Document 06BRrb93XmaoBdBGeEJ1n5NLk

oezui BIO-MEDICAL RESEARCH DOCUMENT DESCRIPTION FORM 53 68 69 Duplicate in all cards: --> 0U0(F3oU year as-1961- File number [Right justify [Numeric only] Author(s), as Last Name FS (No Punctuation) and coden for journal as JAMA preceeded by one blank space ceoe-ziL#/*. * X'-OC/Q 77 78 Sub-Index Code 60 61 62 11 12 13 Title of Report; end with space-hyphen-hyphen-space. Follow with Index Terms, separated from each other with comma-space. Avoid other punctuation; do not abbreviate. g2 Brief Sumaary 12 10 SUMMARY: 31 32 61 62 61 62 63 64 ri~'#/?7Cr J /7-f7<f tl KemaNord Ett foretag i KemaNobelO A member of ^ the KemaNobelOgroup Mr. T R Tofkelson The Dow Chemical Ccnpany Bennett Building 2030 Dow Center Midland, Michigan 48640 USA Plastics Division Yourtal. Owrt CSj/kb 1978-04-25 Dear Mr. Torkelson: Enclosed you will find die pilot study concerning die Mortality In Njurunda Cammlty diat you were interested in. X ign enclose a statement by the National Swedish Nature Conservancy Board, Envirannental Department, vhich is made in connection with a claim fcr a "health examination" of the people in the vicinity of the KemaNord plant outside Sundsvall, Sweden. Both reports have been translated by KemaNord. X hope this information will be of sane value to you, and if you have any further questions I will be happy to answer you. Sincerely, KanaNord Plastics Division R&S 112231 Ends. Copy to Mr A R Adams, Air Products Postal Addr*Aa KemaNord AB Plastics Division Fack, S-S5013 Sundsvaif 13 Tdiptom 060/539700 CaUM Kemanord Sundsvall Tolox 71004 KemaNord <*> Ell fort4g i KwrwNotol/A nwmbar of th K*m*NoL*( group UitJitU** Ivwed UV Plastics Division /tp lk^.rittai.1<^./txunwl Istmt tra'Onttslv lation 1978-04-17 The National Swedish Nature Conservancy Board Environmental Department Dr. Rune CederlOf Dnr 194/75. SNV nr 70-6615-75 Finn-- 1 (8) Hi ^ Htj 0000360 R&S 112232 Regarding a Possible, Epidaniologic Examination of the Population Living in the Vicinity of the KemaNord Plants in stoekvik. ( 1 1. Background and Problem In the decision of the concession committee to grant KenaNord the permit for production at the plants in Stockvlksverken -- the following conditions, among others, are given for production of polyvinyl chloride (PVC). For the final conditions as to the emission into air a probationary period was granted tin the end of 1978. During this period the ocropany shall supplanant the present study material with an in vestigation of the health of those living in the vicinity with respect to a possible influence of the emission of vinyl chloride. Therefore, KemaNord asked the Environmental Department of the -- National Swedish Nature Oonservancy Board to give their views as to the need for an epidemiologic investigation of the population living around the KanaNord plants in Stoekvik. If it is considered that such a need exists, KanaNord also wishes to receive suggestions hat to plan and to carry through such an epidemiologic investigation. The production of PVC In Stoekvik was started during 1945, and In 1973 it amounted to about 60 000 t. Vinyl chloride (VCM) was used in the production. Inanumber of investigations this material has been linked to the occurrence of anginsarcana of the liver of workers who have been exposed because of their work. VCM is also suspected of inducing other types of cancer. So far four . - certain cases of angiosarcoma have been reported among the appr. S_, 750 workers who have been engaged in the PVC production at Stockviksverken. The ccnpany has carried out a series of investigations of the employees and because of these, to a limited extent, found certain symptoms of disease which might be linked to the exposure to VCM. However, no investigation of the people living around the plants has taken place. The VCM values of the air have been measured recently and are to-day very low. In only 1.7% of the actual time of measuring, the values within a radius of 1 km exceeded the detection limits of the measuring methods, 10 ppb. Values of 50-90 pgb were only obtained twice (0.04% of the time). - This means that at present a maximal mean value does not exceed 10 ppb. However, without doubt at the beginning of the production period the contents were higher and probably much higher than to-day. -2 KemaNord O Ett 1 K*m*NoO*l/A rTwmbsr of th# KomaNobot group Plastics Division /tp TRANSLATION 1978-04-17 2 (8) At? There are no figures that shew without a doubt that VCM has had effects on the people living in the vicinity of the V04-enitting