Document zQj972J70dK1bnJg997oBmJ6g

Scand J Work Environ Health 1989;15:38"/--394 Incidence of cancer among welders and other workers in a Norwegian shipyard by Anne Johan N SMteralkyilTdo, nMnDe,s1seSnv,eMrrDe3Lang&rd, MD, MSc,1 Aage Andersen? MELKILD A, LANGARD S, ANDERSEN A, STRAY TDNNESSEN JN. Incidence of cancer among welders and other workers in a Norwegian shipyard. Scand J Work Environ Health 1989;15:387--394. The incidence of cancer among 4778 male shipyard workers, including 783 mild steel welders, was investigated in a historical cohort study. The workers had been employed for at least three months between ! January 1946 and 31 March 1977. The incidence of cancer was observed from 1 January 1953 through 1986. The loss during fol!ow-up was only 0.9 ~/o. There were 53 observed cases of lung cancer in the whole cohort versus 31.3 expected on the basis of the national rates for men. There was an increased incidence of lung cancer among the welders, with seven observed cases versus 3.2 expected. Twenty-two cases of bladder cancer were observed versus 15.2 expected. Two malignant mesotheliomas had occurred (0.7 expected). Smoking was likely to be a confounder in the present study. Due to concomitant exposure to asbestos, the results are inconclusive concerning the possible relationship between exposure to welding fumes and lung cancer. Key lerms: asbestos, bladder cancer, cohort study, lung cancer, welding. There has been concern for some years about a possi- yard workers. The welders were all shipyard workers ble excess of cancer, especially lung cancer, among who had mainly welded mild steel. The recognized ex- welders. Several epidemiologic studies indicate that cess of lung cancer in the shipbuilding industry is gener- welders exhibit a higher incidence of lung cancer than ally attributed to asbestos exposure (9, 17--22). Hence selected reference populations (I--9), while others have it was considered important to include contemporary not reproduced these findihgs (10, ! i). Asbestos and workers at the shipyard, not only for a possibility to smoking have been possible confounders in most of study their incidence of cancer as such, but also to serve the studies (12). it has, however, been suggested that as internal reference groups for the welders. welders in general exhibit ~n approximately 30 % greater incidence of lung.ca..n.c.~.~'.(than the nonwelding populatiou) that cannot be ascribed to smoking or other occupational hazards (13). Background information Fumes from stainless steel welding are mutagenic in Shipbuilding bacteria tests (14) and cause transformation in mammalian cell transformation assays (15). Fumes from mild steel welding have been found to be inactive in these test systems (15). O'n the basis of this back- ground, it has been suggested that the observed excess o[" lung cancer among welders can be ascribed to highrisk subgroups who weld stainless steel and are thus exposed to hexavalent chromium (16). If this suggestion is correct, one could expecl a cohort of mild steel \t elders to exhibit only a moderate or no excess of lung The study was conducted at a shipyard located on the southwestern coast of Norway. The yard was first es- tablished in 1896. Initially the main products were fishing boats and steamboilers. The company expanded rapidly after World War II, partly due to a large amount of repair work. A total of 953 ships were repaired from 1947 to 1951. Having produced mainly freighters after the war, the shipyard specialized in building oil tankers from 1950 onwards. From 1971 to 1979 the yard's main product was gas carriers. Since cancer. -1978 the major products have been large offshore The purpose of the present study was to investigate structures for the North Sea oil industry. the incidence of cancer among welders and other ship- New ships were built in sections. Steel plates and Tdemark Central Hospital, Department oi' Occupational Medicine, Porsgrunn, Norway. The Cancer Registry of Norway, Momebello, Oslo, NorHcallh Deparlment. Rosenberg Verfl, Stavanger, Norway. pipes first entered the workshops for cutting, welding, and assembly work. The assemblies were then trans- ported to the docks for completion, including fitting out and insulation. Repair work was performed at a separate dock, but parts, including asbestos-covered Reprint requests ~o: Dr S Lang~rd, Telemark Cenlral Hospita!. Department of Occupalional Medicine, N-3900 Pots~rimn, Norway. pipes, were transported to the workshops for further handling'. The main workshops at present are the plate shop covering 10 000 m~, the welding shop (9000 m~), 387 CAT-1016793 029 and the pipe shop (3700 m~). The yard has a separate machine shop of 2000 m~ and sandblasting and painting halls. Ventilation and personal protection In the 1940s there was a system for leading fresh air passively into confined spaces during welding at the docks. This system was replaced with fans in the 1950s. Welding In the beginning the welders used dust masl~s made of felt and later switched to carbon filters. After exhaust Only a few workers welded before 1940, and riveting