Document vBE0v9KmLgJb5N07q9Dgppajm

JULY, 1981 ISSUE #1 VOLUME #1 This first in a series of "advisories" is designed to keep water utilities and public health officials current on information regarding usage of asbestoscement pipe. Each Advisory issue will factually summarize data contained in scientific papers, government studies, media reports, and technical and educational source material. JNSimiTB OP OCCUPATIONAL AND GOVERNMENT ANIMAL FEEDING STUDIES SHOW NO CARCINOGENIC EFFECT OF INGESTED ASBESTOS Preliminary results of government-sponsored studies on the effect of ingested asbestos on laboratory animals show no carcinogenic or cocarcinogenic effect. The studies are being conducted by-the National Institute of Environmental Health Sciences (NIEHS) under the direction of the Public Health Service's National Toxicology {5 Program. The Environmental Protection Agency (EPA) has contributed a portion of gj ^;he funding. 3? 25 The aim of the NIEHS/EPA studies is to assess the biological (carcinogenic) effects jjfof asbestos fibers which are ingested, or taken into the body through the digestive tract,. | cas in drinking water. V(C02c . > 3 The studies call for asbestos to be fed continuously in the diet over the entire lifespan of the animal, which is defined as the age at which the animal begins eating solid food until its death. A total of 1,850 male and female hamsters were fed asbestos, and an equal number of hamsters were fed a control diet with no asbestos over their lifetime of 18-23 months. Two types of asbestos fibers (chrysotile and amosite) commonly used in the manufacture of asbestos-cement products were administered. The dose of asbestos (1% of total diet by weight) fed to the hamsters was millions of times greater than levels which occur in drinking water consumed by the public in a lifetime, in areas where asbestos occurs naturally in the drinking water. NIEHS reports, "There was no indication of major differences in the mortality rate between the hamsters receiving the asbestos diet or the control diet." Preliminary analysis of the hamster data indicates that no carcinogenic or cocarcinogenic effect was observed. 1600 Wilson Boulevard^ Suite 1008*& Arlington. Va 22209 A (703) 841-1556 o- / FMSI 05514 The analysis included thorough gross pathological evaluation at time of autopsy, as well as microscopic examination of some thirty tissues from test animals. In addition, preliminary results of rat studies in which a total of 5,158 rats were fed asbestos and control diets were also reported. The lifetime exposure phase of the study concluded, ". . . longevity was not affected by exposure to the various types of fibers." These findings are consistent with a number of other studies of animals fed asbestos in food or drinking water. In 1980, the Health Research Institute at Fairleigh-Dickinson University reported no malignant tumors related to treatment in hamsters maintained on drinking water containing 13 billion asbestos fibers per liter. References: 1. NIEHS, Biological Effects of Ingested Asbestos Status Report January 22, 1981, National Toxicology Program, Public Health Service. 2. NIEHS, Biological Effects of Ingested Asbestos Status Report February 15, 1980, National Toxicology Program, Public Health Service. 3. Moore, J.A., NIEHS Oral Asbestos Studies, National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD July 18-20, 1977. FMSI 05515 BRITISH MEDICAL JOURNAL, 1981 (August 15); 283: 457-458. IOEH 5520 Code 8 ing is massive, and it has the added advantage chat dottif agents may be injected.* If bleeding continues skilled judg is needed, and joint consultation is essential; a balance I be scruck between early haps unnecessary) intervention and repeated resii transfusion,.wl the patient's resy intensive treament they irily in an ordinpy ward. In .very Jones,7 showed that mortality, the duagerouK myth persists bleed from the gu/shouldjree given "nil by is also widespread^faith ityflkalis, despite the clinical observation that in a patien has bled previous dyspeptic pain (and therefore ^ptesi >ly add) disappears. The immediate and long-term rof bleeding on gastric acidity bear re-examination, jk simij uncritical attitude has grown up to the use * ' ' trials* have shown that it is valueless in the.treac as opposed to the pre vention, of bleeding, and its/`routine" use intravenously should be condemned cm the /founds of both illogicality and expense. For mosc patients admission to hospital should mean a f`ew days of~ o`bservap*bn than treatment, and ifbleeding does not recur they can 1 discharged home. 1 Cocoa PB, Raaanbat ; Waldnm RPL, Aaoa ATR. Baity eadoseopy of ooMphatus, 4a and rluodmil bulb in psdener vhfa haama 5r AhdJ l973at:J03-9. 