Document a11x76d0a448NDVVNYgkGpjMB

British Journal of Industrial Medicine 1987;44:119-123 Occupational risks for renal cancer in Sweden J K McLAUGHLIN,'2 H S R MALKER,3 B J STONE,2 J A WEINER,3 BIRGITTA K MALKER,4 J L E ERICSSON,4 W J BLOT,2 J F FRAUMENI JR2 From the Biostatistics Branch' and Epidemiology and Biostatistics Program, Division of Cancer Etiology,2 National Cancer Institute, Bethesda, Maryland 20892, USA, and National Board ofOccupational Safety and Health,3 Solna, S-17184, and National Board ofHealth and Welfare,4 Stockholm, S-106 30, Sweden ABSTRACT A systematic assessment was made of the occurrence of renal cancer among men by industrial and occupational classification using the Cancer-Environment Registry, which links cancer incidence (1961-79) and census data (1960) with industry and occupation for all employed individuals in Sweden. Data were analysed separately for cancers of the renal parenchyma and pelvis. Significantly increased risks for renal cell cancer were observed for several professional and white collar occupations, including physicians and others in the health care industry. By contrast, the risks for renal pelvis cancer tended to be higher among blue collar workers, especially in the machine industry. Deficits of both cancers occurred among farmers. The findings of the survey are considered as aetiological clues that may deserve further study, although some associations support observations in other countries. Little is known about the occupational determinants of renal cancer.' Raised risks have been reported among coke oven workers,2 laundry and dry cleaning workers,34 newspaper pressmen,5 and workers handling asbestos6 and petroleum, tar, and pitch products.7 8 Although the data were usually not reported separately by site within the kidney, the vast majority of tumours in these studies arose in the renal parenchyma (renal cell cancer). An excess riskof renal pelvis cancer has been reported for leather workers9 and a number of workplace exposures associated with bladder cancer.10 The Swedish Cancer-Environment Registry (CER), which links essentially complete cancer incidence data for 1960-79 with employment data obtained from the 1960 national census,' 1 provides a unique opportunity to assess separately the occurrence of renal cell cancer (ICD 7th rev 180-0) and renal pelvis cancer (ICD 180- 1) by industry and occupation for an entire country. We report here the results from this hypothesis generating investigation, the first to evaluate systematically occupational risks for renal cancer using large scale population based incidence date. The location of the study is of special interest, since Sweden and other Scandinavian countries have the highest rates of renal cancer in the world.' Methods A description of the CER and the statistical methods Accepted 6 May 1986 used in calculating the standardised incidence ratios (SIR) have been published before.12 SIRs were calculated for all major (one digit) and general (two digit) industrial and occupational codes. For specific (three digit) employment groups only those with 500 or more individuals were evaluated, as there are over 300 industrial and 300 occupational categories at this level of coding. Few raised risks were observed among employed women; hence only the results for men will be presented (except in those instances where female risks may clarify the observed patterns). Results Among Swedish men employed in 1960, there were 7405 cases of renal cell cancer (99% microscopically confirmed) and 821 cases of renal pelvis cancer (98% microscopically confirmed) during the 19 year follow up. Ninety five per cent of the cases of renal cell cancer were adenocarcinomas and 4% unspecified epithelial cancers; 80% of the cases of renal pelvis cancer were transitional cell carcinomas, 5% squamous cell cancer, and 12% unspecified epithelial cancers. For the