ventilation was provided at the docks from the mid- and welding were performed at the same time during 1960s on, the conditions improved markedly. Personal World War It and the subsequent years. The number respiratory protection and exhaust ventilation were of welders increased substantially during the postwar provided in the shops in the early 1970s. period. Metal-arc welding was the method mostly ap- Workers who regularly worked in the vicinity of a plied in the first years of welding. Rutile and acidic- welder were also exposed to welding fumes, but at coated electrodes were used in the 1940s. In the early lower concentrations. These workers mainly included 1950s basic-coated electrodes partly replaced the ru- pipe fitters (who performed some welding themselves), tile and acidic-coated electrodes and remained the most shipwrights, burners (also exposed to fumes from frequently used electrode. Gas-shielded welding was employed from the late sixties, initially metal inert-gas acetylene cutting), and their helpers and apprentices. and metal active-~as welding. Later, from the mid1970s on, tungsten inert-gas welding also came into Use of asbestos USe. Asbestos was used at the yard until the early 1970s for Up until the early 1970s welding was Pe~'formed almost exclusively on mild steel. During the production insulating in machine rooms and for personal protec- tiom Chrysotile was almost exclusively used. All insulating work with asbestos was carried out by the of gas carriers, two ships delivered in 1973 and 1974 were equipped with tanks consisting of 9 % nickel steel This steel contained.no more than 0.2 % chromium. yard's own employees until the late 1960s, when an outside firm was engaged to insulate the new ships. In ships under repair asbestos was removed by the Three ships (1975--1977) were equipped with alumi- num tanks. Welding on stainlexs steel became more and more common during this period, particularly in pipe welding. Stainless steel welding is now widely applied at the yard. However) up to 1973--1974, there had been virtually no stainless steel welding at the shipyard. boilermakers and, io some extent, by pipe fitters, mechanics, welders, shipwrights, and by their helpers and apprentices. Some asbestos was removed in the workshops, especially in the pipe shop. Insulation was mainly performed by the boilermakers and, to some extent, by pipe fitters. Boilers and pipes were covered with asbestos plates, and the joints were filled with as- Environmen(al monitoring In 1973 a survey of welding fume exposure was findertaken, initiated by concern about the untoward health effects of exposure to nickel-containing welding fumes. The chromium and niclcel content of the deposit was not known, but the welding wire that was most frequemly used contained 70 ~o nickel and 13 % bestos powder mixed with a binding agent. This mix- lure was made on the spot and generated substantial amounts of dust. Measurements of asbestos dust concentrations from the period of interest were not available. Presumably the docks have been the area most contaminated by asbestos, followed by the pipe shop. chromium. Nine welders using metal-arc or metal inert- gas techniques on steel with a content of 9 % nickel Use of primers were monitored for one week. The total welding fume concentration and the nickel and chromium content of the fumes was measured with personal samples. Welding was performed at the docks (in gas carrier tanks), in the p!ate and pipe shops, and in open and confined spaces. Total fume concentrations ranged from 3.6 to 23.6 mg/m), taken as weekly means, with 7.3 mg/m) as the median value for all the welders. No correlation was found between the fume concentration and the place of work. The concentrations of nickel were between 0. I I and 1.97 (median 0.43) mg/m~, and the In general the steel was initially unprimed. However, lead (not lead chromate) was applied to some extent. Zinc chromate was introduced for priming ship shells at the docks in the mid-1950s, and the application was performed by painters and dock workers. Some welding was performed on these primed shells. Iron-based primers were introduced in the 1960s and coincided with the priming being performed before the steel plates entered the shops for ~ttting and assembly work. Zinc has also been a fume constituent in the occasional welding of galvanized steel. total chromium content ranged from 0.03 to 0.65 (me- dian 0.12) mg/m'. An analysis for hexavalent chro- Subjects and methods mium was not carried out. No measurements were The personnel register was used for compiling the list available for earlier periods. of names of the employees at the shipyard. The regis- 388 ter em the plo Th, ist( cat mit od~ 31 I We~ Th 31 ch~ me ten an at for 12~ em co~ sht feb mt w ed tit to pl: pe vh in, jo ca pe sn tit CAT-1016794 o29 ter was established in 1944--1945 and included both operators, foremen, tool makers, welding supervisors, dr employees working at that time and new workers. For electricians). he the present study data on all employees with an em- The Cancer Registry of Norway has recorded all new