories after 1949 (n=165) (1.07, 95% CI 0.92 to 1.25), nor in facilities in which those substances had never been used (n=96) (1.10, 95% CI 0.90 to 1.34).
Examination of SMRs in the tyre and non-tyre sectors of the rubber industry (table 4) indicated significant differences for stomach cancer, which was higher in the tyre (n=343) (1.38, 95% CI 1.23 to 1.53) than non-tyre industries (n=425) (1.19, 95% CI 1.08 to 1.31) (p=0.042), but still significantly elevated in both sectors. The risk of NMRDs, while also significantly raised in both sectors, was higher in the non-tyre (n=2856) (1.17, 95% CI 1.13 to 1.21) compared with the tyre sector (n=1874) (1.09, 95% CI 1.04 to 1.14) (p=0.019). The NHL SMR was found to be significantly lower in the tyre sector of the rubber industry (n=47) (0.69, 95% CI 0.52 to 0.92), but was not different from unity in the non-tyre sectors (n=96) (0.99, 95% CI 0.81 to 1.21) (p=0.037). No differential excess risks were observed in the tyre or non-tyre sectors for cancers of the oesophagus, larynx or prostate, or for leukaemia, multiple myeloma or CDs.
Conducting analyses by departments (table 5) showed that SMRs tended to be highest in ‘crude materials’ and lowest in ‘finishing’ across many of the a priori important disease groups. Formally assessing SMRs gave statistically significant heterogeneity for all causes (p<0.001), all malignant neoplasms (p<0.001), all causes excluding neoplasms (p<0.001), lung cancer (p<0.001), CDs (p=0.003) and NMRDs (p<0.001). The pattern was noticeably different for bladder cancer, with the highest SMR in ‘preprocessing’ (n=107) (1.41, 95% CI 1.16 to 1.70; p=0.131). For the other a priori important disease groups, risks were homogeneous between departments.
The results by individual job groups mainly presented the same patterns as the analysis by department. However, there were additional instances of significantly increased SMRs that had not been evident in the analysis by department: leukaemia in commercial staff and oesophageal cancer in both extruding and inspection workers (see online supplementary table 3).
This study sought evidence in relation to the risks of working in the British rubber and cable manufacturing industry using a large cohort that has not been studied since 1980,6 by extending the follow-up to 49 years and which resulted in 93% mortality. The overall SMR for bladder cancer of 1.16 is lower than that for lung cancer, and is also potentially influenced by tobacco smoking. In contrast to lung cancer, the SMR by decade first employed has decreased substantially, from just under 2.0 for those first employed before 1930 to not significantly raised >1 for those first employed in any subsequent decade. Bladder cancer SMRs were significantly elevated for the tyre and non-tyre sectors, but only elevated for preprocessing operations, and, in particular, those working in extruding and component buildings. The data support the notion that working in the industry conferred an increased risk of bladder cancer, but only for those exposed to antioxidants containing 1-naphthylamine and 2-naphthylamine (and maybe others), which were withdrawn from use in 1949. Research from other countries has also identified increased bladder cancer risks only for pre-1950s employment in the industry.4
This study provided some evidence of raised risks of ischaemic heart disease and bronchitis, risks of which are both strongly influenced by tobacco smoking. The finding for ischaemic heart disease seems to be uniformly elevated across all decades workers first started employment in the industry, whereas for bronchitis the risk increased with decade starting employment. The excesses of both of these end points were found in the non-tyre industries and in the four main job groups for ischaemic heart disease, but, for bronchitis, only in the crude materials and preprocessing groups. An excess of mortality from chronic, but not acute, ischaemic heart disease was identified in an earlier study,5 while a meta-analysis (excluding the earlier study) identified lowered risks in rubber industry workers11; meta-SMRs for bronchitis were as expected in that study. Our findings of increased mortality in these non-cancer end points warrant further investigation to see if they are related to working in the industry.
Our study found statistically significantly increased SMRs for cancers of the stomach, lung and bladder, but not for multiple myeloma, leukaemia, NHL or for cancers of the oesophagus, larynx or prostate. Thus, our findings are largely consistent with those of IARC, except that we found no increase in risk for leukaemia, NHL or multiple myeloma.
