Tobacco Documents Online

ORIGINAL ARTICLE Evaluation of Certain Risk Factors for Lung Cancer in Cracow (Poland) A Case-Control Study Janusz Pawl~ga, Jadw[ga Rachtan and Tadeusz Dyba Unit of Epidemiology, Centre of Onco[ogy, Cracow, Poland. Fax: 048-12-213S91 Correspondence to: Dr Janusz Pawl~ga, Clinic of Oncology, Collegium Medicum, Jagiel[onian University, Kopernika 19, 31 531 Krakdw, Poland Acta Oncolo~Jca Vol 36, No. 5, pp. 4/I-476, I997 The relationship of lung cancer risk to dgamtte smoking, occupational exposure, air pollution at permanent residence and usual frequency of consumption of fruits and vegetablc~ was analysed in a case-control study in Cracow, Poland. The cases were 176 male lung cancer patients diagnosed in 1992-1994 with histological confirmation of the diagnosis, obtained from the population-based Cracow Cancer Registry, and 341 controls randomly selected from the general population. Classification of exposure to outside air pollution was based on measured level of total suspended particular matter and sulphur diox/de by particular districts of town (from 1973 to 1980). Except calculation of odds ratios for male lung cancer associated with exposure= among all study subject=, the assoc/at/on between lung cancer r~sk and frequency of consumption of fruit and vegetables was examined among drinkers of vodka above average. Risk of lung cancer wa= increased significantly with increasing number of pack-yeax~ of smoldng (OR = 18.1' for more than 40 pack-year~). The significant inverse assoc/ation of frequent usual consumption of bolIed vegetables was present both among all subject~ and araong vodka drinker= (OR -- 4.6 and 12.5, re,vpectively, for a rare consumption). The risk of male lung cancer ~va= negatlvely associated with the levd of air pollution but positively with the percentage of occupationally exposed. Our study prov/de~ no ev/dence of a sign/ficant hernfful effect of air pollution and found a strong inverse assoc/afion between frequent fruit and vegetables consumption and lung cancer risk. Received 13 June 1996 Accepted 9.4pril 1997 Since the early fifties studies have shown that tobacco smoking is the most important cause of lung cancer (I, 2). According to an estimation from 1990 elimination of tobacco smoking worldwide could potentially reduce lung cancer cases in males by 80-90% and by 60-80% in females. Elimination of certain occupational exposures in industrialized countries could reduce lung cancer cases by about 10%. However, the effect of reduction of air pollu- tion on the lung cancer risk is still uncertain (3). Previous, mainly observational, epidemiologlc studies showed almost consistent and substantial evidence of a strong inverse relation between the consumption of carotene-rich food and the incidenca of lung cancer (4, Y). Still, the results of some later trials raised the possibilky that beta-carotene may even have a harmful effect increasing the risk of lung cancer (6-9). A significant association b~twccn lung cancer mortality risk and air pollution was previously reported from Cracow (Poland) by J~drychowski et al. (10). This was, however, a retrospective case-control survey of lung cancer deaths with information collected from next-of-king and without data about nutrition. Scandinavian University Press 1997. I$SN 0284-186X Because of the still existing.controversy concerning the role the air pollution, and recently suggested ha~ul effects of beta carotene, in terms ofinng cancer risk, we report here the results of our study from Cracow with specia~ reference to the effect of air pollution, vodka and diet. NIATERIAL AND METHODS From January I, 1992 to December 31, 1994, 250 male lung cancer patients with histological confirmation of diagnosis were reported by the Cracow Cancer Registry (I I). To each case two controls matohcd by age (within 5-years) were randomly selected from the g~neral population by using the electoral roll 1992-].994 and systematic sampling. The standarized