ployment time exceeding three months were collected. cases of cancer since 1 January 1953. The registration of The following information was obtained from the reg- is more than 99 % complete. In the present study the ister: name, address, date of birth, personal identifi- incidence of cancer among the national male popula- d- cation number if known, dates for starting and ter- tion serves as a reference entity. The expected num:al minating work, and job titles for all employment peri- ber of selected cancers was estimated with the use of ods. Information on employment was complete until national five-year age-specific inddence rates for each 31 August 1987. year from 1953 to 1986. Events occurring between 1946 a Information was obtained on 6239 workers. Five were excluded due to missing names or date of birth. to 1952 were not available. Each worker was under observation from the ~ginning of 19~3 or, if he was first ed The study was restricted to men who were first em- employed at a later date, from the middle of the year ployed at the shipyard between 1 January 1946 and he was first employed until the occurrence of death, m 31 December 1976, leaving 488? eligible for further emigration, or the end of follow-up. analysis. The starting date of i January 1946 was Several workers had held more than one job at the chosen because it coincided both with the establishment of an informative personnel register, which is cur- yard. All calculations were made with each occupational category being viewed as a separate entity. Thus rently in use, and the introduction of welding at the one worker may have contributed an observed cancer yard. The closing date o f 3 ! December 1976 allows for a minimum assumed development time for cancer of to more than one occupational category, but not more than once within each category or when the incidence nle tn at least 10 years for all participants. The [53 female workers were excluded, as were 41 foreigners no longer resident in the country, and 24 of cancer of the total cohoi't was considered. Confidence intervals were estimated on the assump- tion of a Poisson distribution of events with the use of a two-sided test of significance. workers who were deceased or had emigrated before ,re the establishment of the Cancer Registr), in 1953. Those s~ 126 who had emigrated after 1 January 1953 were con- rs sidered to be under follow-up until their date of Results emigration. Only 44 (0.9 %) of the eligible subjects There were 716 deaths from all causes in the cohort ie could not be traced. The formation of the cohort is versus 681.3 expected, and 252 cases of cancer were shown in table I. observed versus 244.1 expected (table 2). s- Retired workers, including a former administrator Fifty-three ~ases of lung cancer were observed in the from the personnel office, were temporarily engaged whole cohort against 3 ! .3 expected, which is statisti- to evaluate_the quality of the personnel register with cally significant at the 99 % level (table 2). The inci- respect to job titles and approximate dates of employ- dence of bladder cancer was also elevated, 22 cas~s m ment. Job titles had changed during the study period. sus 15.2 expected. The number of observed cases of ly With the assistance of these retirees' thorough knowl- colon cancer was significantly lower than expected (ob~y edge of previous work conditions, the diversity of job served 8, expected 17.9), but the difference was less titles in the register could be limited to 32. The inclu- apparent when cancers of the colon and rectum were sion criteria were set and the assignment of workers merged (observed 24, expected 30.2). There was also to the final job titles was completed before the data a deficit of stomach cancer (observed 14, expected on health status were collected. 22.1). On the basis of common exposures at the study [. plant, all job titles were merged into l0 different oc- .Is cupational categories. These categories contribute more ~s person-years than the single job titles and thus pro- Table 1. Formation of the cohort. J- vide a more robust basis for estimating an expected "d incidence of cancer. The occupational categories and N :d job titles were as follows: welders; metal workers, oc- Initial numbers of personal records 6239 el casionally welding (caulker burners, pipe filters, cop- Incomplete data k. persmiths); metal workers, seldom or never welding Women 5 153 al (shipwrights, riveters, boilermakers, grinders, black- Men employed betore 1 January 1946 or smiths); machine shop workers (mechanics, turners, after 31 December 1976 Men employed between 1 January 1946 1194 fliers); carpenters (carpenters, joiners, pattern mak- and 31 December 1976 4887 ers); dock workers (laborers, scaffolders, dock workers); sandblasters and painters; assistants and appren- tices; office and administrative personnel; and others (transport workers, guards, dustmen, storemen, crane Foreigners Dead or emigrated before 1 January 1953 Unidentilied Ultimate cohort 41 24 44 4778 389 CAT-1016795 o29 Table 2. Observed and expected cases of cancer at selected sites and total mortalil~Y