Cancer Incidence Among Firefighters: 45 Years of Follow-Up in Five Nordic Countries
Total Page:16
File Type:pdf, Size:1020Kb
Downloaded from oem.bmj.com on February 8, 2014 - Published by group.bmj.com OEM Online First, published on February 6, 2014 as 10.1136/oemed-2013-101803 Workplace ORIGINAL ARTICLE Cancer incidence among firefighters: 45 years of follow-up in five Nordic countries Eero Pukkala,1,2 Jan Ivar Martinsen,3 Elisabete Weiderpass,3,4,5,6 Kristina Kjaerheim,3 Elsebeth Lynge,7 Laufey Tryggvadottir,8,9 Pär Sparén,4 Paul A Demers10 For numbered affiliations see ABSTRACT end of article. Objectives Firefighters are potentially exposed to a What this paper adds Correspondence to wide range of known and suspected carcinogens through their work. The objectives of this study were to Professor Eero Pukkala, Finnish ▸ This first study on cancer risk among Nordic Cancer Registry, Institute for examine the patterns of cancer among Nordic firefighters indicates a small excess in the Statistical and Epidemiological firefighters, and to compare them with the results from overall cancer incidence. Cancer Research, previous studies. Pieni Roobertinkatu 9, ▸ There are more prostate cancer cases Methods Data for this study were drawn from a Helsinki FI-00130, Finland; diagnosed in ages <50 years than in the men fi linkage between the census data for 15 million people eero.pukkala@cancer. in the general population. from the five Nordic countries and their cancer registries ▸ Risk of myeloma, adenocarcinoma of the lung Novelty & Impact for the period 1961–2005. SIR analyses were conducted and mesothelioma are increased in older ages. with the cancer incidence rates for the entire national This study describes the cancer ▸ By contrast with earlier studies, the incidence study populations used as reference rates. incidence pattern among over of testicular cancer was decreased. 16 000 firefighters with more Results A total of 16 422 male firefighters were than 2500 cancer cases. There included in the final cohort. A moderate excess risk was was a small excess in the seen for all cancer sites combined, (SIR=1.06, 95% CI overall cancer incidence, but fi quite high and statistically 1.02 to 1.11). There were statistically signi cant excesses significant excesses, for in the age category of 30–49 years in prostate cancer provide 24 h fire protection, which may involve example, for prostate cancer in (SIR=2.59, 95% CI 1.34 to 4.52) and skin melanoma disruption of circadian rhythms.1 younger ages; and for multiple (SIR=1.62, 95% CI 1.14 to 2.23), while there was Many cohort studies have been conducted, but myeloma, adenocarcinoma of the lung, and mesothelioma in almost no excess in the older ages. By contrast, an the results have been inconsistent, and the risk of older ages. increased risk, mainly in ages of 70 years and higher, cancer among firefighters has been a controversial Observations from this large was observed for non-melanoma skin cancer (SIR=1.40, topic, especially in North America where workers’ cohort with long and complete 95% CI 1.10 to 1.76), multiple myeloma (SIR=1.69, compensation has been a contentious issue. follow-up strengthen the 95% CI 1.08 to 2.51), adenocarcinoma of the lung Recently, two large, comprehensive reviews of International Agency for fi fi Research on Cancer (SIR=1.90, 95% CI 1.34 to 2.62), and mesothelioma cancer in re ghters have been completed. The classification that occupational (SIR=2.59, 95% CI 1.24 to 4.77). By contrast with first was a meta-analysis of 32 studies conducted by exposure related to firefighting earlier studies, the incidence of testicular cancer was LeMasters et al5 which evaluated the risk for 21 is possibly associated with decreased (SIR=0.51, 95% CI 0.23 to 0.98). cancers. Based on the strength of the association, cancer risk. Conclusions Some of these associations have been consistency and study type, the authors classified Received 18 August 2013 observed previously, and potential exposure to polycyclic multiple myeloma, non-Hodgkin lymphoma, pros- Revised 2 January 2014 aromatic hydrocarbons, asbestos and shift work involving tate cancer and testicular cancer as probably being Accepted 13 January 2014 disruption of circadian rhythms may partly explain these associated with firefighting. Additionally, cancer of results. the skin, brain, rectum, buccal cavity and pharynx, stomach and colon, as well as malignant melanoma and leukaemia were classified as possibly related to firefighting. INTRODUCTION The second comprehensive review was con- Firefighters are potentially exposed to a wide range ducted by an International Agency for Research on of known and suspected carcinogens through their Cancer (IARC) monograph working group (2010) work. The constituents of fire smoke vary accord- as part of its evaluation of firefighting. The group ing to combusted materials and other factors, but reviewed the results of 42 studies, including the benzene, formaldehyde, polycyclic aromatic hydro- results of two large