The overall SMR for lung cancer of 1.25 is at a level where the excess risk could be due to confounding by tobacco smoking; however, we did not have data on the smoking history of cohort members. A study examining the role of tobacco smoking and lung cancer in the rubber industry found excess rate ratios in the curing department after adjusting for smoking, and suggested that the excess could be partially due to exposure to n-nitrosamines in the process.12 In the present study, there was a highly statistically significant trend for increasing lung cancer SMRs from those first employed in the industry before 1930 through to those first employed in the 1960s, which mirrors the general rise of per capita tobacco product use in the UK. 13 This SMR is significantly raised in the tyre and non-tyre sectors and, although raised in all job groups, is lower in finishing than the others. Such analyses need to be treated cautiously since surrogate indicators of exposure in occupational cohort studies can lead to an overestimation of risks, especially for low-risk jobs and work areas.14 Studies published since the latest IARC review2 have also found increased lung cancer risks.15 16
The overall SMR for stomach cancer was similar to that for lung cancer; tobacco smoking also influences the risk of this tumour. No trend in SMRs by decade of first employment was seen for stomach cancer and, as in the earlier follow-up,6 the excess was higher in the tyre sector, and significantly raised in the non-tyre sectors. Stomach cancer risk was elevated for workers in the crude materials and preprocessing departments and particularly for compounding, latex mixing, extruding, inspection, site working and engineering jobs. Recent studies have not found an increased stomach cancer risk in the rubber industry.15–20
None of the overall SMRs for multiple myeloma, leukaemia or NHL was in excess. Another study based on the UK rubber industry identified a significantly increased risk of mortality from multiple myeloma, but this was based on only seven cases.21 In the current study, there was a significant excess of leukaemia among commercial staff, but this could be due to chance rather than employment in the rubber industry given the low number of cases involved. Since the IARC review, only one study has found a relationship between time since first exposure and leukaemia risk, but there was no overall excess.20 Although a meta-analysis found consistent evidence of raised risks for laryngeal cancer, with some positive findings for oesophageal cancer and prostate cancer from working in the rubber manufacturing industry,22 such risks were not identified in the present study. The absence of laryngeal and oesophageal cancers may suggest that smoking levels may not have been higher in the rubber industry compared with the general population, since these cancers are also linked to tobacco smoking.23
To the best of our knowledge, this study constitutes the longest and most complete follow-up of the mortality of men employed in the rubber manufacturing industry in the UK. We carried out sensitivity analyses relating to deaths in older ages and this had no impact on any of our analyses and, therefore, our conclusions. In spite of the strengths, the study had a number of limitations that warrant discussion. First, this was a prospective cohort based on a census of the industry carried out in 1967. As such, there is evidence, including increasing trends with later decade of employment start, which suggests there is a healthy worker survivor effect operating24–26 in this study; workers in the industry would have had to have survived and still be working in the industry by 1967 to have been included in the cohort. At the same time, it should also be noted that there is a distinct absence of a healthy worker effect due to the long follow-up of the cohort26 27 as exemplified by significantly elevated SMRs from NMRDs and CDs, and so it is unlikely that the SMRs in this study will be biased towards the null by this. Second, we did not have complete work histories for the cohort, only job held at the time of recruitment into the study. This limited our ability to take account of different jobs over time and to explore carcinogenic and other mortality risks by job title in more detail. Third, we had no information on smoking or any other important risk factors, such as work in other industries, exposure to asbestos or diet. This omission could have resulted in a bias in those SMRs potentially affected by these risk factors. Our findings will also be subject to multiple significance testing.28 In addition, this analysis only contained mortality data, and a future analysis of cancer incidence data is warranted for less fatal cancers, such as leukaemia.
An internet search of a sample of factories suggested that most incorporated in the original study have since closed, and remaining factories in the UK will have experienced improved hygiene standards and/or employed enhanced automation, both contributing to observed reductions in overall exposures.29 Nevertheless, with global production of natural and synthetic rubber in 2016 in excess of 25 million tonnes, many workers around the world remain exposed to process generated emissions in rubber manufacturing.30 Our results indicate potentially higher risks in the earlier stages of manufacturing, so continued research is needed to identify causative agents, especially in this stage of the process. Efforts should continue to be made to reduce employee exposures to dust and fumes throughout the production process.
This study provides evidence that bladder cancer risk has declined in the industry, and that there are increased risks for mortality from ischaemic heart disease and bronchitis and supports IARC’s conclusions in relation to risks in the rubber industry for cancers of the bladder, lung and stomach. IARC’s conclusions for leukaemia, NHL or multiple myeloma were not supported. Extension of this study to include cancer incidence and its inclusion in an international pooled study would further help clarify carcinogenicity from working in the industry.
The authors would like to thank Marlyn Davis for administrative support and Dario Consoni for assistance with implementing heterogeneity and trends tests of SMRs in Stata. The authors would additionally like to thank staff at the Health and Safety Executive, The University of Manchester and the Institute of Occupational Medicine for computerising the many thousands of study cards.
Contributors DMME and FdV conceived of the study. FdV, DMME, JC and RA obtained funding for the study. WM conducted the statistical analyses and wrote the first draft version. All authors commented on the methodology and interpretation of the results. All authors commented on the draft versions of the manuscript.
Funding This study was funded by Cancer Research UK (C29425/A16521). Additional funding for tracing of the cohort was provided by the UK Health and Safety Executive (PRJ787).
Disclaimer The views expressed are those of the author(s) and not necessarily those of CRUK or HSE.
Competing interests None declared.
Patient consent Not required.
Ethics approval The study was approved by an NHS ethics committee (Ref: 13/NW/0543), the Health Research Authority’s Confidentiality Advisory Group (Ref: CAG 5-08(d)/2013), the Office for National Statistics and NHS Digital’s Data Access Advisory Group (now Independent Group Advising on the Release of Data, Ref: NIC-323309-L2G9T).
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Additional unpublished data are not available for this study.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.