Occupational Medicine Advance Access originally published online on January 22, 2007
Occupational Medicine 2007 57(3):177-185; doi:10.1093/occmed/kql168
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mortality of UK oil refinery and petroleum distribution workers, 1951–2003
Institute of Occupational and Environmental Medicine, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
Correspondence to: Tom Sorahan, Institute of Occupational and Environmental Medicine, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. Tel: +0121 414 3644; fax: +0121 414 6217; e-mail: t.m.sorahan{at}bham.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
Background The health of UK petroleum industry workers has been monitored for many years.
Aim To identify any long-term adverse health outcomes from occupational exposures in this industry.
Methods The mortality (1951–2003) and cancer morbidity (1971–2003) experienced by cohorts of 28 555 oil refinery workers and 16 477 petroleum distribution workers has been investigated. Study subjects were all those males first employed in the period 1946–74 at one of eight UK oil refineries or 476 UK petroleum distribution centres; all subjects had a minimum of 12 months employment with some employment after 1 January 1951. Observed numbers of cause-specific deaths and site-specific cancer registrations were compared with expectations based on national mortality and cancer incidence rates.
Results Standardized mortality ratios (SMRs) were significantly <100 for all causes both in oil refinery workers (Obs 11 156, SMR 89) and in petroleum distribution workers (Obs 7320, SMR 96). Significantly elevated SMRs were shown in oil refinery workers for cancer of the pleura (mesothelioma) (Obs 64, SMR 261) and melanoma (Obs 48, SMR 168). Significantly elevated SMRs were not found in petroleum distribution workers for any site of cancer. Significantly elevated standardized registration ratios (SRRs) were only shown in oil refinery workers and for cancer of the pleura (mesothelioma) (Obs 115, SMR 274), melanoma (Obs 85, SMR 129) and other skin cancer (Obs 983, SRR 117).
Conclusions The only findings that showed clear evidence of an occupational cancer hazard were those for mesothelioma in oil refinery workers.
Keywords Cancer; oil refinery workers; petroleum distribution workers
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
In the 1970s the Institute of Petroleum (now known as the Energy Institute) developed epidemiological cohort studies on the mortality and cancer morbidity experience of male employees from eight oil refineries and 476 petroleum distribution centres in the UK [1–7]. A large number of cohort studies of petroleum industry workers in different parts of the world have been carried out and meta-analyses of findings by site of cancer are available [8–10].
The original cohorts comprised 34 569 oil refinery workers [1] and 23 358 petroleum distribution workers [2]. All these male employees had a minimum period of employment of 12 months in the period 1950–75; some study subjects were first employed around the turn of the century. Analyses are now limited to those workers first employed after 1 January 1946; the rationale for this re-definition has been supplied previously [5].
A further 5 years of mortality data (1999–2003) were available for analysis in the overall period of follow-up (1951–2003), together with previously unreported cancer registration data for the period 1971–2003. The objectives of the study were to summarize available mortality and cancer incidence data and to determine whether any part of this experience might be related to occupational exposures (in which event, further analyses capable of investigating the potential role of occupational exposures might be needed).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
The computer file for the revised cohorts contained identifying particulars, work history information (oil refinery or petroleum distribution centre, dates of commencing and leaving employment and job title in 1975 or last job if left employment before 1975) and follow-up information (date of death, underlying and contributory causes of death and cancer registration particulars) for 28 555 oil refinery workers and 16 477 petroleum distribution workers first employed in the period 1946–74. All subjects had a minimum of 12 months employment with some employment after 1 January 1951. Six of the oil refineries were in England and Wales; the remaining two refineries were located in Scotland. A total of 403 of the petroleum distribution centres were in England and Wales; the remaining 73 centres were located in Scotland.
In preparation for the current analysis a number of procedures were carried out to improve data quality including searches for duplicate entries. A total of 75 duplicates were removed from the revised cohort of oil refinery workers that now numbers 28 555 employees and three duplicates were removed from the revised cohort of petroleum distribution workers that now numbers 16 477 employees. [A total of 142 oil refinery workers are also members of the distribution workers cohort on the basis of earlier employment as a petroleum distribution worker; 162 petroleum distribution workers are also members of the oil refinery cohort on the basis of earlier employment as an oil refinery worker. Descriptions of the original study computer files indicate that the same person is allowed to be a member of both cohorts (oil refinery workers and petroleum distribution workers).]