industries. An evaluation of the risks has been done by A.M. Kuzmac and R.E. MoGaughy at the U.S. Environmental Protection Agency. The risk of the people living in the neighbourhood has been estimated on the basis of experiences from both professional exposure of workers and from experiments with animals. In the USA 4.5 million people are supposed to live within a radius of 5 miles (about 9 km) from a number of PVC-plants. The average exposure within this radius has been estimated at 17 ppb vinyl ( chloride, a figure that will be difficult to verify. As to the ' effect of the carcinogen materials it is now generally believed that the effect is proportional to the dose, even at extnsnely lew levels of dose, i.e. that the so-called zero effect only exists when the dose Is 0. Using such a Hnaar- form of evaluation one arrived at the fact that between 4 1 and up to 10 cases of liver-angiosarcasa should occur annually in the population in vestigated as a consequence of the exposure to vinyl chloride, this corresponds to a maximum of 2 cases per one million people per year. It is further emphasized that the frequency of malignant tumours of other types than angiosarcoma of the liver should occur twice as often, and also that serious liver injuries among occupationally exposed persons should be 30 times as frequent as angiosarcoma. Cue is cautious in drawing conclusions and mentions a lumber of error possibilities that ^ffy^influence the evaluations. If the Information concerning the frequency of angiosarcoma is correct it means that at an average content of vinyl chloride of no irore than 10 ppb in the air and In an area with the nunber of people as in Njurunda no more than one case of angiosarcoma and no more than two cases of other malignant tumours per 100 year should be expected as a oonsequence of the exposure. This / maximal! estimate is based on the highest American figure of risks supposing the unlikely high exposure of 10 ppb laxoucd KemaNord. in spite of this it is evident that this low frequency cannot be proved within a reasonable time of observation. ( The number of liver injuries of other types of cancer should, w according to the American investigation, show a 30 times higher annual incidence. It does not seem appropriate to use a linear extrapolation model here, but in case such a model is used, the exposure at a very high estimate means a case of liver injury every three years, or without regard to latent periods and higher mortality among the cases of liver injury which have already occurred, a total prevalence of no more than 10 cases in the new living population. If the liver injury was specific for the exposure to vinyl chloride it should necessitate a spot check -- . of about 1 200 individuals in order to prove with 95% probability -3 R & s 112233 t03l./av-> I R&S 112234 KemaNord O Ett to#*i*g t K*mNobl/A mmbv of tha KemiNobM group Itaruvtl^lcAXut Dy Plastics Division /tp |>iIihimiIIW,<^Ow vin| K,inf TRANSLATION Dalum/Oatr 1978-04-17 3 (8) ftoVihrH^j trf that liver injury does exist in the population. If a still higher probability is desired, e.g. 99%, a spot check of almost 2 000 individuals is necessary. But if the liver injury is not specific an essentially larger nunber of spot checks trust be taken, the number depending on the "background frequency1* of liver injuries measured by the test in question. Above calculations are very approximative but in the first place they lead to the fact that observation of nortality from cancer, and then specifically angiosarcana of the liver, has to be extended over a long period, and in the second place that proof of a ':.v possible liver injury needs spot checks from as maiy as 5 000 10. 000 individuals from whan specimens have to be taken, i.e. an almost total examination. , In this connection we want to point out that this concerns a prospective Investigation to watch out for the risks that may arise as a consequence of the present exposure - such as has been maximally estimated. With regard to the fact that the exposure may have considerably exceeded the present values previously, a larger nrmber of deaths or liver injuries could be principally expected in future among people who have lived within the area during such periods with higher exposure. ___ In oonnecticn with discussion the possibilities to carry through direct examinations of the now living population has to be mentioned, either through inquiries in order to screen the suspected positive cases according to the inquiry for medical exanlnation, or a caiplete medical examination of a selection of the population. However, there does not seem to be enough specific subjective problems to ask for In an Inquiry to serve the purpose of being an effective means of screening. Whether a medical examination a of the population should be motivated or not also depends an 'w how specific the objective methods of investigation in question axe. This is a medical, clinical question and a decision will have to wait until the final results fran the occupatianal, _ medical examination are available. With regard to above the following planned examinationa aim only at a registration examination of the population. 