major divisions of industry and occupation in Sweden (table 1) the highest risks for the renal cell cancer occurred among professional, technical, and related workers (SIR = 1-20, p < 0-01). There were also significantly raised SIRs for the white collar occupations of administrative workers and sales workers, and for the services industry. By contrast, the highest risks for renal pelvis cancer occurred 119 120 McLaughlin, Malker, Stone, Weiner, Malker, Ericsson, Blot, Fraumeni Table 1 Standardised incidence ratios (SIR) for renal cell and renal pelvis cancers by major division ofindustry and occupation among me Code Major industry Renal cell Renal pelvis cancer cases SIRt cancer cases SIRt Code Major occupation Renal cell Renal pelvis cancer cases SIRt cancer cases SIRt 0 Farming, forestry, 1287 0.88* 105 hunting, and fishing I Mining and quarrying 56 086 8 2 Manufacturing I 1024 3 ManufacturingII 1658 4 Construction 983 1 01 119 1-06* 211 1-02 113 5 Electric, gas, water, 131 and sanitary services 6 Trade, finance, 773 insurance, real estate 7 Transport and 606 communication 8 Services 868 1-17 13 1-04 83 1 04 71 1.13* 97 9 Non-classifiable services 19 0-85 1 0.67* 0 1 31 1 1-05 2 1.22* 3 1-07 4 1-07 5 1 01 6 1-11 7 1-14 8 0-41 9 Professional, technical, 995 and related workers Administrative, executive, 319 and managerial workers Clerical workers 309 Sales workers 581 Farmers, fishermen, 1255 hunters, and related workers Miners, quarrymen, 50 and related workers Transport and 532 communication workers Craftsmen, production 2048 workers, and labourers Craftsmen, production 965 workers, and labourers, NEC Services, sport, and 351 recreation workers 1 20* 106 1-14** 29 1-08 36 1 14* 64 0-87* 102 1-04 7 1 00 55 1 02 240 0-94 140 1-08 42 1-17 093 1 13 1-14 0-66* 1 45 0 94 1-08 1-22** 1-18 *p < 0-01; **p < 005. tAdjusted for age and region. NEC = Not elsewhere classified. Table 2 Standardised incidence ratios (SIR) for renal cell and renal pelvis cancers by general (two digit) manufacturing industries and craftsman tradesman occupations among men Code General industry Renal cell Renal pelvis cancer cases SIRt cancer cases SIRt Code General occupation Renal cell Renal pelvis cancer cases SIRt cancer cases SIRt 20 Food products 167 21 Tobacco and beverages 29 22 Textile 100 23 Apparel 110 24 Lumber and wood 215 products 25 Furniture and fixtures 59 26 Paper and allied 202 products 27 Printing and 125 publishing 28 Leather 17 30 Rubber 28 31 Chemical 92 32 Coal and petroleum 12 refining 33 Stone, clay, glass 113 34 Fabricated metal 398 products 35 Machinery and 611 electronics 36 Transportation 335 equipment 37 Miscellaneous 67 manufacturing 39 Manufacturing, NEC 2 40 Home building 508 construction 41 Other construction 475 092 29 089 4 1 11 8 097 15 1-04 16 0-86 8 1-07 23 102 14 1-15 2 095 4 1-07 7 077 2 086 16 099 49 1 14* 85 1-10 34 1-19 14 0-76 0 1-02 52 1 02 61 1 41 70 1-08 71 0-84 72 1-25 73 0 70 74 0-94 75 1-12 76 102 77 1-23 78 1-10 79 0-74 80 1-16 81 1-10 82 1-17 83 1-42* 84 0-98 85 2-15** 86 - 87 095 88 1 20 Spinners, weavers, 57 and knitters Tailors, cutters, furriers 45 Shoe and leather workers 46 Metal making and 137 treating workers Precision instrument 35 makers Toolmakers, machinists, 699 plumbers, welders Electricians, electronic 173 workers Woodworkers 452 Painters, paperhangers Bricklayer and construction worker, NEC Graphical worker Potters, kilnmen, and glass workers Food industry worker's Chemical and cellulose workers Tobacco workers 124 280 55 24 108 107 0 Craftsmen & production 77 process workers Labourers, NEC 191 Stationary engine and equipment operators Stationary engine and equipment operator 170 233 1 24 3 0-79 7 1-06 5 0-96 20 1-07 4 1-07 95 108 22 1-02 37 094 10 090 37 084 11 073 8 095 16 096 14 -0 0.95 10 083 25 115 17 0-93 39 0-63 1-12 1-07 1-34 1 11 1 31 1-29 075 069 1-08 1.50 2-12 125 1 15 - 1 09 099 1-07 