due !o cancer among shipyard workers. (SIR = standardized incidence ratio, 95 % CI = 95 % confidence Interval) Cancer aile~ Obse~ed Expected SIR g5 % Cl Stomach (151) Colon (153). Rectum (154) Nose (160) Larynx (161) Lung (162) Mesothelloma (163) Kidney (180) aladder (181) All cancer (140--204) t4 22.1 63 35--106 8 17.9 45" 20--89 16 12.3 130 74--211 2 1.0 211 21--758 6 3.6 167 61--364 53 31.3 169"" 127--217 2 0.7 267 27--960 5 9.6 52 17~12t 22 15.2 145 91--220 252 244.1 103 90--116 Code of the Inlernatlonal Classification of Diseases (sevenlh revision} in parentheses, P<0.05, P<0.0t. Two cases of malignant mesothelioma had occurred (expected 0.7), one in a turner employed in 1946 for less than one year and one in a shipwright employed from 1950 to 1965. The time between first employment at the yard and the occurrence of cancer was 34 and 35 years, respectively. Lung cancer was in excess in several occupational categories (table 3). The excess was statistically significant among the machine shop workers (observed 8, expected 2.9) and among the assistants and apprentices (observed 38, expected 19.2). As most of the worl~ers started out as unskilled, this latter group comprises three-fourths of the entire cohort. Therefore, for all types of cancer, the results of" this group closely resem- Table 3. Observed and expected cases of lung cancer Icode 162 of the International Classilication of Diseases (seventh revl- slon)] among the shipy~d confidence interval) :~ workers according to their occupational category. (SIR = standardized - incidence ratio ' 95 %~ CI = 95 % OccUpational category Number . of Observed Expected workers SIR 95 % CI Person-years (1953-- 1986) Welders Metal workers occasionally welding Metal workers seldom/never welding Machine shop workers Carpenters Dock workers Sandblasters, painters Assistants and apprentices Office and administrative workers Others Total 783 7 3.2 244 1 1.8 722 6 2.8 279 8 2.9 105 -- 1.3 253 7 3.3 43 0.4 3610 ~ 19.2. 304 1.2 684 "~ 6.4 4778 53 31.3 221 57 214 276" 210 198"" 109 169"" 88--454 ----316 79--468 121--545 "84~432 140--272 44--224 127--217 15 807.0 5 583.0 16 460.0 7 384.5 2 652.0 5 754.0 898.5 84 892.0 6 597.5 16 566.5 110407.0 'P<O.Of,"P<0.01. Table 4. Observed and expected cases of cancer at selected sites among the shipyard workers according Io their occupational category. Occupational calegory Welders Metal workers occasionally welding Metal workers seldom or never welding Machine shop workers Carpenters Dock workers Sandblasters, painters Assistants and apprentices Office and adminis. trative workers Others Total Stomach Colon Ob- Ex. Ob. Ex. served pected served pected 1 1.8 2 1,7 1 1.2 1 1.0 1' 1.8 2 2.5 3 1.1 1 2.8 1 0.3 6 12.4 -- 0.7 --" 4.6 14 22.1 1 1.6 -- 1.8 -- 0.8 2 2.0 -- 0.2 2"" 10.7 1 0,7 3 3.7 8" 17.9 Cancer site Reclum Kidney Ob- Ex- Oh- Ex- served pected served pected 1 1.2 -- 1.0 2 0.7 -- 0.5 1 1.1 0 1.2 " 0 0.5 5" 1.3 0 0,2 10 7.4 0 0.5 2 2.5 16 12.3 1 1 1 --- 2 -1 5 0.9 0.9 0.4 1,0 0.1 6.1 0.4 1.9 9,6 Bladder All sites Ob- Ex- Ob- Ex- served pected served pected 2 1,5 30 24.5 1 0.9 12 13.6 4 1.3 1 1,4 -- 0.6 3 1,7 -- 0,2 15 9.1 1 0.6 6 3,2 22 15.2 26 22.9 27 23.5 6 I0.0 31 26.0 1 3.0 166 149.7 9 9,6 45 48.9 252 244.1 "P<O.Of,"P<0.01. 390 bled those of th, cidence ratios fo the welders (obs( ers welding onl.~ The excess in. pronounced am ployed in 1950P<0.01) and in P<0.05). The ei the machine sh, ers firs( employ of lung cancer ~ subcohort of w Table 4 show: by occupationa significant varia occupational gl cally significanl docl~ workers. was ~nerged wi excess was no [, 3.3). A deficit the assistants a Table 5. Observe men(. A latency interval) Cancel site Stomach (1511 Colon 0.53) Rectum (154) Nose (t60) Larynx (161) -- " Lung (162) Kidney (180} Bladder O81) All cancer Code o! the P < 0.01. Table 6. Obser sion)l for selec o( 10 years Occupational category Welders Metal workers occasionally v Metal workers or never weld Machine ShOl: Dock workers Assistanls an apprentices P<0.05, "" CAT-1016796 029 fred 6 for oycd merit and ~onal ~nifiI, tires ~ers raH selrt - revi95 % ears .0 .0 .0 .5. .0 .0 .5 0 5 5 0 ~nal bled those of the total cohort. High standardized inCidence ratios for lung cancer were also found among the welders (observed 7, expected 3.2}, the metal work- ers welding only rarely, and the dock workers. The excess incidence of lung cancer was the most pronounced among the shipyard workers first em- ployed in 1950--1959 (observed 27, expected 13.0, P<0.01) and in 1960--1969 (observed 8, expected 3.l, P < 0.05). The eight cases of lung cancer found among the machine shop workers had all occurred in work- ers first employed before 1960. There was no cluster of lung cancer cases by period of employment in the subcohort of welders. Table 4 shows the incidence of cancer at selected sites by occupational category. There was no statistically significant variation in risk for any cancer site between occupational groups, with the exception of ~ statistically significant excess of cancer of the rectum among dock workers. However, when cancer of the rectum was merged with cancer of the colon, the persistent excess was no longer significant (observed. 