1 Schiller KFR* Tn i SC* VUUans OC. Hacmaccmcw and mriuii, with specialyrefei l97O;ii:7-l0. A to bents infliwncim da outcome. Br Med j Conn HO. T> so/or not to scope. .V EnfiJMtd 1981 ;304:967-9. * Peterson VL* Mractt CC* Smith HJ AUen MH* Corbett D8. Routine early encuachpy in upptr^amoimfvinakraa bleeding: a randomised, controlledjpd. Y E**iJ Mtd 1981 84:925-9. Hum PS1! TJoniinj j, Kerman MG. MonaUty in patients with heemat- ena: a prospective study. Br Mtdj 19793:1238-40. 1 Ring ^It^eaga JA, Baum S. Inccnrcncaocsai radiology in gastnwntesdnal _ in: Berk JE, ed. Devtiopmtmu m digaavt rfiieew Phila- : Lea and Pabigcr, 1977:59-72. r Jofl&uFPA. Haamatamaai* and melaana with special reference to binding pcic ulcer. Br MtdJ 1947pi:4414, 477-82. r SJt Misiewkz JJ, Edwards J, it of. Controlled trial ofrjmrtidlne & upper gastrointestinal haemorrhage. Gmt 1979;20:892-3. Smoking, coal, asbestos, and the lungs Over the past 20 years the number of British coalminers with pneumoconiosis has fallen substantially. In part the fall is due to the fact chat fewer men work in the pits, but it is also a response to effective measures taken to reduce* dust levels underground--on the evidence of research showing the association between exposure to respirable dust and the risks of developing pneumoconiosis.1 Some 500 men a year are still diagnosed as having the disease by the pneumoconiosis medical boards, though the average ages at which men show signs of the different stages ofthe disease have been increasing steadily, largely reflecting the higher dust levels of earlier yean.* Asbestosis, the other important pneumoconiosis in Britain, continues to be diagnosed by the pneumoconiosis boards in about 200 people a year; no decline in incidence has yet been seen, but current dust-control policies in the industry are expected to produce such an effect in the near future. In theory occupational diseases arc wholly preventable, but in practice mi long as society requires an industry's products some men will fail ill and even die as a result. A cost in terms ot and accidents is exacted for the benefits provided by most productive industry. Society and its elected representa tives need to be informed of the risks in order that work can be as safe as possible, consistent with the need to continue or increase production3; but the ultimate responsibility in a democratic society for occupational morbidity and mortality rests with every individual. Two problems that exemplify the complexity of these issues are bronchitis in the coal industry and lung cancer in the asbestos industry. Chronic bronchitis is one of the most mmmon disabling diseases in Britain and, though mortality from it has been failing, it still ranks high as a cause of death. Its relation to cigarette smoking is well known,4 but it is also related strikingly to the Registrar General's social groupings (which are based on occupation), to dust exposure in industry, and to area of residence,4 implying some influence of general ionospheric pollution (ar least in the past), overcrowding, and other.social factors. Cigarette smoking, with its effects on bach smokers and chose who surround them, is also strongly related to social class.