1-38 *p < 001; **p < 005. tAdjusted for age and region. NEC = Not elsewhere classified. Occupational risks for renal cancer in Sweden 121 Table 3 Statistically significant (p < 0 05) standardised incidence ratios (SIR) for renal cell and renal pelvis cancersfor specific (three digit) industries and occupation among men Code Specific industry Cases 015 Veterinary medicine 7 220 Wool industry 33 350 Machine industry 435 352 Other electronics industry 165 362 Railroad and trolley works 48 412 Building sheetmetal construction 20 702 Automobile transportation 65 804 Police work 45 820 Health care 851 Fine arts and music composition 853 Theatres 103 20 11 102 Other ore mining 350 Machine industry 370 Scientific and surgical instrument construction 413 Plumbing installation 802 Legal services 853 Theatres *p < 0-01. tAdjusted for age and region. 4 70 6 18 7 3 SlRt Code Specific occupation Cases Renal cell cancer 2-59 006 1 61 031 1-13 054 1-20 085 1-45 111 179 294 1-33 313 1-45 755 902 1-28 981 1-75 Engineers and technicians in 80 other technical fields Physicians 33 Teachers of practical subjects 21 Journalists and editors 34 Business executives 201 Dispatchers and shipping agents 28 Advertisers 13 Welders and metal cutters 71 Police 41 Military service 52 2 14 Renal pelvis cancer 4-49 005 Engineers and technicians in mining and metallurgy 1.60* 071 Judges 5.18* 331 Wholesale buyers 6 5 19 2.37* 750 4-67* 754 5-23 Machinists and toolmakers Plumbers 31 15 SIRt 1-28 1-48 1 71 1.88* 1.23* 1.99* 1-98 1 31 1-52 1-35 3-29 9.16* 1 91 152 2 17 among blue collar occupations (craftsmen, production workers, and labourers, SIR = 1 22, p < 0 05) and for mining and manufacturing industries, although non-significantly raised SIRs were seen among professional and service workers. For both renal cell and renal pelvis cancers, significantly decreased SIRs were found among farmers, forestry, fishing, and related workers. Because previous clues to occupational factors centered on jobs and exposures in manufacturing, SIRs were calculated for all general (two digit) manufacturing industries and craftsman tradesman occupations (table 2). Significantly raised risks for both cancers were seen for men in the machinery and electronics industry. SIRs exceeding 10 were more often observed for renal pelvis cancer, however, with the excesses being significant among those employed in miscellaneous manufacturing, and as toolmakers, machinists, plumbers, and welders. Table 3 (top panel) shows in which specific (three digit) industries and occupations there are significantly raised (p < 0 05) SIRs associated with renal cell cancer. Several health related professions and the health care industry appear to have a raised risk for renal cell cancer. A significantly high SIR was found for physicians and the risk for dentists was also raised (SIR = 1-62; 20 cases) and of borderline statistical significance (p = 0-06). For the general (two digit) occupation of medical professionals, which combines physicians and dentists, the SIR was increased and highly significant (SIR = 1-53; 53 cases; p < 0-01). Most of the other occupations with raised SIRs were white collar professions. The bottom panel of table 3 shows specific industries and occupations with significantly raised SIRs for renal pelvis cancer. SIRs for this cancer were generally higher than those for renal cell cancer, although based on many fewer observations. Discussion This hypothesis generating study showed several statistically significant occupational associations for cancers of the renal parenchyma and renal pelvis. In general, the occupational patterns of renal cell cancer differed from those of renal pelvis cancer. The SIRs for renal cell cancer tended to be high among white collar workers whereas the SIRs for renal pelvis cancer were generally higher among blue collar employees. The increased risk of renal cell cancer