A comparison of study computer files with vital status data maintained at the National Health Service Central Register (NHSCR) is often carried out, particularly when a number of different personnel have been involved with maintaining a study either at the same institution or at different locations. A computer file was obtained from the NHSCR showing the current vital status of all study subjects who were alive at the time of the NHSCR computerization in 1991. There were 534 oil refinery workers and 57 petroleum distribution workers in the study computer files under the heading ‘traced alive at NHSCR’ who were not found in the NHSCR files. Identifying particulars for these study subjects were sent to NHSCR for ascertainment of vital status.
At the time of the last update it was noted that cancer registration data appeared to be incomplete [5] and steps have been taken to remedy this situation. All subjects (417 oil refinery workers and 322 petroleum distribution workers) dying with cancer (underlying or contributory cause) in the period 1974–2002 but without any cancer registration details were identified. [Non-melatanous skin cancers (‘other skin cancer’) were excluded from lists of both registrations and deaths.] Details of these former employees were sent either to the NHSCR or to the General Register Office (GRO) for Scotland so that a further search could be made for the ‘missing’ cancer registrations. A total of 232 cancer registrations in oil refinery workers and 159 cancer registrations in petroleum workers were identified and details were incorporated into the main study files.
The NHSCR and the GRO for Scotland provided vital status information on the closing date of the survey, 31 December 2003. For those refinery workers who had died (n = 11 754), a copy of the death certificate was supplied with the underlying and other causes of death coded to the contemporaneous revision of the International Classification of Diseases (ICD); the cause of death was untraced for only 20 deaths (0.2%). A total of 1713 refinery workers (6.0%) had emigrated and 285 refinery workers were untraced (1.0%). A total of 7685 decedents were identified in the petroleum distribution workers; the cause of death was untraced for only 13 deaths (0.2%). A total of 329 distribution workers (2.0%) had emigrated and 365 distribution workers were untraced (2.2%).
Expected numbers of deaths were calculated from male mortality rates [specified by 5-year age groups, 5-year calendar periods and country (England and Wales, Scotland)] applied to similarly defined arrays of person-years-at-risk (pyr) generated by the data. Workers entered the pyr at the end of the 12-month minimum period of employment or on 1 January 1951, whichever was the later date. They left the pyr on the closing date of the study (31 December 2003), the date of death, the date of emigration or the date last known alive, whichever was the earlier date. These procedures were accomplished by means of the PERSONYEARS software. No contributions were made to observed or expected numbers past the age of 85 years; this censoring was applied for the reasons supplied previously [5].
Observed and expected numbers for workers in England and Wales were then combined with those for Scottish workers. Overall standardized mortality ratios (SMRs) were calculated as the ratio of observed deaths to expected deaths, expressed as a percentage. The significance of the differences between observed numbers and their corresponding expectations was assessed by means of the Poisson distribution. Tests for trend and heterogeneity have been described previously [5]. All significance tests were two tailed.
Corresponding analyses were carried out for cancer registrations except that the analyses were limited to the period 1971–2003; the national cancer registration scheme began in 1971.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
Overall observed and expected numbers of deaths for the main disease groupings are shown in Table 1. Mortality from all causes was significantly below that expected on the basis of national mortality rates both for oil refinery workers (Obs 11 156, SMR 89) and for petroleum distribution workers (Obs 7320, SMR 96). Significant deficits are shown for most of the important non-cancer disease groupings for both cohorts. Overall observed and expected numbers of deaths for individual cancer sites (three-digit ICD codes) are also shown. In oil refinery workers, significant deficits are shown for all neoplasms (Obs 3492, SMR 95), cancer of the liver (Obs 23, SMR 60), cancer of the pancreas (Obs 123, SMR 81) and for lung cancer (Obs 1106, SMR 87). Significant excesses are shown for cancer of the pleura (mesothelioma) (Obs 64, SMR 261) and melanoma (Obs 48, SMR 168). Leukaemia mortality was unexceptional (Obs 96, SMR 107). In petroleum distribution workers, there were no statistically significant differences between observed and expected numbers for any individual site of cancer, although an SMR of 124 was shown for leukaemia (Obs 67). [The removal of age censoring on observed and expected numbers had little impact on SMRs for cancer of the pleura (Obs 65, SMR 262) or lung cancer (Obs 1132, SMR 89) (not shown in Table 1).]
|
Expectations for leukaemia subtypes were not available for the full period of study, but these could be calculated for the period 1968–2003 (not shown in Table 1). In refinery workers, there was no suggestion of excess risks for individual leukaemia subtypes [chronic lymphatic leukaemia (CLL): Obs 21, Exp 19.6, SMR 107, 95% CI 66–164; acute myeloid leukaemia (AML): Obs 30, Exp 34.0, SMR 88, 95% CI 60–126; acute non-lymphocytic leukaemia (ANLL): Obs 36, Exp 38.7, SMR 93, SMR 65–129]. In distribution workers, non-significantly elevated SMRs were found (CLL: Obs 18, Exp 12.0, SMR 150, 95% CI 89–237; AML: Obs 29, Exp 20.9, SMR 139, 95% CI 93–200; ANLL: Obs 30, Exp 23.7, SMR 127, 95% CI 86–181).