2. Planned Investigation 2.1. The Position Taken as a Matter of Principle Epidemiologic studies of the population can be done in several ways, in this connection two of than may be of interest. 4 leoi.ia om R&S 112235 KemaNord O Eli forug 1 KtmtNobtf/A nwntw 1 KmimNoM group r^tkmultol>Hi U>iftMrw/l*tMta try Plastics Division /tp frlwew<tin^tn/Ooa*n*>l rum* OTiRliiAA/CNkSK LATION 1978-04-17 One type (A) means study during a certain period of the mortality within the geographical area which has been supposedly exposed. The mortality figures are related to the average population in the same area during the period. The relation between exposure and mortality is difficult to evaluate in such a material as the mortality also refers to persons who have recently moved into the area. Besides persons who have lived for a long time in the. exposed area may have moved away and died in an area that is not being observed. This type of investigation is, however, relatively easy to carry through based on the existing population statistics. Another type (B) is to define the target population as the number of people having lived within the exposed area during a certain time a number of years back, and then observe the mortality of these individuals, no matter if it has happened in the area or in another place (cohort studies). In the latter case the number of people is clearly defined, and the time for living in and outside the exposed area can be given for each Individual. However, the latter type of investigation is much more time-consuming, mostly because of the necessity of checking persons who have moved away frcxn the area. In both types of investigations it is usual to compare results from selected reference populations which have been examined in the same way. In a suitable way it is also possible to compare the mortality investigation of type A with age- and sex-standardized incidence figures for all the nation. However, it has to be pointed out that especially certain industrial areas may have a population which in other relevant aspects than age and sex nay differ from the average of the nation. E.g. it may be the case for a lumber of behaviour variables like smoking and consumption of alcohol. Therefore, i,t is important that together with comparisons of incidence figures for all the nation also to study the difference between the target population and a comparison population which has been selected from an area with broadly speaking the same population structure. The above concerns mainly mortality studies, but the same principles should be applied even when it concerns checking of sickness when registering cancer, deformities, etc. 2.2. Preliminary Mortality Investigation, Type A In a preliminary study (C-G. Elinder and G. Pershagen) the mortality during the years 1961 to 1974 in Njurunda caimunity has been ocnpared partly to the mortality in the nation as a -5 4 (8) ifBt.rt.na.ft R&S 112236 KemaNord O Ett (Orttag i Km*Nobi/A 0< ih* Plastics Division /tp group fWki TRANSLATION 1973-04-17 trulls' 'RwlWauA 5 (8) Hwg n l*wij f--j whole and partly to the mortality in a reference area during this period. As reference area the Sjdlevad camunity outside Omskdldsvik was chosen because after going through a great number of camunities and counties along the coast of Norrland this canrunity turned out to be the best alternative to the reference area. SjSlevad has a similar population distribution and pattern of enplqyment as the population of Njurunda and is also geographically similar: a rmst-ai district close to a larger industrial town. Information of mortality has been obtained through using the so-called register of cause of death (Central Bureau of Statistics) where every death since 1961 is available on computer. When it had been proved that 10-1 might cause cancer of the liver this diagnosis attracted the most attention, of course. L However, during this