mailed self-administered questionnaire was used to obtain information on cases' and controls' persona[ characteristics (age, permanent tea/deuce with special emphasis on the district of town, marital status and education), smoking habits, consumpt/ou of vodka, occu- pational exposures and usual weekly frequency of con- sumpdon of carrots, fsesh and boiled vegetables and fruit. .4eta Oncologica This article is for individual use only and may not be further reproduced or stored electronically without wdtten permission from the copy~ght hNder. Unauthorized repreduction may result in financial and other penalties, (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY

4"/2 J. Pawlega et Men were asked whether they had cv¢r b~n regular smokers and, if so, about the age at which they had started smoking, the total number of years of smoking, whether they currently smoked or which year they had last smokext, and the average amount of cigarettes smoked per year. The smoking variable was catcgorizcd by pack-years (numbers of cigarettes smok*d per day x years of smoking divided by 20). Questious about vodka drinking gave information on the usual intake in grams, frequency of drinking per month and years of drinking. As concerns occupation, subjects we.m asked about the years of exposure to asbestos, certain particIes (cement, coal, metal, stone., leather, cotton, glass wool), gasea (car- bon oxide, sulphur dioxide, natural gas), ftml (gasoline, oil, diesel), pesticides, coal-tar products, paints and other com- pounds identifi¢d by respond*rs. Occupational variables were grouped by y~ars of exposure (non-e~xposcd, exposed leas than 20 years, and 20 years or more). The frequency of usual consumption of vegetables and fruit was quantified on a 6-grad, scale: 1) daily; 2) 4-6 times per week; 3) 3 times per week; 4) 2 times per '5") onco a week; 6) leas frequently. The classification of ~xposum to community air pollu- tion was based on mcasur~ mean yearly levels of total suspended particlea (TSP), and sulphur dioxide (SO2). The ambient air characteristics for the districts of Cracow City for the period 1973-1980 were determined by a network of 20 sampling stations distributed uniformly in town, d*- signed to measure TSP ad SO2 ew.ry day (12). The place of I~rmanent residence, in particular district of town, was linked with thes~ data. The avcxage delay b~twcen diagno- sis of cancer and interview was 3 months, and notification at cancer registry and int¢rviow 1 month. Th~ odds ratios (OR) with their 95% confidence inmr- vals were calculated for smoking, occupational ~xposure and consumption of fruit and carrots (mor~ than once per week, one.* per week, less than onc~ p~: week), green and boiled vegetables (more than 3 times per week, 3 times pex week, leas than 3 times per On calculation on an average usual frequency of drink- ing of vodka was one* per month and the average amount was 200 g, the odds ratios of lung cancer for usuaI freqtiency of consumption of fruit, carrots, grin and boiled vegetables among drinkers of vodka above average (mor~ than once per month and more than 200 g pex drink) were computed. The odds ratios of lung cancer wcrc calculated sepa- rately for pollution by TSP and SO2 respectively aft¢r ranking the districts of town from the loweat to the highest lev*l. The percentage of those occupationally exposed cases was estimated from controls. In addition to calculations of the ag~-adjusted odds ratios, unconditional multiple logistic regression was used to account simultaneously for the potentially confounding Aeta Oncologlea 36 (199"0 effect of s~verai factors. The significance of the Rtmar trend was also tested (13). All calculations wer~ made by using the statistical package Statistica Stasoft (14). The propor- tion of lung cancer cases attributable to various risk factors were estimated according to Bruzzi et al. (t5). The q~estioanaire was returned by 176 ma_le lung canc, r patients (respons¢ rate 70.4%). The