among health professionals (physicians, dentists, and veterinary surgeons) is consistent with findings from a recent occupational mortality analysis in the United States.13 It is not clear whether this association reflects occupational exposures or, perhaps more likely, lifestyle risk factors for this cancer, including smoking habits, body weight, meat and fat consumption, analgesic use, or diagnostic surveillance.8 14 15 Tobacco smoking appears an unlikely 122 McLaughlin, Malker, Stone, Weiner, Malker, Ericsson, Blot, Fraumeni explanation, however, since smoking rates for Swed- ing, however, was higher among individuals in the ish health professionals in 1963 were comparable with specific occupations of toolmakers and machinists those for the general population (J Carstensen, per- (59% smokers) and plumbers (57% smokers) than sonal communication). The high rates observed among the general male population (46% smokers) (J among welders, military personnel, journalists and Carstensen, personal communication). Smoking may editors, and dispatchers and shipping agents, have thus partly account for the excess risks associated been reported in previous occupational surveys. 16 - 18 with these job categories. Smoking may also con- Some occupational categories have been previously tribute to the deficit of both cancers (most pro- linked to renal cell cancer but did not emerge as high nounced for renal pelvis cancer) among farmers, since risk groups in the present survey. For example, the as a group, farmers smoke less.2' The multiple com- raised risk observed among coke oven workers in the parisons made in this study also affect the inter- United States2 was not detected in the categories of pretation of the findings, since there were few a priori iron and steel making (SIR = 0 98; 145 cases) or leads and a large number of employment categories hearth and furnace workers (SIR = 0-94; 7 cases), were reviewed. It is difficult to know, therefore, although no specific code exists for coke oven work- whether to place much credibility in statistically ers. The SIRs for the laundry and dry cleaning indus- significant associations for specific (three digit) indus- try were 099 (18 cases) among men and 0-86 (25 tries or occupations when almost 800 such categories cases) among women, indicating no increased risk were studied for the two cancers, and many associ- such as has been observed in the United States.34 ations are expected on the basis of chance alone. It is Unlike previous reports,5 we found no excess risk noteworthy, however, that several findings of our sur- among printing or graphical workers (SIR = 0-83; 47 vey are consistent with previously reported associ- cases), insulation workers (SIR = 0-98; 2 cases), or ations for renal cancer. among workers in petroleum refineries (SIR = 0-92; 8 cases) or gasoline stations (SIR = 0 59; 9 cases). Requests for reprints to: Dr J K McLaughlin, Our earlier communication on kidney cancer Biostatistics Branch, National Cancer Institute, reported excesses of renal cell cancer among tanners Landow Building, Room 3C16, Bethesda, Maryland and of renal pelvis cancer among shoe factory work- 20892. ers between 1961 and 1973.19 With the longer period of follow up, these associations persisted but were no longer statistically significant among tanners (SIR = References 156; 9 cases) or shoe factory workers (SIR = 2-68; 4 I McLaughlin JK, Schuman LM. Epidemiology of renal cell car- cases). The inclusion of women in the latter category, cinoma. In: Lilienfeld AM, ed. Reviews in however, did result in a significant association (SIR = 2-82; 6 cases). This appears consistent with a case- control study indicating that leather workers are prone to renal pelvis cancer.9 The significantly raised cancer epidemiology. Vol 2. New York: Elsevier, 1983:170-210. 