period only 7 such cases had occurred, which have to be considered as Ism, Indicating that the previous. high exposure at any case has not yet been reflected in Increased mortality from angiosarcoma within the population in general, cn the other hand the investigation showed a relatively increased incidence of cancer of the pancreas for men, where the 5MRA' within Njurunda amounted to 148, i.e. an increase of about 50%. Similar calculations frtac SjSlevad showed no noteworthy proportion. When technically, statistically evaluating this increase of cancer of the pancreas - which is significant - reservations have to be made to the fact that the result has energed in connection with an explorative procedure where a great number of diagnoses have been observed. At a level of significance of 5% 1/20 of all observations are expected to be significant at rarxian. There has been found no reason to put forward any specific hypothesis concerning only cancer of the pancreas. In every explorative procedure of this kind "significant" results shall be followed up and, if possible, be the object of tests of independant naterial. Thus it is not possible _ now to canoent cn the causes of the result. 1) Standard Mortality Ratio expresses the relation between the observed number of *<* and the expected number of cases based on a specific figure of incidence related to age and sex for all the nation. -6 1<DI.7(0U KemaNord O Ew t K*mN0b#i/A irambif of thm K*mjih*ob*l group Plastics Division /tp TRANSLATION ftaHirti/Ukita 1978-04-17 fctMimt/Wepitam 6 (S) When going through the cases it turned out that 5 of the 20 men who died from cancer of the pancreas were found among those who had been employed by the industry. Hus does not mean a reliable increase among the enployees. However, if these are not taken into consideration the SMR does not differ significantly frcm 100. 2.3. Investigation Plan for a Mortality Study, Type B As has been pointed out above the mortality study, type A, is not acnprehensive. A study, type B, means that the population o registered within Njurunda in 1960 has been defined as the target population. This can be done by making lists frcm the census of 1960, which can evsi give certain structural and aocioeconcmical data concerning the individuals. Whm canhining a register based on this and the so-called cause of death c ocnputer tapes (see 2.2.), then a list of all deaths in the registered population is obtained in principle. For control "the survivors" are ocnpared to the register of the total population where, except for emigrants, everybody should be found. The same procedure is carried through within a reference area. The statistical analyses were carried through as a omparison between both the regions of figures of sexand age-standardized incidence. A similar checking of the individuals listed in the personal registers can also be acne of the cancer register founded by the Social Welfare Board. Besides information about canoer mortality this register also contains data concerning cancer sietaess. The analysis mat even take into account whether those living within the Njurunda area during 1960 sane time or other before or after that time have been aployed by KemaNord AB. An investigation of this type is estimated to take one year. The total costs should be around Sw.Crs 100 000. 3. Steinary and Discussion In spite of the many investigations the knowledge of the health risks for the population because of the discharge of the vinyl chloride frcm a PTC-industry is still very limited. To judge frcm the American estimates the risk for angiosarcoma of the liver In a population the size of the one in Njurunda is very snail. If the average content of the air is at most 10 ppb vinyl chloride then no more than one case should occur every 100 years. The risk for malignant turours of other -7 R&S 112237 taii.7*o?. [ KemaNord <^ Ett 1 KmaNobl/A m*mbr of th K*m*Not*4 group Plastics Division /tp . Vr Ck<tviikit|fi4m/Dv(uiMi rwnw TRANSLATION 1978-04-17 Efuitv 'Awpmi* v>- 7 (8) K*g>.i types than angiosarocma of the liver should be approximately twice that number. These values have been given assuming a linww relation of dose response without threshold values. As to other liver injuries these are indicated to be 30 times as frequent as liver angiosarcoma, but it can be assured that there is a threshold value which is probably higher than the contents which appear in the air. Besides the liver in juries are not specific for the exposure to vinyl chloride, and there does not seem to be any point