* In clinical practice patients disabled by chronic bronchitis who have never been smokers are extremely rare. Talk of occupational bronchitis, as though occupation were the sole cause of the potentially disabling or fatal disease in an individual, is Claims that exposure to dust does noc contribute to the disease ate, however, equally false, since there is good evidence in the coal industry of relations between measured dust exposure and symptoms, impairment of lung function, and mortality from chronic bronchitis.7 * To disentangle the relative effects of dust and cigarettes in causing the disease is difficult epidemioiogicaily* and im passible in the individual exposed to both; nevertheless, both have played their part--at least in the past. Whether current levels ofdust exposure in the coal industry will be sufficient to cause important clinical effects on the airways remains to be seen, but the evidence suggests that this is unlikely. Similar arguments apply to the relation between exposure to asbestos, cigarettes, and lung cancer. Here the two causal factors seem to act in a multiplicative manner. Exposure to asbestos probably has a linear relation with the risk of develop ing hmg cancer.1811 This risk is very High indeed in the heavily exposed smoker, but the risks are also substantial in a heavily exposed non-smoker. Two forms of action would, therefore, reduce the number ofworkers at risk oflung cancer; reduction of asbestos levels in the industry and reduction of smoking. Cutting the proportion of smokers would have more effect in workers exposed to asbestos than in the general population. Dust control has been much improved in both industries in recent years, and further improvements will be made in the near future voluntarily and in response to public pressure or legislation. As well as reducing the risks ofthe specific pneumo conioses in the workers, this will also reduce the risk of gl|r^n*c bronchitis in miners and lung cancer in asbestos woilu-ii. Yet these latter diseases will continue to be a problem in industry as well as in the general population so long as people continue to smoke. An important reduction in their mcirlmce. will occur only when the same public and govern mental pressure is exerted on smoking as on dust control. Bmnung tobacco advertising would be a sensible next step. R-- S, Walton WH, Rofan JM. New Sum standard, for , M Bnmh co*1 mine*. Nairn 1970;227:445-7. * Coal Board. Mtdiciru ttrviet jimtutl report 1979-80. London: , National Coal Board, 19l. * *!ni,5?0"v Th* riikj of a, .odn* Sr \UdJ 19H1I '81:1374. oyul College of Physicians. Smokt*ttf anJ health m./. Tunbridge Veils: Pitnun Medical, 1971. 10 32 36 82 FMSI 05516 458 BRITISH MEDICAL JOURNAL VOLUME 283 15 AUGUST 1^81 4 Holland W\f. Epidemiology of chronic bronchitis. In: Scaddins J^' Cummin, G, Thurlbeck TM, cds. ScientificfitmtUticns of mpi'olory medicine. London: Heinenann, 1981. ,, ,, * Capcll PJ. Trends in cigareite smoking in the United Kingdom. Health Trends 1978;10:49-54. . 1 Rae S, Talker DO, Aafield MD. Chronic bronchitis and dust exposure m British coalminers. In: Taltoa TH. ed. inhaled particles. 1- old Tolling, Surrey: Unwin, 1971. " Rogan JM, Airfield MD, Jacobsen M. Rae S, Talker DD, Talron TH. Role of dust in the working enrironmenx in development of chronic bronchitis in British coal minen. SrJ Ittd Med 1973 0:217-26. * Tacobscn M. Smoking and dtsahtitry ii wiiiw r --mano,.740. 11 Saracci R. Asbestos and lung jacu; sn ..!, of the epidemiological evidence on the asbcstot-emoldnc Intcractioii- far T Cancer 1977 0: 523-31. 11 Liddell D. Asbestos and public health. Thorax 1981;30:241-4. hypertension, for example, have been reported in some thougn not all industrial studies, and recordings of indus^al noise have produced increases in diastolic blood pressure and total peripheral resistance lasting longer than the noise^Work performance, accident rates, and behaviour map, also be adversely affected, though confounding factorshard to exclude in real-life studies; and Broadbcnt a study by G Jansen suggesting that steelworkers workhfg in noise had more domestic disputes than others/3 / . Reducing noise and even prodding^rersonal protection, however, cost money--how much is notneirt but the notes in the background paper accorapanyin^^he consultative docu ment suggest large sums-^^ith liq/tcd resources more spent on reducing noise could.mean leas for other health and safety Noise at work measures. The commission maintains that few if any workers will be exposed for long to exactly 90 decibels since in practice the design target jinll needno be two or three deobels lower Enormous numbers of people are