Overall observed and expected numbers of registrations for incident cancers (three-digit ICD codes) in the period 1971–2003 are shown in Table 2. In oil refinery workers, registrations for all neoplasms were significantly below that expected on the basis of national incidence rates (Obs 4631, SRR 97). Significant excesses are shown for cancer of the pleura (mesothelioma) (Obs 115, SRR 274), melanoma (Obs 85, SMR 129) and other skin tumours (Obs 983, SRR 117). Leukaemia incidence (all types) was close to expectation (Obs 114, SRR 102) as was incidence of lymphoid leukaemia (SRR 92, Obs 47) and myeloid leukaemia (SRR 103, Obs 53). In petroleum distribution workers, registrations for all neoplasms were significantly above expectation (Obs 3105, SRR 105), as were registrations for non-melatanous skin cancer (Obs 594, SRR 113). A significant deficit is shown for cancer of the pleura (mesothelioma) (Obs 9, SRR 33). Leukaemia incidence (all types) was not significantly elevated (Obs 81, SRR 118). The SRR for lymphoid leukaemia was 101 (Obs 32) while that for myeloid leukaemia was 133 (Obs 42).
|
These overall findings (Tables 1 and 2) were reviewed in order to select causes of death that warranted further investigation. Cancer of the pleura (mesothelioma) and melanoma were selected on the basis of the excess SMRs and SRRs in refinery workers. Leukaemia and ANLL were selected on the basis of continuing interest in late effects of benzene exposure [11–13]. All causes mortality was also selected to gauge the likely influence of selection and socio-economic effects. Non-melatanous skin cancer (other skin cancer) was not selected for further national comparisons as registration practices for such tumours are notoriously variable and unreliable.
Mortality from selected causes is shown by the period from commencing employment in Table 3. Highly significant positive trends are shown for all causes mortality both in refinery and distribution workers and a significant negative trend is shown for melanoma in oil refinery workers. In refinery workers, there were significantly elevated SMRs of 286 (Obs 55) for mesothelioma with late follow-up (
30 years) and 536 (Obs 9) for melanoma with early follow-up (1–9 years).
|
Mortality from selected causes is shown by year of hire in Table 4. Highly significant trends are shown for all causes mortality both in refinery and distribution workers, such that later decades of hire tend to be associated with particularly low mortality. A similar tendency is shown for mesothelioma in refinery workers, although the trend is not statistically significant.
|
Mortality in refinery workers from selected causes is shown by job type in Table 5. There is highly significant heterogeneity shown in the set of SMRs for all causes mortality, with particularly low mortality in workers with white-collar/white-coat jobs. Mesothelioma is elevated for most job types except scientific and technical workers, and administrative and clerical staff. Melanoma is significantly elevated in operators, other craftsmen and administrative and clerical staff.
|
Mortality in petroleum distribution workers from selected causes is shown by job type in Table 6. There is highly significant heterogeneity shown in the set of SMRs for all causes mortality with particularly low mortality in workers with white-collar/white-coat jobs.
|
Similar analyses to those shown in Tables 3–6
were also carried out for cancer registration data and these will be made available in technical reports. Little additional information of note was obtained from these further analyses except that the elevated incidence of cancer of the pleura in oil refinery workers was not limited to earlier years of hire (1946–49: Obs 11, Exp 6.2, SRR 178, 95% CI 89–318; 1950–59: Obs 87, Exp 26.5, SRR 329, 95% CI 263–406; 1960–69: Obs 11, Exp 6.7, SRR 164, 95% CI 82–293; 1970–74: Obs 6, Exp 2.6, SRR 228, 95% CI 84–497). In addition, the elevated incidence of melanoma in oil refinery workers was not limited to the early period of follow-up (1–9 years: Obs 4, Exp 1.0, SRR 404, 95% CI 110–1035; 10–19 years: Obs 5, Exp 5.3, SRR 95, 95% CI 31–221; 20–29 years: Obs 15, Exp 15.7, SRR 96, 95% CI 54–158; 30 years: Obs 61, Exp 44.0, SRR 139, 95% CI 106–178). |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
This study has found a marked excess of mesothelioma in oil refinery workers but not in petroleum distribution workers and no clear-cut excess of AML (or the slightly larger grouping of ANLL) in either group of workers.
