Document gaDqbqV9zMQXRvbQZ03bajvpG

l v 88 - to adhere to all these criteria mentioned. 2ielhuls et al discussed other methodological aspects of study design: it is essential that where pos sible the history taking should be performed by an experienced and inquisitive physician) it should not only be based upon an extensive standard li6t of pos sible sources of exposure, but it should also Include a free interview, explo ring unexpected unlisted possibilities of exposure. In addition the controls should be carefully matched with regard to age, sex, place of residence, time of diagnosis, and state of health (for each deceased case of mcsothelioma a deceased control, etc.). An absolute underreporting of post (para-)occupation al exposure may easily occur when for instance a too rigid standard question naire is followed or when only hosxit*! records are screened, a relative under reporting of past asbestos exposure occurs if deceased mesothelioma cases are matched against live controls. Main et al (1974) reported on a case-control study from the Hamburg area: 150 autopsy cases wore compared with the same number of live controls} 85 cases had occupational exposure against 35 of the controls; 70 cases had neighbour hood exposure, 1 case domestic exposure? no indication for exposure category J, 7, 3 had been found in 44, i.c. 79t of mesothelioma cases and in 114, i.e. 76% of controls. The maximal prevalence of true environmental exposure there fore w%s less than 30% however, relative .underreport i ng of (para-)occupational, neighbourhood and possibly domestic exposure had very probably taken place, be cause live controls were compared with deceased mesothelioma cases. In the Dresden and Hal 1e-Kagdunburg (DDR) area StJrm (1975) performed a study in 315 mesotheliomas (788 pleura, 76 peritoneal, 1 pericardial); in 746 cases, in which an exhaustive evaluation could be performed (69 cases no history!), 67% had past direct occupational exposure, 70% indirect occupational exposure, 4% domiciliary exposure, 61 were neighbourhood cases. Studies in Italy (Rubino et al (1972) snd Switzerland (ROttrser et al 1974) used non-sensitive history taking procedures which very probably led to serious underreporting: the percentages of r.bsence of past occupational exposure, 90 and 96% respectively, arc certainly much too**'igh to be accepted as valid. In the Netherlands Ziclhuis et ol (1975)' d rot establish evidence of definite or probable occupot ic:*n) cxyusure in lot o: 31 cases from the Rotter dam area (herbsur, shipyards, heavy industry) against 711 in matched controls? the percentages were 39 and 83 respectively in 36 coses and matched controls in a a u J Hi- ( St o. t! t n> n T1 t* hr li t' g: o! 900] 37ki PRODUCED BY FORD