in showing possible liver injuries in the neighbouring population, even if an / almost total examination takes place. '*-** A' comparison of causes of death in Njurunda ccnmunity during the years 1961 to 1974 and in the reference area Sj&levad _ during the same period has not shown any noticeable results / except for an Increase of 50% of cancer of the pancreas within - Njurunda. However, available data do not indicate if there is a connection between the type of tumour and the exposure to viryl chloride. It turned out that 5 of the 20 cases had been enployed by the industry. This proportion does not give any safe reason for a causal connection, but together with the totally higher frequency within the area it is a point to be observed fruii an occupational-medical point of view as well as from one of public hygiene. In these circumstances it is now the question to what extent further examinations of the population in the vicinity* of tiie KemaNord plants will be able to clarify the evaluations of risks based on the American estimates. It has to be pointed out that the evaluations of risks are based on a supposed content of vinyl diinrtde in the air of about 10" ppb, which is to be as the maximum content to-day. The information l to be obtained from studies of the mortality during the years 1961 to 1974 must be related to the proportion of exposure *Ti the and possibly even to values before 1961. It seems difficult to make an adequate description of exposure prior to this time, and therefore, the epidemiologic investiga tions should hardly be of any value to evaluate the dose \' I response relation. On the other hand a coipletely negative result should speak strongly in favour of the fact that no risks can be expected in future as a consequence of the exposure to vinyl chloride. In support of above it can be summarized that further investi gations of mortality of the population living in the vicinity of the KemaNord plants in Stockvik as to angiosarcoma of the liver and other malignant tumours cannot he expected to give results valid for dose response calculations. The circumstances -8 I. I I l I I R&S 112238 1OI.II01S KemaNord O Ett fort*Q i K*m*Nob*iM mtmtwr of fh# K*m*Nob*4 group Plastics Division Ap Qtr> tiflmiiiiivi/IXiciwi>cf<< iitm TRANSLATION 1978-04-17 concerning cancer of the liver which have appeared in the pilot study that not even the higher doses which have occurred earlier give any effect on the population in general. Of course, we cannot predict what will happen for the future why continued observations can be justified. If other authorities under these circumstances think further investigations should be made, a cohort study, type B, should be the'one to be considered. Ihe matter has been doodad by assistant department manager. Dr. Thonas Lindvall with Dr, Rune CederlSf as reporter. Professor, Magnus Piscator and medical doctor Carl-Gustaf EUnnsr have also taken part in the decision. 8 (8) & w 6SZZU \ .71,03.1 wzzvv S*U KemaNord <?> fi fOTHiag i K*mwNoti*(/A of th* K*m*Noc*rt group Plastics Division /tp I ju&it><4h>i*nn.|>k-iWnw<l ***' TRANSLATION lULmulut*- 1978-04-20 f Pilot Study concerning the Mortality in Njurunda Camunity fcy Carl-Gustaf Blinder and Goran Pershagen Introduction In the decision of the concession ccnudttee to grant KamaNord the pennit for production at the plants in Stockviksverken the following conditions, among others, are given for production of polyvinyl chloride IPVC): For the final oonditicns as to the discharge into air a pro bationary period was granted till the end of 1978. [Xiring this period the ccnpany shall supplement the present study material with an investigation of the health of those living in the vicinity with respect to a possible influence of the emission of vinyl chloride. KamaNord AB asked the Environmental Department of the National Svedish Nature Conservancy Board, the Karolinska Institute, to investigate the conditions for an investigation of the kind to which the concession relates. The oomnittee is expected to suggest some kind of an epidemiologic investigation in prder to demonstrate possible health risks to the population in the vicinity because of the anission of vinyl chloride. American investigations show a connection between angiosarcctna of the liver and exposure to vinyl chloride (\M) in the industrial environment during a long time and high contents (Lloyd, 1975). It is further suspected that high contents of VCM can Induce other malignant tumours (Moore, 1975). So far four cases of angiosarccne of the