Earlier analyses had found overall excess mortality from cancer of the gall bladder in refinery workers, although more detailed analyses had not indicated the influence of an occupational cancer hazard and it was concluded that the overall excess SMR may well have been no more than a chance finding [5]. This conclusion is supported by the updated analyses which found no further deaths in refinery workers from this disease in the new period of follow-up (1999–2003) and no significantly elevated SRR for incident cancers of this site.
An overall excess mortality for melanoma in UK oil refinery workers has been noted previously [1,3,5] and this excess has been maintained in the updated analyses. The detailed mortality analyses found especially elevated risks in the early period of follow-up; such a pattern is not typical for an occupational cancer hazard. However, the new cancer incidence findings suggest that an elevated risk is not limited to early periods of follow-up. Both assessments of risk (mortality and cancer incidence) found elevated risks in a mixture of blue-collar and white-collar jobs and this would argue against the involvement of a chemical hazard. Unfortunately, the present study does not include any information on variables such as sunbathing practices and number of foreign holidays and periods of working in hot climates; these factors may well be involved in the overall excesses. A meta-analysis of mortality from skin cancer (dominated by melanoma) in 350 000 oil refinery and petroleum distribution workers from many countries reported a slight overall excess (Obs 340, Exp 308.9, SMR 110, 95% CI 99–122) [9]. Removing the contribution made by UK oil refinery workers to this overall finding reduces this meta-SMR to 106 (Obs 310). It would appear that elevated melanoma is not a general problem in this industry in other parts of the world. An updated analysis of mortality and cancer incidence in Australian petroleum industry workers has, however, been published more recently [14]. There was no overall excess of melanoma mortality (Obs 19, SMR 91), although a significant excess was shown for melanoma incidence (Obs 191, SRR 137). No occupational cause was apparent for this latter excess.
The effects of an earlier asbestos hazard are now shown clearly in the findings for cancer of the pleura (mesothelioma) in oil refinery workers. Observed and expected numbers of deaths refer to cancer of the pleura in the period covered by the 8th and 9th revisions of the ICD (1968–2000) and all mesothelioma (pleura, peritoneum, other and unspecified sites) for the period covered by the 10th revision of the ICD (2001–03). The findings for incident cancers indicate that elevated risks are not limited to those first employed before 1960 and it is not possible to predict from the current data as to when the risk will have played itself out. At face value, the findings by job type suggest that asbestos exposure may have been a plant-wide problem rather than concentrated in one or two jobs, although the study is limited by having only a single job title for each worker. It is possible, of course, that some of the excess in refinery workers is due to asbestos exposure received in other industries but a major contribution can be excluded with some confidence on the basis of the contrast with distribution workers. With such a large excess of mesothelioma, it is difficult to believe that there have been no occupational asbestos-induced lung cancers. However, analyses of death certificates in terms of any mention of lung cancer and any mention of asbestosis indicate that numbers of such cancers may be quite small; this would be consistent with the low overall SMR for lung cancer. There were 21 death certificates that included a mention of asbestosis of which eight also included a mention of lung cancer; the latter figure is to be compared with an expectation of 2.0, assuming these diseases are independent. Clearly, six extra deaths from lung cancer would not be discernible on an SMR of 87 based on 1106 deaths. However, not all cases of asbestosis would appear on the death certificate so the estimate of six occupational lung cancers in oil refinery workers is probably an underestimate.
For leukaemia, the findings were a little more complex. There were a number of isolated excesses but little evidence of important patterns of risk. It is to be expected that benzene exposure, particularly the higher levels found in various parts of the petroleum distribution industry many years ago, will have had some impact on leukaemia risks in this study, although this could hardly explain why administrative and clerical workers had higher leukaemia risks than drivers. The recent historical study of UK benzene-exposed workers [15] found only modestly elevated risks of ANLL and this latter group of workers must have had much higher mean benzene exposures than the cohorts under study. The overall mortality excess of about six ANLL deaths could be used as an upper limit for the extent of any occupational influence from benzene exposure in petroleum distribution workers. Meta-analyses of leukaemia mortality in petroleum industry workers have found little evidence of adverse health outcomes (AML, SMR 96; CML, SMR 89; ALL, SMR 116; CLL, SMR 84) [10].