response rate for con- trois was 68.2% (n =~ 34.1) (Table I). The controls were on average slightly older than the cases, with a~rnean age of 62 years (SD 7) and 59 years (SD 9), respectively. There were only small differences between cases and controls concern- ing marital status. The cases wer~ less educated than controls. Ninety-eight percent of lung cancer cases were cigarette smokers. The corresponding figure for controls was 73% (Table 2). There was a significant increace in lung cancer risk with increasing number of pack-years of smoking. We did not find any significant impact of length of occupational exposure on the risk of lung cancer. Lung cancer risk increasad significantly with failing fre- quency of consumption of fruit and wg~tables. However, in case of grcen vegetables the risk increase became non- significant after adjustment for the available confounding factors. When only drinkers of vodka above average were taker into account (Tabl~ 3) multivariate analysis showed an lurers, association betw~n consumption of groan and boiled wgctablea and lung cancer. The odds ratio of lung cancer significantly decreased with increasing level of air pollution measured by TSP and $02" in particular districts of town (Tables 4 and 5). However, it was significantly lower in districts with signifi- cantly lower percentage of residents being occupationally exposed (Table 6). The attributable risk of male lung cancer in Cracow was 90% for smoking, 26% for occupational exposure, and 60%, 25% and 49% for rare consumption of carrots, fruit Table 1 Age d!xtributlon of I76 cases of lung cancer and 34l controla (year) Cas~* Control~ (n = 170 (n- ~l) n % n % 30-44 10 6 2 1 45-49 14 8 lff 3 50-54 25 14 26 7 55-59 39 22 74 22 60-64 41 23 I 1 33 65-69 ' 30 17 68 20 70 + 17 10 48" 14 Mean + SD '~9+9 62+7 This article is for individual use only and may not be further reproduced or stored electronically without written permission from the copyright holder. Unauthorized reproduction may result in financial and other panaities. (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY

•Scta Oncolo&tca 36 (1997) L~mg cancer r~k m Cracow Table 2 Odds rattox (OR)for male hmg cancer according to ~molcing, occupational e,'~posure and consumption of fruit and ~egetablez 473 (n = 175) (n - 341) (95% CI)** Adjusted OR* (95% Smoking (pack-years) 0 1 - 20 I 21-40 80 More than 40 81 Z= trend Occupational exposure None 83 <20 years 42 ~20 years 51 %= trend Consumption > once/week 85 once/week 20 <once/week 71 ~ trend Carrots >once/week once/week 65 <once/week Z~ trend Green vegetables > 3~times/week I 3 times/weak 53 <3 times/week l I0 ~ trend Boiled vegetables >3 times/week 41 3 times/week 53 <3 times/week 82 Z~ trend * Adjusted for multiple risk factors: 92 1.00 1.00 69 3.0 (0.0-9.9) 2.9 (0.8-10.4-) 94. 16.6 (5.8-4.7.7) 15.2 (4.8-4.7.5) 86 23.2 (8.l-66.4) 18.7 (6.0-58.2) 76.4 (p < 0.05) 48.4 (p < 0.o~ 180 1.00 1.00 77 1.1 (0.7-1.8) 0.7 (0.4-1.4) 84 1.4 (0.9-2.1) 1.0 (0.5-t.8) 2.1 (p>0.05) 0.I (p>0.05) 270 1.00 1.00 25 2.6 (1.4-5.0) I.I (0.5-2.4.) 45 5,I (3.2-8.1) 2.4 (1.3-4.4) 54.6 (p < 0.05) 7.8 (p < 0.05) 131 1.00 1.00 97 4.7 (2.6-8.5) 3.5 (1,7-7.3) 112 6.2 (3.5-11.0) 2`9 (1.4-6.1) 40.9 (p < 0.05") 5.5 (,p < 0.05) 87 1.00 1.00 135 2.5 (1.3-4.9) 1.7 (0.8-4.0) llS 6.6 (~.4-12`5) 1.1 (0.5-2.7) 48.4 (p <0.05) 0.1 (p <0.05) 1~0 1.00 /.do 97 2.4 (1.5-3.9) 1.8 (1.0-3.2) 53 7.3 (4.5-12.1) 4.6 (2.3-9.2) 68.4 (p < 0.05) 18.5 (p < 0.05) - smoking:, for age, education, place of permanent residence, years of occupational exposure, frequency of R'uit and vegetable consumption - oceupational exposure: for age, education, place of permanent residence, pack-years of smoking, frequency of fruit and vegetable consumption - fruit and vegetable consumption: for age, education, place of permanent residence, p.ack-years of smoking, years of o~cupational exposure ** CI denot~ confidence interval and boiled vegetables, respectively. This figur~ could not be calculated for air