2 Redmond CK, Ciocco A, Lloyd W, Ruch HW. Long-term mortality study of steelworkers. VI. Mortality from malignant neoplasms among coke oven workers. J Occup Med SIRs among machinists and toolmakers, plumbers, and workers in the machine industry are new aetiological leads, although machinists and plumbers are reported to be at increased risk of bladder can- 1972;14:621-9. 3 Katz RM, Jowett D. Female laundry and dry cleaning workers in Wisconsin: a mortality analysis. Am J Public Health 1981 ;71:305-7. 4 Duh RW, Asal NR. Mortality among laundry and dry cleaning cer,20 a tumour with risk factors resembling those of renal pelvis cancer. 1 Although this record linkage study may provide occupational clues to the origins of renal cancer, the workers in Oklahoma. Am J Public Health 1984;74:1278-80. 5 Paganini-Hill A, Glazer E, Henderson BE, Ross RK. Cause- specific mortality among newspaper web pressmen. J Occup Med 1980;22:542-4. 6 Selikoff IJ, Hammond EC, Seidman M. Mortality experience of CER has limitations that preclude making causal inferences. 2 For example, there is no information on the duration of employment or on non-occupational factors that may influence the risk of renal cancer. insulation workers in the United States and Canada, 1943-76. Ann NY Acad Sci 1979;330:91-116. 7 Savitz DA, Moure R. Cancer risk among oil refinery workers: a review of epidemiologic studies. J Occup Med 1984;26:662-70. 8 McLaughlin JK, Mandel JS, Blot WJ, Schuman LM, Mehl ES, Since the effect of cigarette smoking appears greater for renal pelvis cancer10 than renal cell cancer,8 occupational differences in smoking prevalence Fraumeni JF, Jr. A population-based case-control study of renal cell carcinoma. JNCI 1984;72:275-84. 9 Schmauz R, Cole P. Epidemiology of cancer of the renal pelvis and ureter. JNCI 1974;52:1431-4. would more likely influence the associations with 10 McLaughlin JK, Blot WJ, Mandel JS, Schuman LM, Mehl ES, renal pelvis cancer. The 1963 survey of smoking habits in Sweden found similar proportions of smok- ers in both blue collar (exclusive of agricultural jobs) and white collar workers.2' The prevalence of smok- Fraumeni JF, Jr. Etiology of cancer of the renal pelvis. JNCI 1983;71:287-91. II Wiklund K, Einhorn J, Wennstrom G, Rappaport E. A Swedish cancer environment registry available for research. Scand J Work Environ Health 1981;7:64-7. Occupational risks for renal cancer in Sweden 12 Malker HSR, McLaughlin JK, Malker BK, et al. Biliary tract cancer and occupation in Sweden. Br J Ind Med 1986;43:257-62. 13 Dubrow R, Wegman DH. Occupational characteristics of white male cancer victims in Massachusetts 1971-3. Cincinnati: Department of Health and Human Services, National Institute of Occupational Safety and Health, 1984. 14 Wynder EL, Mabuchi K, Whitmore WF, Jr. Epidemiology of adenocarcinoma of the kidney. JNCI 1974;53:1619-34. 15 Curtis RE, Boice JD, Jr, Kleinerman RA, Flannery JT, Fraumeni JF, Jr. Summary: multiple primary cancers in Connecticut, 1935-82. NatI Cancer Inst Monogr 1985;68:219-42. 16 Milham S. Occupational mortality in Washington state, 1950-79. Cincinnati: Department of Health and Human Services, National Institute of Occupational Safety and Health, 1983. 123 17 Dubrow R, Wegman DH. Setting priorities for occupational cancer research and control: synthesis of the results of occupational disease surveillance studies. JNCI 1983;71:1 123-42. 18 Petersen GR, Milham S. Occuptional mortality in the state ofCalifornia 1959-61. Cincinnati: Department of Health, Education, and Welfare, National Institute of Occupational Safety and Health, 1980. 19 Malker HSR, Malker BK, McLaughlin JK, Blot WJ. Kidney cancer among leather workers. Lancet 1984;i:56. 20 Matanoski GM, Elliott EA. Bladder cancer epidemiology. Epidemiol Rev 1981;3:203-29. 21 Survey Research Institute. The cigarette-a study ofpersonality and motives. Stockholm: National Control Bureau of Statistics, 1972. (English summary only.)