In conclusion, the findings of this analysis should be welcome news for UK oil refinery workers and petroleum distribution workers. Their overall and cause-specific mortality and cancer morbidity compares favourably with the national average. However, there were 125 cases of cancer of the pleura/mesothelioma in refinery workers on the basis of death certificate or cancer incidence data (including deaths from mesothelioma of unspecified site, classified as ‘unspecified cancer’ in the eighth and ninth revisions of the ICD); the overwhelming majority of these cases are likely to be due to earlier incidental asbestos exposure received at oil refineries. It has been suggested that asbestos causes two lung cancers for every case of mesothelioma [16]. This is clearly not the case for UK oil refinery workers. More sophisticated analyses of cancer risks involving retrospective quantitative assessments of asbestos exposure could make a useful contribution to understanding the circumstances in which asbestos exposure does or does not lead to major lung cancer risks. Given that asbestos is believed to be the only important cause of mesothelioma in the UK, industries tend to be categorized as those with asbestos problems and those without; oil refinery workers are part of the former and petroleum distribution workers are part of the latter.
|
Conflicts of interest |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
None declared.
|
Acknowledgements |
|---|
The author thanks the Office for National Statistics for supplying follow-up information. The author thanks Margaret Williams for data consolidation and word processing. The author is grateful to Lesley Rushton and the late Michael Alderson for their earlier contributions to this study. The costs of this analysis were defrayed by a research award from the Energy Institute.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
- Rushton L, Alderson MR. An epidemiological survey of eight oil refineries in Britain. Br J Ind Med (1981) 38:225–234.[Web of Science][Medline]
- Rushton L, Alderson MR. Epidemiological survey of oil distribution centres in Britain. Br J Ind Med (1983) 40:330–339.[Web of Science][Medline]
- Rushton L. Further follow up of mortality in a United Kingdom oil refinery cohort. Br J Ind Med (1993) 50:549–560.[Web of Science][Medline]
- Rushton L. Further follow up of mortality in a United Kingdom oil distribution centre cohort. Br J Ind Med (1993) 50:561–569.[Web of Science][Medline]
- Sorahan T, Nichols L, Harrington JM. Mortality of United Kingdom oil refinery and petroleum distribution workers, 1951 -98. Occup Med (Lond) (2002) 52:333–339.[CrossRef][Medline]
- Sorahan T, Nichols L, Harrington JM. Mortality of United Kingdom Oil Refinery Workers, 1951-98 (2001) London: Institute of Petroleum.
- Sorahan T, Nichols L, Harrington JM. Mortality of United Kingdom Petroleum Distribution Workers, 1951-98 (2001) London: Institute of Petroleum.
- Wong O, Raabe GK. Critical review of cancer epidemiology in petroleum industry employees, with a quantitative meta-analysis by cancer site. Am J Ind Med (1989) 15:283–310.[Web of Science][Medline]
- Wong O, Raabe GK. A critical review of cancer epidemiology in the petroleum industry, with a meta-analysis of a combined database of more than 350,000 workers. Regul Toxicol Pharmacol (2000) 32:78–98.[CrossRef][Web of Science][Medline]
- Raabe GK, Wong O. Leukaemia mortality by cell type in petroleum workers with potential exposure to benzene. Environ Health Perspect (1996) 104:1381–1392.[CrossRef][Web of Science][Medline]
- Rushton L, Romaniuk H. A case-control study to investigate the risk of leukaemia associated with exposure to benzene in petroleum marketing and distribution workers in the United Kingdom. Occup Environ Med (1997) 54:152–166.
[Abstract/Free Full Text] - Schnatter RA, Armstrong TW, Thompson LS, Nicolich MJ, Katz AM, Huebner WW. The relationship between low-level benzene exposure and leukaemia in Canadian petroleum distribution workers. Environ Health Perspect (1996) 104:1375–1379.[CrossRef][Web of Science][Medline]
- Wong O, Harris F, Smith TJ. Health effects of gasoline exposure II. Mortality patterns of distribution workers in the United States. Environ Health Perspect (1993) 101(Suppl. 6):63–76.
- Gun RT, Pratt N, Ryan P, Roder D. Update of mortality and cancer incidence in the Australian petroleum industry cohort. Occup Environ Med (2006) 63:476–481.
[Abstract/Free Full Text] - Sorahan T, Kinlen L, Doll R. Cancer risks in a historical UK cohort of benzene exposed workers. Occup Environ Med (2005) 62:231–236.
[Abstract/Free Full Text] - Report Consensus. Asbestos, asbestosis, and cancer: the Helsinki criteria for diagnosis and attribution. Scand J Work Environ Health (1997) 23:311–316.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||