7., expected 3.3). A deficit of colon cancer was only seen among the assistants and apprentices. For cancer of the blad- der, the standardized incidence ratio (SIR) was nonsignificantly increased among the metal workers rarely welding (obser~'ed 4, expected 1.3). No statistically significant excess or deficit of cancer in all sites was found in any occupational category. To allow for different presumed development times for cancer, analyses were performed according to time since first employment. The incidence of lung cancer and cancer of the bladder in the whole cohort was sig- nificantly increased l0 years and more after first em- ployment in the yard. Seven of the eight lung cancers among the machine shop workers occurred 20 years or more after first employment. In table 5 the cohort has been stratified according to total length of employment, and table 6 shows the incidence of lung cancer by duration of employment in selected occupational categories. A presumed 10-year development time is simulated by the exclu- sion of person-years until l0 years after first employ- ment. There w. as no significant trend in risk with duration of employment for any cancer site except lung cancer (table 5). The excess incidence of lung cancer was the most pronounced for workers employed for Table mont. 5. A Observed and latency period eoxfp1e0ctyeedacrassheassob! ecaenncaecr caot usenlteecdtefodr.si(tSesIRam= ostna9ndtahredsizheipdyianrcdidweonoceersraaticoc,o9r5din%g to CI length ot crop|by= 95 % confidence interval) Lenglh of employmenl Cancer site* < 1 year 1--5 years Oh- Ex- sin 95 % CI sOehrv- eExd- SpIeRcted 9.5 % CI served peeled Stomach (151) Co, on (153l 4 4.8 1.2 ~13 - 23--213 3 1 5.7 52 11--154 5.0 20 --112 Nose (160) Larynx (161) Lung {162~ Kidney (180) Bladder(181) 2 0.2 918 1~0--3600 0.3 - 1 0.9 106 .-::-411 -~ 1.1 191 16--655 21 7.5 280" 173--428 14 9.1 154 85--258 -- 2.3 - 3 2.8 108 21--314 4 3.6 112 31--283 7 4.4 160 64--327 All cancer {140--204) 67 5.4.6 123 95--156 62 66.8 93 71--119 >5 years Oshe- Ervx-eSdIRperiod95 % Ct 6 8.6 70 26--152 6 7.2 83 3'{--182 0.4 -~ 1.4 213 43--629 15 12.6 119 67--196 1 3.6 27 --147 I1 6.3 175 87--313 103 94.6 109 88--131 Total Obs- eExr-vSeidR pecte9d5 % CI 13 19.1 68 3~ 116 7 13.4 52 21--108 2 0.9 222 22--800 6 3.4 178 65--385 50 29.2 171" 127--226 4 8.9 45 12--115 22 14.3 154 97--233 232 216.0 107 94--122 Code of the International Classilication of Diseases (seventh revisionl in parentheses. P<O01. ted Table 6. Observed and expected cases of lung cancer [code 162 of the International Classification of Diseases ~seventh revi4.5 sion)] tot selected occupational categories according to length of employment in the selected category only. A latency period of 10 years has been accounted for. 3.6 Length of e~ployrnent Occupational -~.9 i category < 1 year Observed Expected 1--5 years Obsertied Expected > 5 years Observed Expected Total Observed .Expected :,.0 Welders -- 0.4 5"" 0.9 1 1.7 6 2,9 Metal workers occasionally welding -- 0.1 -- 0.3 1 1.3 1 1.7 Metal workers seldom or never welding I 0.5 1 0.7 3 1.5 5 2.6 6 9 Machine shOp workers Dock workers 2 0.8 4" 0.8 4" 0.8 2 1.3 2 1.1 1 1.0 8" 2.7 7 3.1 Assistants and apprentices 18"" 6.3 13 8.8 5 3.0 3~"" 18.0 P<0.05, "P<0.01, 391 CAT-1016797 029 less than one year (observed 21, expected 7.5, SIR 280), A primary positive selection of workers may have and the standardized incidence ratio decreased to 119 affected the cohort at its inception, and this phenome- for workers employed for more than five years (table non could have concealed a possible association be- 5). The majority of these short-time employees had tween work-related exposure and illness. However, its been engaged as assistants or apprentices (observed 18, impact on the incidence of cancer is assumed to be expected 6.3), and the excess was also significantly limited. In some of the analyses a presumed "latency elevated among the short-time dock workers (observed period" of 10 years from first date of employment in 4, expected 0.8) (table 6). For employees with a dura- an occupational category was accounted for. Conse- tion of employment of one to five years there was a quently, these effect estimates have not been attenuated statistically significant excess of lung cancer among the by work experience etiologically unrelated to exposure welders (observed 5, expected 0.9) and the machine in the job or jobs at issue (23, 24). shop workers (observed 4, expected 0.8). The incidence The main finding of the present study was an ex- of cancers of the colon and rectum was marginally in- cess incidence of lung cancer in the total cohort of ship- creased among the Workers employed for more than yard workers (SIR 169). The magnitude of the excess five years (observed 14, expected 12.2) (table 5). was higher than that reported in some epidemiologic studies on shipyard employees (9, 17, 20), while other authors have demonstrated an approximately twofold risk of lung cancer for shipyard workers (19, 21, 22). Discussion ~~ Exposure !.o asbestos