liver have been observed among the appr. 750 workers who have been enployed in the vinyl chloride anchor the polyvinyl chloride production since production started 1945 at the plant in question (deaths occurred 1970, 1972, 1976 and 1977). This pilot study is Independant of the investigation regarding epidemiologic investigation around Stockvlksverken which has been made by the department, but it should give valuable background information. -2 1 (5) n*n n/S*g f*j M1U402; t I i KemaNord O Ett forttog i KmaNobtl/A nnmb of tt <tmNobtt group Plastics Division /tp l*,itun4iihiii>ai/(>Aunol mmi TRANSLATION i^iiwidiiiu 1978-04-20 Exposure of the Neighbourhood The production of polyvinyl chloride in Stockvlksverken started during 1945, and during 1973 it amounted to about 60 000 t. The VCM content in the industrial environment was very high during the 50s and the 60s but has been reduced considerably during the 70s, first to 10-20 ppm during 1974 and front January 1, 1976 to below 1 ppm which is the limit in force new. A mxrber. of measurements of imnission in the housing area close by was made in 1976, and the result showed that the imnission of vinyl chloride in question was low, mostly beyond detection by the measuring method, the limit being 0.01 ppm (0.03 mg/m ). However, the emission and the exposure of the environment was considerably higher earlier before vinyl chloride monomer was identified as a carcinogen material. Besides polyvinyl chloride even caloiun carbide, sodium chlorate, formalin, formaldehyde resins of urea, phenol and melamine, as well as special chemicals consisting of, among other things, fatty chemical products and agricultural chemicals are pro duced at the KanaNord plants in Stockvlksverken. The production of nest of the chemicals was started in the 40s and the 50s, but sane of the special chemicals were not produced until the last decade. However, it has to be pointed out that other industries have or have had manufacturing within the area in question. Thus Granges Aluminium, a producer of aluminum, is located 3-4 ion north of Stockvlksverken, and Svartviks Cellulosafabrik, which was closed down in 1973, 2-3 km south of Stockvlksverken. until the mid-60s there was a production of cellulose in Essvik 2-3 km southeast of Stockvlksverken and until the early 40s in Nyham about 2 km also southeast of Stockviksverken. Chemical products have been produced in Nyhann since the production of cellulose was closed down, and for the last years sulphur dioxide has been manufactured in Essvik. In table 1, which shows the population distribution within Njurunda cannunity in 1974, it can be seen that about 2 500 of the Njurunda population of 12 000 people live close to Stockvlksverken, i.e. within a distance of about 2 tan. Program for the Pilot Studies When studying possible health effects around a certain polluting industry it is necessary to define an exposed population. The exposed population must not be so numerous that possible health effects are "wiped out", but on the other hand not so small that It does not cover the whole area where health effects may occur. -3 t031tt02.ft 2 (5) KemaNord <!$ En twoug i KontoNobol/A nwntbot o( th* KamaNatHl group f-1 UH#d<4'ltP<i<i by Plastics Division /tp TRANSLATION Udoan/OaH 1978-04-20 Efutw /HhaaM S4WP** 3 (5) N4 Of practical and of work resource reasons the exposed population In this study is defined as Njurunda camunity in Sundsvall county. Information on mortality in this area has been compared to the mortality in the nation as a whole and to a control area. In this case the mettods are very rough as only a limited number of the population of Njurunda camunity lives close to Stockviksverken (see table 1). As reference area the SjMlevad camunity outside QmshBldsvik , was chosen. Having checked a large nutter of camunities and V counties along the coast of Norrland this camunity appeared to be the best alternative as a reference area. The population distribution and industrial profile of SjMlevad are similar _ to the ones of the population in Njurunda, and besides it is geographically alike: a coastal camunity close to a larger industrial town. Information about the mortality has been obtained through using the so-called cause of death register (SCB) where all deaths have been oarputerized since 1961. ___ The methods used in the study in question are the same as the ones used for a similar study around the RSnnskMrsverken, outside SkellefteA. This has recently been published (Pershagen, Blinder and Bolander, 1977}. \ Result and