91

fraction might be nearly the same for smokers and non-smokers, and for malesandfemales.Tosocalled ‘dosimetricapproach’ yieldsariskvalue at the lower end of the given range.

Beir IV Committee estimates of lung cancer mortality associated with exposure to radon progeny. EllettWH.FabrikantJI,CooperRD.Radiat. Prot. Dosim. 1988;21:445- 9.

The National Research Council’s BEIR IV Committee has developed a new model for estimating risks due to the inhalation of radon daughters. This model takes into account the temporal pattern of excess relative risk that has been observed in studies of underground miners. Parameters in the Committee’s Time Since Exposure (TSE) model are cumulative exposure, age at risk, and time since a prior cumulative exposure. Smoking is modelled as a multiplicative interaction with cumulativeexposure, but the dataanalysed by the Committee werealso consistent with a submultiplicative interaction. Risk estimates based on the TSE model are compared with those put forward by the ICRP and others.

Radon and lung cancer: An epidemiological study in Norway. Suanden E, Magnus K, James AC, Green BMR, Strand T. NafionaI Insfitwe ofRadiarion Hygiene, N-1345 Osteras. Radial Prot Dosim 1988;24:471-4.

The objectives and strategy of an epidemiological study on the effects of exposure to radon in Norwegian dwellings is presented. The study is a cooperation between the National Institute of Radiaton Hygiene and the Norwegian Cancer Registry in Norway and the National Radiologi- cal Protection Board of the United Kingdom, with funding by the Norwegian Cancer Society. Measuremenls of radon are being made in 10,000 dwellings representing all Norwegian municipalities. The houses have been selected by a stratified random sampling procedure based on data from the Central Bureau of Statistics of Norway. The number of measurements in each municipality is proponional to the number of inhabitants. The total population of Norway is about 4 million. Thus one measurement will be performed per 400 inhabitants and one in 150 homes will be measured. The potential for detecting an effect of radon exposure by such a study in Norway is unique because: (1) Radon concentrations are high and there are large regional vari- ations. (2) Data from the Norwegian Cancer Registry is of high quality: allcancers havebeen subjecttocompulsoryreportingsince 1955.These data can be broken down according to municipality, sex and age. (3) In 1964/1965 a large scale survey of smoking habits was carried out in Norway. Thesedatacan also bebroken down according to municipality, sex and age, and by types of smoking and smoking rate. It is intended to examine the correlation between lung cancer incidence and geo- graphical variation in radon levels after making allowance for smoking habits. Radon measurements were started in early 1987 and the results of the study are expected to be published in 1989.

Lung cancer and indoor radon in Florida. StockwellHG,NossC1,RossEA,PetersJT,CandeloraEC.Deparmtenr ofEpidemiology andBiostatistics, College ofPublic Health, University of Florida, Tampa, FL 33612. Radiat Prot Dosim 1988:24:475-7.

An epidemiologic case control study was conducted in Florida to determine if living in counties with potentially high radon levels increased the risk of lung cancer. Fifty-three Florida counties were identified for study. Eighteen of these counties were determined by a statewide radon mapping study to have potentially elevated levels of indoor radon. Thirty-five counties without elevated indoor radon levels were chosen as the comparison population. A 25% increase in lung cancer risk was observed among residents of the three county area with the highest potential for elevated levels of indoor radon. The risks appeared highest in men and were increased for all lung cancer cell types combined and for squamous cell and small cell carcinomas of the lung. Among the remaining 15 counties with elevated radon levels, no elevation in the risk of lung cancer was observed.

The distribution of domestic radon concentrations and lung cancer mortality in England and Wales. Haynes RM. School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ. Radial Prot Dosim 1988;25:93-6.

Using aggregate data for the counties of England and Wales, a negative association is found between mean radon concentrations in dwelling and lung cancer standard&d mortality ratios, when regional smoking variations, diet variations, social class variations and popula- tion density are controlled. Cornwall and Devon have the highest mean domesticradongasconcentrations,yetthenumberoflungcancerdeaths there was within the range to be expected from relationships not involving radon observed in the rest of the country. While high values of radon exposure appear to concentrate in particular localities, the variations in lung cancer mortality between districts in Cornwall and Devon are small. These findings do not refute the linear exposure-risk hypothesis, but the evidence suggests that relatively few, if any, radon related deaths were associated with the dwellings where radon gas concentrations exceeded the recommended action level.

The incidence of malignant mesothelioma in Australia, 1947-1980. Musk AW, Dolin PJ, Armstrong BK, Ford JM, De Clerk NH, Hobbs MST. Deparrment of Respirarory Medicine, Sir Charles Gairdner Hospiral, Nedlands, WA 6009. Med .I Aust 1989;150:242-6.

Details of patients with malignant mesothelioma that was diagnosed in Australiabefore 1981 wereobtainedbysearchingallpossiblesources throughout Australia as far into the past as possible and up to and including 1980. The earliest patient with mesothelioma who was identified was diagnosed in Victoria in 1947. By 1980,535 (81%) men and 123 (19%) women had been diagnosed with the disease; only 14 persons were aged less than 35 years at the time of diagnosis (the youngestpersonwaslSyearsofage).Theincidencerateinsubjects were 35 years or older at diagnosis was less than 1.0 cases per million parson-years until 1964-1968, and then it rose progressively to 15.5 cases per million person-years in 1979-1980. The highest rate (69.7 cases per million person-years) was observed in 65- to 74-year-old men in 1979-1980. The incidence rate in Western Australia was greater than were the rates in other states of Australia after the mid 1960s. Pleural mesotheliomas accounted for 88% of cases in which the site of the turnour was known: peritoneal mesotheliomas accounted for 10% of such cases and ‘other’ sites for 2% of such cases. In 6% of cases the site was not specified. The exposure to asbestos was stated as ‘definite’ in 59%ofthecaseswitharecordedhistotyofexposure:8%ofallthecase.s in the study had been exposed to crocidolite (blue asbestos) from WittenoomGorgein WestemAustralia.Theageatdiagnosisofpatients with known exposure to asbestos was similar to that in those without known exposure. The increases in the incidence of malignant mesothe- lioma in Austria follow the published trends in the production and use of the amphibole varieties of asbestos in this country after a lag period of between 20 and 30 years.

Riik of lung, larynx, pharynx and buccal cavity cancers among carbon electrode manufacturing workers. Moulin JJ, Wild P, Mar JM et al. Service d’Epi&miologie. fnsrilut NalionaldeRechercheefdeSecwi~e,54501 VandoeuvreCedex.Scand. J Work Environ Health 1989:15:30-7.

Among workers employed in factories producing carbon graphite products the risk of cancer due to exposure to polycyclic aromatic hydrocarbons was estimated. In one cohort (plant A), a cancer incidence study was carried out: the number of cases were not significantly increased for lung cancers 17 cases, standardized incidence ratio (SIR) 791 orforcancersoftheupperrespiratory andalimentary tract(lOcaseS, SIR 103). In another cohort (plant B), a mortality study was carried out: neither the mortality from lung cancer [ 13 deaths, standardized mortal- ityratio(SMR) 1181 nor that from upperrespiratoryandalimentary tract cancers (10 deaths, SMR 125) was significantly higher than expected. Withineachcohort,acase-referent study wascarriedout. Inplant A the odds ratios were high but nonsignificant for lung cancers (odds ratio 3.42) and upper respiratory and alimentary tract cancers (odds ratio 2.19) and they showed a nonsignificant relationship with duration of exposure. In plant B, the odds ratios were low for every cancer site.