is likely to explain part of the excess of lnng cancer found in this cohort. The demon- In the pr~ent study an observed case of cancer Was stration of an increased incidence of lung cancer in ascribed to an occupational category when a workerseveral occupational categories at the shipyard is in achad at least three months of employment in one of thecordance with the results of other investigations (9, 17, jobs constituting the category. Assignment to one cat- 18, 20, 21). However, the results from these studies egory was independent of employment in another jobare inconsistent as to which occupational groups had at a different~riod of time at the yard. One workeran excess of asbestos-related cancers, probably due to may thus have contributed with a cancer case in morediverse routines for asbestos handling or possibly to than one occupational category. With the use of this methodological shortcomings. method the observations within the categories were not At the shipyard, asbestos had been handled mainly diluted by the pooling of the worker's total experience by the boilermakers, but also to some extent by the into one category only, and the observed number of pipe fitters, mechanics, welders, and shipwrights. By- cases in each category is a true representation of the cancers that may have occurred as a result of work exposure in the category at issue. The personnel register was the only source of information about each worker, cun-ent use of the register and the fact that it was established comparatively stander exposure was common, and the docks was probably the area where most of the asbestos was handled. As might be expected, an increased incidence of lung cancer was found among the machine shop workers, dock workers, metal workers who rarely welded (including boilermakers and shipwrights), and late gives us reason to trust ;Js completeness. The flies welders. Exposure to zinc chromate primer might have on former employees have been kept fully independent been a minor additional causal factor among the dock of the reasons each one had for terminating work, and workers (251. this practice implies that a possible loss of personnel The overall standardized incidence ratio for lung cards has not been dependent on outcome of illness cancer among the welders did not exceed the cor- or death. As the register contained a substantial num- responding standardized incidence ratio of the other ber of files on workers employed for less than three occupational categories. The mean age at employment months, we have no reason to believe that there is any and the secular distribution of person-years were not deficit of short-time employees in the cohort. so different as to invalidate tentative internal com- As a result of the control and corrections of the in- parisons. Five lung cancers versus 0.9 expected oc- dividual personnel,cards by former employees, mis- curred in the stibcohor! with I--5 years of employment classification of workers by the job title given in the . (table 6). Although this excess was highly significant, personnel register seems unlikely. the result is difficult to interpret as there was no ex- Only 44 (0.9 o70) of the workers were lost to follow- cess of lung cancer among the welders employed at the up. Twenty-nine of these were assistants or appren- shipyard for more than five years. The extent to which tices. Only 15 skilled workers were not traced. These possible previot~s welding experience at other shipyards 15 unidentified workers were scattered among several or in other industries has influenced the results is not occupational categories; hence a possible outcome de- known for the different welder subcohorts. pendency of the loss will have had very little influence There was no excess of lung cancer among the work- on the results. With respect to the emigrants it is un- ers assigned to the category of metal workers occasion- likely that the outcome under study has had any bear- ally welding (table 4). The pipe fitters, who had also ing on the decision to go abroad. been exposed to some extent to asbestos, constituted 392 a great pr not be exl employmt years. Wil pered by Tire ext hort decrc and was si le~s than ~ among lh anaou~ th homogellt rices was I stantial m been eng. workers I the overa skilled w~ which m; terminant places. F, was the n tion ofen focus on portanl c, marion ~ present c In 196formed i= The perc 77.6 70. 