exposed to potentially than the limn. If" also emphasises the general requirement to reduce noise afi far as ^/reasonably practicable, and adds that damaging levels ofnone at work. In Britain, in manufacturing industry alone about 600 000 work in noise levels above an average of 90 decibels*--noisy enough to make shouting it "will keep the auction under review, and will consider whether to/propoy >ome lower value if this seems correct in the lightoffunm^evclopmcms." The present proposal, then, necessary for talking to someone standing at arm's length. Over 21 million more work in levels over 80 decibels. The Health and Safety Commission has now1 proposed could indeed he seen as "a sensible first step in legislation." Neverth^ss, a general obligation to keep noise levels as low' as possible and below the statutory limit would not legislation based on the 1972 voluntary code of practice* and the specific regulations that apply to a few industries. The central provisions are that exposure to noise must be reduced to the lowest level that is "reasonably practicable"--whether / influcncythc poorer or less conscientious firms as powerfully as wmfld a lower statutory limit; and once the figure of 90 dedmas is enshrined in legislation we cannot realistically expect any early change; The document speaks ofthe crippling by reduction of the actual noise or, failing that, by ear protec* tion--and that no one must be exposed to more than.90 decibels. Is this level too high ? '/ J Aosts to industries if the regulations were very stringent and 'the likelihood that some would have to dose. The social The risks of damage after given exposures are known quicy precisely, largely as a result of the work of Burns jfid Robinson.3 After a lifetime's exposure to 100 dedbehf the document states, 32% of people will have a hearing threshold level of 50 decibels or mote (averaged over 1, 2/andTC kHz), and with 90 and 80 decibels the proportions will be 11% and 3%. A hearing threshold of 50 decibels is cljd n/nr at which implications of different limits, in terms of direct and indirect costs and benefits, need to be spelt out quite fully in the coming months, so that society can make an informed choice. We need too to hear more of the "strongly held and differing views" about adoption of 90 dedbels as the main action level, to which the commission refers so laconically. Its final decision to propose 90 decibels must reflea Britain's present economic the DHSS starts to pay disability benefitsifhut this is a considerable level of disability. A hearing'le^l of 30 decibels circumstances and a substantially lower limit may not be feasible. But would not 85 dedbels, which would nearly treble or more represents impairment of th^nmderstanding of the number of workers protected by legislation, be preferable ? conversation even in low backgrounj^noise, and is now At the very least, this could be written into the regulations as recognised as such by the Industrial Injuries Advisory a "warning level." At all events, we hope that many of those Council/ A careful look at figure 1 ja*he document shows that in the health professions will send their views to the Health on this basis at least 40% of workers exposed to 90 decibels and Safety Commission before the end of April 198Zf will have some, handicap. Not ajr af this will be due to noise, but the steepness of the curvrfitHn 90 down to 80 decibels, 1 Health and Safety Conumssson. Protection of fueling ax work. Content of proposed regulations and draft approved code ofpractice and guidance note. at which some 26% are affected, points to the size of the problem. Moreover, somc^/those in noisy jobs, now more than. in the past, have appimsble amounts of noise in their leisure London: HMSO, 1981. 1 Department of Employment. Code of practice for reducing the exposure of employcd.persom to noise. London: HMSO* 1972. 