pollution as the ~timated effect was inverse. DISCUSSION Our study prov£des no evidence of significant l~armftfi effects of ambient air pollution on male lung cancer risk. On the contrary, we observed that in the most polluted district of town the lung cancer risk was significantly lower than that among men living in the least polluted part. The variation in the risk of lung cancer was more consistent with the percentage of occupational exposure than with the lov¢l of air pollution. These results do not support the findings of a previous case-control study from Cracow (10). In 1990, J~dry- chowski et ai. (10) found a significant increase in lung cancer risk for the highest air pollution exposure level. males (OR of 1.48 with 95% CI 1.08-2.01). Our stndy was based on incident cases and their personal interview with ~e use of mailed self-admlnistered questionnaires, whereas the Jedrychowsld study was a case-cuntro! survey of lung cancer deaths with collection of information from next-of-kins. Therefore, the quality of the information, This a~cle is for individual use only and may not be further reproduced or stored electronically without written permission from the copyright holder. Unauthorized reproduction may result in financial and other penalties. (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY

474 Pawlega et el. Acta Oncalogica 36 (1997) Table 3 The ¢~ect of frult and t~getable consumption on rink of male lung cancer among drinkera of vodka above average > once per mouth Age-adjusted OR Multiple factor > 200 g/per drink (95% CI)** adjusted OR* (95% C1)** (n-36) (20% of total) (21% of tote1) Fruit >once/week 20 58 once/week 6 7 <once/week I0 8 x~ ~nd > once/week I 35 onee/week 16 20 < once/week 19 18 Green vegetable~ • > 3 times/week 1 3 time~/week 5 29 < 3 time~/week 30 23 Z= trend Boiled vegetable~ >3 time~/week 7 38 3 tirae~/week 8 20 <3 times/week 21 15 x= ~d * Adjusted for age~ edu~on, pl~e of permanent ~idcuce, ** CI deno~ ¢on~denee integral 1.00 1.00 4.4 (I.1-18.0) 0.5 (0.5-2.4) 4.5 (1.4-13.7) 0.6 (0.1-4.1) 8.4 (p < 0.05) 0.4 (p < 0.05) 1.00 1.00 32.5 (3,8-277.2) 13.7 (0.6-302.4) 41.9 (5.0-354.6) [7.9 (0.9-369.2) 21.8 (p <0.05) 3.5 (p >0.05) 1.00 t.00 4.4 (0.4-46.5) L4 (0.1-29.6) 53.5 (5.2-551.9) 6.4 (0.8-352.6) 3.4 (p > 0.05) 5.8 (p < 0.05) LO0 1.00 2.2 (0.7-7.0 0.5 (0~I-4.1) 12.2 f3.'/-39.5) 12,5 (1,2-t32.1) 21.2 (p < 0.05) 4.3 (p <~0.05) yea~ of occupational exposure and pack-years of smoking Table 4 Odds rat~ of hmg cancer and air pollution by total suspended particulates (TSP) Distdct Cases Controls of town n (%) a (%) Odth ratios (95% CI) adjusted for:. Age Age and smoking Age, smoking and occupational cxposu~ Age, smoking, fruit and vegetable consumption and occupational exposure Age, .~moMng, fruit and vegetable cousumption, occupational expoxere and education Nowa Huta 61 05) 69 (20) L00 1.00 1.00 (0.078)* KrowodtT.~ 35 (20) 87 (26) 0.42 0.47 0.47 (0.094) (0.25-0.71) (0.26-0.86) (0,25-0.86) Podg6rze 45 (26) 68 (20) 0.65 0.64 0.63 " (0.I04) (0.3S-I.09) (0.35-1.1~) (0.35-1.15) ~r6dm~e-e~e 31 (17) 117 (34) 0.28 0.29 0.28 (0.142) , (0.17-0.48) (0.16-0.52) (0.15-0.51) Unknown 4- (2) 0 za trend 15.8 13.3 13.0 p < 0.05 p < 0.05 p < 0.05 * Yearly mean of TSP in 1973-1980(mg/m3) 0.38 (o.19-o.76) 0AS (0.24-0.95) 0.24 (0,[2-0,48) 13.6 p <0.05 0.42 (0.2~ =8.S4) 0.48 (0.24-0.97) 0.24. (0.12-0.483 12.9 " p < 0.05 This article is for individual use only and may not be further reproduced or stored electronically without wdtten permission from the copyright holder. Unauthorized reproduction may result in financial end other penalties. (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY

Acta Oncologlca 36 (1997) L,a~g cancer risk In Cracow Table 5 Odds ratte$ of hmg cfmcer and air pollution by total sulphur dioxtd~ (~0~) District Caste Controls of town n (%) n (%) Odds ratios (95% CI) adjusted for. Age Age and smoking Nowa Hum 6I (35) 69 (20) 1.00 1.00 (0.071)* Podg6rzv 45 (26) 68 (20) 0.65 0.64 (0.076) (0.38-1.09) (0.36-1.15) Krowodl2a 35 (~) 87 (~) 0.42 0.48 (0.092) (0.25-0.71) (0.26-0.8~ ~r6dmie~ie 31 (1~ 117 ~34) 0.~ 0.28 (0.H0) (0A7-0.48) (0A6-0.50 Z~ t~nd 24.3 t7.3 < 0.05 < 0.05 P Age, smoking Age, smoking, fruit Age, smoldag, fruit and occupational and vegetable and vegetable exposur~ consumption and consumption, occupational occupational exposure exposure and education * Yearly mean of ~02 in 1973-1980 (mg/m3) 1.00 1.00 1.00 0.63 .0.48 0.48 (0.35-I.15) (0.24-0.95) (0.24-96) 0.47 0.38 0.42 (0.25-0.86) (0.19-0.76) (0.2t-8.84) 0.28 0.24 0.24 (0.15-0.51) (0.12-0.48) (0.12-0.48) 17.2 16.3 14.5 p < 0.05 p < 0.05 p < 0.05 especially regarding occupational exposure, could be worse than ~hat in our study. Nu~tional factors were not /n- eluded in ~he previous Cracow study. The classification of exposure to comrnun/ty a/r pollu- tion and the source of information were identical in both studies. Therefore, the bias connected with possible move- ment of male population from on~ district of town to another should not be the cause of different results in these two studies concerning air pollution. Furthemore, it is worth noticing that the population of Cracow is relatively stable with yearly migration rates from one district to another of about 1.5%. The passive smoldng and radon exposure were not evaluated in either studies. Thus, the differences between two Cracow lung cancer studies re- garding air pollution seem to arise mainly from metholog- ical problems.. Since the time of publishing of the previous Cracow lung cancer study (i.e., between 1990 and 1994) we have 475 not found any case-control studies in the literature sup- porting a significant role of outdoor air pollution as a risk factor of lung cancer. Chinese studies found aa important impact of indoor air pollution by smoky coal and inade- quate ventilation of houses, an exposure which ~ rather rare outside ~sia and Africa (16, 17). The case-control studies from the Niag~tra Region, Canada, five German cities and US-Cook County did not demonstrate any significant associations between outdoor air pollution and lung cancer risk (18-20). As in our study, the ltmg canc2rs we, re mainly attributable to smok- ing and more r~lated to ocoupational exposures thau to air pollution. However, probably due to the small numbers of exposed people, we were not able to demonstrate the effect of ocoupational exposures. In accordance with other epi- demiologieal studies we found a strong inverse effect of the intake of fruit and vegetables against lung cancer (5). l.u spite of rather consistent epidemiological evidence, the Table 6 Odds ratio (OR) of lung cancer according to place of reMdence and percentage of the occupationally exposed trend Plac~ of Cases Controls Odds ratios (95% residenc~ Oceupationaly exposed**, Age-adjusted MuldpIe factor* adjusted OR 1.0 71 0.4 (0.2-0.7) 41 10.0 Howe Huta 61 69 1,0 Other 111 272 0.4 (0.3-0.6) X~ trend 15.4 * Adjusted for age, education, smoking and fruit and vegetable consumption ** Estimated from controls Tills article is for individual use only and may not be further reproduced or stored electronically without written permission from the copyright holder. Unauthorized reproduction may result in financial and other penalties. (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY

476 J. Pawl¢&a et aL recent Finnish trial did not find any reduction in the incidence of lung cancer among male smokers after 5-8 years of dietary supplemention with alpha tocopberol or beta carotene. In the case of beta carotene a slight harmful effect was even suggestedl The authors try to explain the lack of reduction in the incidence of lung cancer among men given supplemention of inadequate duration, use of wrong dose, or inappropriate study population (6). Ac- cording to our observation the inverse (statistically not significant) association between lung cancer risk and fre- quent carrot consumption was present also among vodka drinkers. The significant reduction in