82 % ve~ ployees. son (27), b. asis of caused b dardizcd centage tio,a and tively. It likely. The ix ration o co. The vey of s= prccludt of smolstlbgrot, Lung (cg, Grt way. TI Ics,~ tha age Bril ~ICCDLlUl trics. Two 0.7 exp. CAT-1016798 029 have 3me- n be~ _'r~ o be ency ~nse- ~ated ~sure L ex- :hip- ~cess :ogic. .ther fold Z2). f the ~on- :r in ~I dies had 'e tO y to inly the By- was was :rice hop rely and lave ock ttng :orher lent not o- Int, ex- I It e rich ;rds ~tOI ~rk- OB- Iso ted a great proportion of these workers. The result can- has tolake into consideration that the figures for the not be explained in terms of variation in age at first general population are to a great extent due to asbestos employment or by the secular distribution of person: exposure_ There is consistent evidence that exposure years. With only 1.8 cases expected, the result is ham- to chrysotile, which may have occurred in the present pered by considerable uncertainty. cohort, is associated with a smaller risk of mesothelio- The excess incidence of lung cancer in the total co- ma than is exposure to amphiboles (28). As the esti- hort decreased as the length of employment increased mated mean development time is about 30 to 35 years, and was significant only for the workers employed for the observation period in the present study may be too less than one year (table 6). The excess of lung cancer short to reveal a possible excess. The small numbers among the short-term workers was the most evident preclude further interpretation. among the assistants and apprentices, who formed a in relation to the observed deficit of kidney cancer homogeneous group before the education of appren- (table 4), it should be noted that the regional figures tices was properly organized in the early 1960s. A sub- For the incidence of this tumor are low. The part of stantia[ number of these short-time employees had not the country where the shipyard is localed had a defi- been engaged in other jobs at the shipyard. These cit of 29 % of kidney cancer in the period 1966~i975 workers thus contributed several lung cancer cases to compared with the national incidence (29). the overall excess, h is likely that this groul~ oF un- The deficit of stomach cancer may also partly be ex- skilled workers exhibits an increased lung cancer risk plained by regional variations in the incidence of can- which may be related both to individual disease de- cer. The incidence of stomach cancer in the part of terminants and to exposure factors at other work- the counlry where the shipyard is located was 14 */o places. For skilled workers the excess of lung cancer lower than the national figures in the period 1966-- was the most prominent for the workers with a dura- 1975 (29). tion of employment of one to five years. These resuhs The excess incidence of cancer of the bladder (table focus on the possibility of smoking acting as an im- 4) is difficult to interpret in terms of occupational ex- portant confounder in this study. However, no infor- posure, and is suggestive of smoking acting as a con- mation was available on the smoking habits of the founder in this study. Exposure to asbestos and tobac- present cohort. co smoking may explain the excess of larynx cancer in 1964--1965 a survey of smoking habits was per- (table 5), but with only six cases observed the result Formed in the 1893--1927 Norwegian birth cohort (26). may be due to chance. The percentage of current plus previous smokers was It is unlikely that exposure to welding Fumes con- 77.6 %. Industrial workers exhibited a prevalence oF stitutes a major lung cancer hazard in the present 82 ~/0 versus 85 q0 for office and administrative em- study, but a small excess of lung cancer caused by ployees. According to the method described by Axef- welding fume exposure may well be present in the son (27l, the degree of confounding determined on the Figures. Further follow-up with identification of those basis of these numbers with regard Io lung cancer exposed to fumes from stainless steel welding from the caused by smoking alone amounts to a maximum stan- early i 970s onwards, when also asbestos ceased to be dardized incidence ratio of 110. A standardi.zed inci- used at the shipyard, is needed to determine whether dence ratio of about 170 ,,,ould correspond to a per- the excess of lung cancer among the welders is related centage of nonsmokers among the reference popula- to stainless steel welding. tion and the shipyard employees orS0 and 20, respec- tively. In the present study such a difference seems un- likely. Acknowledgmenls The Axelson method (27) does not account for duration of smoking and lifetime consumption of tobacco. The uncertainties in applying the 1964--1965 sur- vey of smokiug (26) to the presem shipyard cohorl also preclude firm couc[usions concerning the contribution of smoking to tl~,: lung cancer excess in the dil'fcrem subgroups. This project was financed by Bundesministerium for Forschung und Technologie and the Royal Norwegian Council for Scientific and Industrial Research. We thank the staff of the shipyard Moss Rosenberg VerFt for their cooperation throughout the study, es- pecially Ms IH Hah'orsen at the medical office. We warmly thank Mr P Theie and Mr P Berg fbr their Lung cancer rates in some other European covmries compt, ter assistance, his P Flor for her linguistic help, (cg, Grca! P, ritain) arc three times as high as in Norway. The h, ng cancer rate of workers employed for and his K Theie for her typiug assistance. less than one year was thtts sintilar to that of an aver- age British man. Differences in smoking could in fact accoum For larger differences in hmg cancer rates in Re [erences Norway than ill inosl olh~r western European couu-I. Beaumont JJ. Weiss NS. Lung cancer among welders. trics. Two cases oF ntcsolhelioma were observed againsl J Occup Med 1981;23:839~44. 