epidemiology studies that had ▸ http://dx.doi.org/10.1136/ carbons and fine particulates are common.1 been published after the review by LeMasters oemed-2013-101662 Exposures can be quite high, although usually for et al.5 Based on internal meta-analyses, consistency short periods of time. Firefighters may also be of the studies and limited dose-response informa- exposed to asbestos, which may be released when tion using duration of employment, the working To cite: Pukkala E, older structures burn, and historically, asbestos was group concluded that the strongest evidence was Martinsen JI, Weiderpass E, a component of some protective equipment, such for testicular cancer, prostate cancer and et al. Occup Environ Med 12 fi fi Published Online First: as gloves. Last, re ghters may be exposed to non-Hodgkin lymphoma. However, the working [please include Day Month diesel exhaust, recently classified as a human lung group, as a whole, classified firefighting exposure as Year] doi:10.1136/oemed- carcinogen, from firefighting vehicles.34Additionally, being only possibly associated with cancer (IARC 2013-101803 most firefighters work rotating or extended shifts to Category 2B).1 PukkalaCopyright E, et al. Occup Article Environ author Med 2014; (or0:1– 7.their doi:10.1136/oemed-2013-101803 employer) 2014. Produced by BMJ Publishing Group Ltd under licence. 1 Downloaded from oem.bmj.com on February 8, 2014 - Published by group.bmj.com Workplace Although many studies on firefighters have been conducted, were then counted until the date of emigration, death or to 31 the great majority were limited to cancer information from December of the following years: in Denmark and Norway death certificates. The use of death certificates limits the ability 2003, in Iceland 2004 and in Finland and Sweden 2005. The to study the cancer sites with high survival rates as well as those source of the data on dates of death and emigration in all coun- based on histology. Additionally, many of the previous studies tries was the Central Population Register. Data on tumours were were not sufficient in size to examine the risk of rare cancers. derived from record linkage with the cancer registries from each Last, few studies of firefighters have been conducted outside of of the five Nordic countries. National cancer registration started North America. The Nordic countries, with their national with cancer diagnosed in 1943 in Denmark, in 1953 in Finland cancer registries and ability to conduct record linkage, would and Norway, in 1955 in Iceland and in 1958 in Sweden. During seem to be ideal places to study firefighters. The Nordic fire- the follow-up period of the NOCCA study, all cancer registries fighter unions have recently—after having had several cases of, received information on cancer cases from general and specialist for example, young-age prostate cancers in one single fire practitioners, hospital departments and from pathology autopsy department—showed special interest in knowing about cancer notifications. All countries except Denmark received notifica- risk in their occupation, which makes this topic especially per- tions from pathology departments. Sweden was the only tinent at this time. country not receiving copies of the death certificates. The com- Two small cohort studies have previously been conducted in pleteness and accuracy of the Nordic cancer registries are very the Nordic countries. Hansen in 1990 observed an excess risk high in international comparison, and much work has been lung cancer mortality among Danish firefighters, but had limited done to reach comparability in cancer classifications between power to examine the risk of other specific cancers.6 Tornling the Nordic countries over the more than five decades of cancer and colleagues in 1994 identified possible dose-response rela- registration.11 tionships with the number of fires fought, for stomach and The cancer cases have been grouped into 49 main categories brain cancer, in a retrospective cohort study of Stockholm fire- and 27 diagnostic subgroups based on the national topography fighters.7 It was among the best studies conducted to date in and morphology coding systems (Appendix tables 5–6 at http:// terms of exposure information, but was limited in size. astra.cancer.fi/NOCCA). The incident cancer cases included in Recently, the Nordic Occupational Cancer (NOCCA) project the present study involve all invasive cancers and also benign (http://astra.cancer.fi/NOCCA) linked the census data on occu- brain tumours. The non-melanoma skin cancers were excluded pations from the five Nordic countries and their cancer regis- from the ‘All cancers’ category, as basal cell carcinomas of the tries for the period from 1961 to 2005.8 In this paper, we use skin could not be separated from the rest of the data collected for the NOCCA study and present the results for ‘Non-melanoma skin cancers’ (primarily squamous cell carcin- cancer incidence among Nordic firefighters. The objectives of omas) in the Danish data.