odds ratios of male lung cancer with frequent consumption of boiled vegeta- bles was observed in Cracow for all study subjects taken together and among vodka drinkers above average. It has been suggested that some thermoresistant component in vegetables (not only beta carotene) can be responsible for this effect (21). In summary, we found that permanent residence in the most polluted part of town does not increase the risk of lung cancer. The frequent consumption of boiled vegeta- bles was inversely assodated with lung cancer risk even among drinkers of vodka. REFERENCES 1. Doll R, Hill/LB. Smoking and carcinoma of the lung. Prelin~ ina~j reports. Br Med .1 1950; 2: 739-48. 2. Wynder EI~ Graham EA. Tobacco ~moking as a possible etiolog!e factor in bronchogenio carcinoma. A study of six hundred and eighty-four proved case~. I Am ~lVled Assoc 1950; 143:329-36 ...... 3. Tomati~ L. editor. Conclusion. In: Cancer:. can~s, occurrence and control. Lyon: International Agency for Re~tch on Cuncer, 1990: 329-33. 4. Bye~ T, Perry G. Dietary carotens, vitamin C and vitamin E a~ protective antioxidant~ in human cancers. Am Rev Nutr 1992; 12: 139-59. 5. Van Poppol G. Carntanoid~ and cancer, an update with ampha$i$ on human intervention studies.,Eur J Cancex 1993; 29A: 1335-44. 6. The Alpha-Tocopherol, Beta Carotene Cancex Px'~venfion Aeta Oncologtea 36 (1997) Study Group. The effect of vitamin E and beta carotene 6u the incidence of lung cancer and other cancers in male smok- ers. N Eng J Med 1994: 330:. 1029-35. . 7. Henekens CH, Buring JE, Mauson JE, et al. Lack of effect of long-term snplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Eng J Med 1996; 334: 1145-9. 8. Omanan GS, Godmen OF., Thomequist MD. Effect of a combination of beta caroten and vitamin A on lung cancer and cardiovascular disease. N Eng J Med 1996; 334: 1150-5. 9. Law C, Read D. Betacarotene's fall from grace drows mixed reactions. 3 Natl Cancer inst 1996; 88: 235. 10. J~:lryehowski W, Becber H, Wahreadorf.l, Busa-Cierpialek Z. A case-control study on lung cancer with special reference to the effect of air pollution in Poland. J Epidemiol Comm Health 1990; 44: 114-20. II. Pawi~ga 3. Cracow City. In: Parkin DM, Muir CS, Whelan SL, GaG YT, Ferlay 3, Powell I. eds. Cancer incidence in five continents, Vol. VI. Lyon, France: International Agency for Research on Cancer, 1992: 678-$1. 12. Szal H, Krzy'zews~iZ. Spartial distribution and trends in air potintion in Cracow over 1973-1980 (in Polish). Ochr Pew 1985; 19: 9-12. 13. Atmitage P. Tests for linear trends in proportions and fre- quencies. Biometrics 1955; 11: 375-86. 14. Statistiea for Windows. Statsoft Inn 1994. 15. Bru~i P, Green sB, Byar DP, Briuton LA, Schalrer C. Estimating the attributable risk for multiple risk factors using case-control data.Am J Epidemiol 1985; 122: 904-14. 16. Lan Q, Chen W, Chen H, He XZ. Risk factors for lung cancer in non-smoker~ in Xuanwei Country of China. Biomed • Environ Sci 1993; 6~ 112:8. 17. Lin Q, Sasco AJ', Riboli E, Hu MX. Indoor air p611ution and lung cancer in Guangzhou, People's Republic of China- Azn J Epideminl I993; 137: 145-54. Holowaty EJ, Risch HA, Miller AB, Butch ,[D. Lung cancer in women in the Niagara Region, Ontario[ A case-control study. Can J" Public Health 1991; 82: 304-~L Jockel KH, Ahrens W, Winhman HE, et aL Occupational and environmental hazards associated with lung cancer, int 3" Epidemiol 1992; 21: 202-11. Keller J'E, Howe HL. Risk factors for lung cancer among non smoking Illinois resi-dent~. Environ r{es 1993; ~0: I-11. Graham S. Diet and cancer: Epidemloingin aspect. In: Lilien- fold AM ed. lleviews in cancer epidemiology. New York, Amsterdam, Oxford: Elsevier, 1983; 2: 1-41. 18. 19. 20. 21. This article i~ for individual use only and may not be further reproduced or stored electronically without written permission from the copyright halder. Unauthorized reproduction may result in financial and other penalties. (c) SCANDINAVIAN UNIVERSITY PRESS NORWAY