Discussion As a first step the mortality in Njurunda camunity fran the years 1961 to 1974 inclusive was oaipared to the average of the whole nation, as well as to tbe reference camunity of 1 , SjMlevad. The scrolled crude ratios (SRR) for a muter of different eaiwpa of death were calculated, see table 2. Besides the total number of TMTM of death table 2 shews different types of causes of death in Njurunda camunity between the years 1961 and 1974 inclusive, also the so-called crude rate ( . ratios (CRR). In this case the CRR means that the specified --' causes of mortality in the population in question during a period of 14 years has been oaipared to the mortality of the reference populations. In the case in question the mortality for various causes of death in Njurunda camunity has been oaipared partly to the camunity of SjMlevad, partly to Sweden as a whole. The table shows that the total mortality in Njurunda camunity gets a CRR of 1.6 which is higher than the expected 1.0, but this can be explained by the fact that the average age within the Njurunda camunity is higher than within the > - SjMlevad camunity and also coopared to the nation as a whole. -4 R&S 112242 t0JI.7l.0M 'i KemaNord O Ett i KamaNobal/A nnmb of th KtimNoM group ThwuiM/Oi by Plastics Division /tp Ciuti(iwntn|inti/Ui(uCH<,i iw TRANSLATION 1978-04-20 When the total mortality is divided into different groips no marked differences from the CPU of 1.6 appear, except cancer of the pancreas, where the CRR is 2.7 and 2.4 when compared to SjHlevad and the nation as a whole resp. It is to be noticed that we have only been able to find a total number of seven cases of cancer of the liver (4 men and 3 women) during the whole period that VO-I has been proved to cause only cancer of the liver, and consequently we had first expected to find an increase of that frcm passible effects cn the population living in the vicinity. w As suggested above the crude rate calculations can be mis leading because of the different age distribution within the investigated population and the reference population (as known the frequency of cancer increases steeply with increasing age) therefore, we continued with an age-adjusted comparison and with calculation of standardized mortality ratios (SMR) for cancer of the pancreas. Tables 3 and 4 show an age-adjusted coiparison of the nurrber of cases of cancer of the pancreas in Njurunda and SjSlevad to what should be expected based on the statistics for causes of death far all the nation. The size of the population and the age distribution in the ecmiunities investigated have been calculated as the mean value of the census of 1960 and of the ones of 1965 and 1970. The expected nurfaer of cases of cancer of the pancreas has been calculated as the mean value of those recorded in the whole nation in the years 1961, 1964,, 1971 and 1374 (Official Swedish statistics). 4^ Tables 3 and 4 show that an ovemortality of cancer of the pancreas exists among men in Njurunda oaomunity (SMR = 148). When the reference area, SjMlevad, was caipared to the national average the observed nurrber of cases of cancer of the pancreas o decreases instead by about 25% more than what was expected (SMR = 74). Even in Njurunda the SMR should be below 100 as the incidence of cancer of the pancreas is higher in big cities. When going through the register of KenaNord employees it turned out that five of the 20 men who had died frcm cancer of the pancreas had been employed by the ocnpany for ncre than two months. ' 4 (5) R&S 112243 i ir t iz w s 'sa am KemaNord O tr Ibrtiag I KwmNobti/A flwnbv of th# KmvNoUt group Plastics Division /tp C^uPWlhHMlti/l^AtfrwiI ruAY TRANSLATION IjHulTWUjt* 1978-04-20 tiU'W/fhvAM 1%** That five of the twenty had been employed by KemaNord does not mean a safe increase of the number of cancer of the pancreas anting the employees, as approximately 25% of the male population in question within Rjurunda caimunity some time or other during their lives have been eoployed by the company. Conclusion This pilot study is very rough and can with certainty neither / prove nor exclude possible effects on the vicinity as a con sequence of the emission from Stockviksverken. However, the study stows that only seven cases of cancer of the liver (4 men and 3 women) have occurred in Njurvmda caimunity In the years from 1961 to 1974, which is not higher than expected. However, the study also shows that the mortality from cancer of the pancreas in Rjurunda cominity is higher than what can be expected. Any conclusions as to the