2. Becket N, Clande J. Frentzd-Be.vme R. Cancer risk of arc ~elders exposed Io fumes containing chromium and 03 expected, When judging the expecled figures, one nickel. Stand J \Vork Environ Health 1985;I 1:75--82. 393 CAT-1016799 029 3, Breslow L, Hoaglin L, Rasmussen G, Abrams HK, Occupations and cigarette smoking as factors in lung cancer. Am J Public Health 1954;44:171--81. 4. Lerchen ML, Wiggins CL, Samel JM. Lung cancer and occupation in New Mexico. J Natl Cancer Insl 198"/; 79:639--.-45. 5. Menck HR, Henderson BE. Occupational differences in rates of lung cancer. J Occup Meal 1976;18:797--801. 6. Polcdnak AP. Mortality among welders, including a group exposed Io nickel oxides. Arch Environ Health 1981 ;36:235--42. 7. SjOgren B. A relrospective cohort study of mortality among stainless steel wdders. Scand J Work Environ Health ! 9g0;6:197--200. 8. SjOgren B, Gustavssou A, Hedstr6m L. Mortalily in two cohorts of welders exposed 1o high- and low-levels of hexavalent chromium. Scand J Work Environ Health 1987; 13:247--51. 9. Tola S, Ka!liom~iki P-L, Pukkala E, Asp S, Korkala M-L. Incidence of cancer among welders, platers, machinists, and pipe filter~ in shipyards and machine shops. Br .i lnd Meal 1988;45:209--18. 10. Dunn JE, Weir JM. A prospective sludy of mortality of several occupational groups: special emphasis on lung cancer. Arch Environ Health 1968;17:71--6. I I. Ott M, Holder BB, Langner RR. Determinants of mortality in an industrial population. J Occup Med 1976; 18:171--7. 12. Lang~rd S. Confounding and effect-modifying factors in epidemiological studies in welders. In: Stern RM, Berlin A, Fletcher AC, .i~rvisalo J, ed. Proceedings of the international conference on health hazards and biolog- ical effects of welding fumes and gases, Copenhagen, 18--21 February, 1985. Amsterdam: Excerpla Medica, 1986:439--43. 13. Peto J. Cancer morbidity and mortality studies of welders. In: Stern RM, Berlin A, Fletcher AC, J,arvisaIo J, ed, Proceedings of the inlernational conference on health hazards and biological effects of welding fumes and gases, Copenhagen, 18--21 February, 1985. Amsterdam: Excerpta Medica, 1986:423--34. 14. Maxild J, Andersen M, Kid P, Slern RM. Mulagenicity of fume particles from metal arc welding on stainless steel in the Salmonellaimicrosome test. Mmat Res 1978; 56:235--,I-3. -15. Hansen K, Stern RM. Welding fumes and chromium compounds in cell ~ransformation assays. J Appl Toy icol 1985;5:306--14. 16. Stern RM. Assessment of risk of lung cancer for welders. Arch Environ Heah h 1983:38:148--55. 17. Beaumont J J, Weiss NS. Mortality of welders, shipfitlets and other metal trades workers in Boilermakers Local no 104, AFL-CIO. Am J Epidemiol 1980;I 12: 775--86. 18. BIo~ WJ, Morris LE, Stroubc R, Tagnon I, Fraumeni .IF. Lung and laryngeal cancers in relation to shipyard employment in costal Virginia. J Nail Caucer lust 1980; 65:571--5. 19. Kolonel LN, Yoshizawa CN, Hirohata T, Myers BC. Cancer occurrence in shipyard workers exposed to beslos in Hawaii. Cancer Res 1985;45:3924--8. 20. Newhous ML, Oakes D, Woollcy AJ. Mortality of welders and other craftsmen al a shipyard in NE England. Br J Ind Med 1985;42:406--10. 21. Puntoni R, Vercelli M, Merlo F, Valerio F, Sand L. Mortality among shipyard workers in Genoa, Ilaly. Ann NY Acad Sci 1979;330:353--77. 22. Sanden ,.~, N~slund PE, Jlirvholm B. Mortalily iu lung and gastrointestinal cancer among shipyard workers. Im Arch Occup Environ Health 1985;55:277--83. 2.]. Langfird S, Andersen AA. Betydningen av bruk av latcnslid, kritisk alder og valg av referansepopulasjon red beregning av cancerinsidens i kohort-sludier IThe import ance of use of latency period, critical age, and choice of reference populations for the estimation of cancer incidence in cohort studies]. In: Arl:w.idsforskningsinstituttene, 29[h Nordic meeting for occupational hygiene. H~nefoss, Norway, 3--5 November 1980. Oslo: Arbeldsforskningsinstiluttene, 1980:92. 24. Rothman KJ. Induction and latent periods. Am J Epidemiol 1981;I 14:253--9. 25. Dalager NA, Mason TJ, Fraumeni JF, Hoover R, Payne WW. Cancer mortality among workers exposed to zinc chromate paints. J Occup Med 1980;22:25--9. 26. Zeiner-Henriksen T. Rokevaner i den norske bcfolkning [Smoking habits in the Norwegian popnlationl. Tidsskr Nor L~egeforen 1976;96:61"/--20. 27. Axdson O. Aspects on confounding in occupational hcahh epidemiology [Letter to the editorl. Scand .i Work Environ Heahh 1978;4:98--102. 28. Peto J, Doll R, Hermon C, Binns W, Clayton R. Goffe T. Relationship of mortality to measures of euvironmeu~al asbeslos pollution in an asbestos textile factory. Ann Occup Hyg 198.5;29:305--55. 29. Cancer Regislr.v of Norway. Geographical varialion,, in tauter incidence in Norway 1966--75. O,;lo: Cancer Regi~tr.v of" Norway. 1978. Received for ptlblicatiott: 7 Fel',ruary 394 Scand J Shif by Tot Few in tionsh ~ollrse night ~ quent: was r aborti no[ C( slight" ing irt day \~ irreg~ tanec, inas night Fren, foun birth chan low I large weig furtl birtt that risk Swe, lar v birtl ~ In ol, Rep olo~ ToF CAT-1016800 o29