aetiology of this cannot be drawn at this stage, however, but as chemical in duction of cancer of the pancreas has been described (Finklea, 1976) such should be suspected. In order to make it possible to trace any possible causes of the increased mortality from cancer of the pancreas a so-called case reference study based on the cases in question is urgent. Besides occupational exposure at KemaNord factors such as employment by another industry, location of housing, alcohol consumption, smoking habits will have to be thoroughly in vestigated in such a study. { 6 R&S 112245 Referenscr Byrfin, D. , fiiretayslakarc; brev, 1977. I- v}- . Polk- och bostadsriikningen, I960, 1965, 1970. SCO, Stockholm. Pinklea, J.F.j in letter of August 20, 1976. National Institute for Occupational Safety and Health, Rockville, Maryland 20852, USA. '{ "r Lloyd, W.J. Angiosarcoma of the liver in vinyl chloride/poly vinyl chloride workers. J. Occup. Med. 17, 333-334, 1975. Moore, J.W. The vinyl. Chemistry, 48;12-16, 1976. # Pershagen, G., Elinder, c.-G. and Bolander, A.-M. Environ. Health Perspect. 19tl33-137, 1977. Statistiska Centralbyran, c/o Ann-Marie Bolander, Stockholm. Sveriges Officiella Statlstiks Dddsorsaker, 1961, 1964, 1968, 1971 och 1974. SCB, Stockholm. 1> KemaNord O lt fcwuUfl i KomNoto*l/A ownblr of th K*m*Nob*f group Plastics Division /tp Ouk(M<(C<>|nj<'Hl/D-L(^Vll iWTm translation Ckn.wn, Qai* 1978-04-20 E< kill*/Pt|HAA Mi TABLE 1 The Population of Njurunda 1974 Stockvik, Bredsand Svartvik Kvissleby, Nolfcy DingersjO, Klockarberget Njurundabcntren, MjOsund Skottsund Essvik Juniskdr Skatan Xngan, Ojestrand VSstana Total nunber 2,355 1,088 3,702 583 2,196 842 674 311 138 100 70 12,059 IflDl.ftOft #---------------------------- - --------------------------------------------------------------------- r i1 < ' r1 X r TABLE 2 The mortality from various causes of death within the Niurunda ccnmunitv 1961 to 1974, as well as Crude Rate Ratios (CRR) for the various ranges of death ^ f Cause of Death Total number Njurunda Deaths, total nurber Tumours, total number 1082 182 Tumours in the digestive duct Tumours in the respiratory organs 81 25 Tumours in the lyirphatic and haemopoietic tissue 5 Tumcurs in the pancreas Tumours in the liver Tumours in the gall ducts and the gall-bladder 20 4 0 CRR Niurunda Sjiilevad 1.51 1.21 1.10 1.62 0.23 2.71 1.40 - CRR Njurunda Sweden 1.61 1.35 1.55 1.22 0.36 2.36 1.70 - Total number Njurunda 829 182 65 9 16 7 3 9 CRR Nluruwla SjSlevad 1.47 1.30 1.27 1.81 1.73 1.09 1.84 1.81 CRR Niurunda Sweden 1.53 1.53 1.60 1.72 1.52 1.00 1.82 1.85 Nervous system and sense organs Heart and circulatory organs Respiratory organs Malformation 146 489 62 9 1.71 1.65 2.30 1.40 2.19 1.71 1.62 1.70 155 328 35 6 1.48 1.84 1.76 0.84 2.12 1.54 1.14 1.66 CRR = 1, means that there is no difference in mortality, while CRR = 2 means that the mortality in the investigated area is twice the cne in the reference area. toZZll S$U P la stics D ivision iff? 3f z iO Iel \ *0 c lt z* r > ?? % i % ?<> 1i i Iftoo it f 1 f,, s1 i< i .* KemaNord O Ett t K*fnaNob*i/A nwrobtr of the KtmaNoW group Plastics Division /tp IjuAufWIWWIft/DUu^lWM n#M TRANSLATION Lutun<ttew 1978-04-20 ErulW/f^iMl TABLE 3 The mortality fran cancer of the pancreas within Njurunda ccnmunity fran 1960 bo 1974 oaipared to the national average: NJURLKDA Average population I960, 1965, 1970 20-29 men women 30-39 men wcmen 40-49 men wcmen 50-59 men wcmen 60-69 men women 70- men women 813 711 744 698 835 796 861 849 704 678 506 511 Expected nixrber of cancers of the pancreas based on the whole country 0.00 0.01 0.06 0.10 0.43 0.38 2.01 1.02 4.06 2.43 6.97 5.00 Observed ambers of cases in Njurunda 2 1 7 3 11 3 . Men SMR 20 TT3? x 100 = 148 Horen SMR =* 7 x 100 78 R&S 112248 I R&S 112249 4 KemaNord O Ett forUQ KtwNobil/A munbw of th* KmiNobtl group r#nioi#lwV^Mi UdaOvt/lkiuMl by Plastics Division /tp [>|fctJlvi.|tM<'WWblAtlt>Wl il TRANSLATION lljt.WtWf1.lt 1978-04-20 ^ m/fUy Nil TABLE: 4 The mortality fran cancer of the pancreas within Sjaievad conuunity from 1960 to 1974 ccnpared to the national average: -- - SJSLEVAD H Average population 1060, 1065, 1970 Expected nutter of cancers of the pancreas based on the whole country Observed nurbers of cases in Sj&levad 20-29 men wanen 30-39 men wanen 40-49 men wanen 50-59 men women 60-69 men women 70- men wanen 1326 1216 1148 1182 1282 1233 1145 1098 772 749 395 490 0*01 0.02 0.05 0.09 0.67 0.49 2.67 1.32 4.46 2.82 5.69 5.04 1 4 3 2 2 7 ' Men SMR - 10 1235 = 74 Wanen SMR - 9 935 = 92 ). 10 f 02.A j i!