5 Bums 9T, Robinson D. Hearing and noise in industry. London: HMSO, hours. An estimated jfr-12% of the large population of disco attenders have noimflobk*; and, though these kinds of pursuits are unlikely to enfind/bver a working lifetime, such combined 1970. * Industrial Injuries Advisory Council. Occupational deafness. London: HMSO, 1978. (Cmnd 7266.) * Bickerdike J, Gregory A. An evaluation of hearing damage risk to attenders exposure shouwnot/be forgotten. ' The Healtir md/Safety Commission says that it selected levels ovenpW deobels for mandatory action "to ensure that the greatar tfforris directed to areas ofgreatest need." But rhis wouldimve unprotected the far greater proportion of workers exposed to noise of80-90 decibels, as figure 2 in the document sha^s: the .proportions suffering damage are smaller but the absolute numbers could be large. A further point, not dealt with by tne commission, is that the damaging effects of noise are notmecessarily confined to hearing. Other health effects at discothetfues. Leeds: School of Constructional Studies, Leeds Polytechnic, 1979 (with Noise Advisory Council memorandum, June 1980). * Gloag D. Pollution and people [two articles on noise). Br Med J 198081 1325-7 and 1404-6. 7 Andr6n L Hsnsson L, Bfbrkman M, Jonsson A. Noise as a contributory factor in the development of elevated arterial pressure. Acta Med Seand 1980 07:493-8. " Broadbenr D. Human performance in noise. In: Harris C, ed. Handbook of noise control. 2nd ed. New York: McGraw-Hill, 197R:ch 17,1-19. * Health and Safety Commission. Some aspects of noise and hearing loss'. notes on the problem of noise at work and report of the HSE working group on machinery noise. London: HMSO, 1981. are less certain*; but higher blood pressures and incidences of `Deobeli" refers throughout to dB (A) over an eight-hour period. t Comments to: Mr C O Lcite, Health and Safety Executive, Hazardous Substances Division, Branch HSD D3, 25 Chapel Street, Londun NW1 5DT FMSI 05517 Canadian Medical Association Journal, 1981 (August 1); 125(3): 237-239. ^ TTOEH 5525 ^ Code 90 81 AND ENVIRONMENTAL HEALTH EDITORIALS Asbestos and public health F.D.K. Liddell, ma, ph d Asbestos workers are known to have an increased risk of pulmonary fibrosis, pulmonary carcinoma and dif fuse primary interstitial mesothelial tumours of the pleura and peritoneum. Suspicions about the role of inhaled asbestos in these conditions date back to 1906, 1935 and 1960 respectively, and these suspicions were confirmed in the 1930s, 1960s and 1970s. However, the health effects of asbestos are highly complex and only because of recent intensive epidemiologic in vestigation has synthesis become possible. Most of the epidemiologic evidence is in the first three references;1-3 clinical implications are discussed by Becklake.4 al though her epidemiology is somewhat outdated. It is now known that the probability of adverse effects from inhaled asbestos depends on the total number of fibres inhaled (a function of both duration of exposure and concentration) and the type of asbestos. Most asbestos used in the Western World is "chrysotile", which is mined mainly in Quebec. The rest is "amphibole" -- essentially crocidolite and amosite -- which is now mined only in South Africa. Although the mechanisms of retention and elimination of asbestos are not fully understood, on average and where expo sures are similar much less chrysotile than amphibole is found in the lungs post mortem.5 Thus, amphibole might appear to be the more potentially hazardous agent; indeed, this has been confirmed in many epide miologic studies. The findings conform to a "fibre gradient", with crocidolite the most hazardous and chrysotile the least. In humans one cannot measure "dose" -- that is, how much dust is retained in the target organ. At best, through personal samplers one can measure fibre con centrations in the air. However, in most studies even workplace asbestos levels have not been measured. Perhaps the best epidemiologic investigation was that of a birth cohort of all 11 379 persons bom between 1891 and 1920 who worked in chrysotile production in Quebec;2 this study was crucial for the Health and Safety Commission's advisory committee on asbestos in the United Kingdom.' The wide range of accumul- From the department of epidemiology and health, McGill University, Montreal Reprint requests to: Dr. F.D.K. Liddell, Department of epidemiology and health. McGill University. 3775 University St.. Montreal. PQ H3A 2B4 ated dust exposure -- that is, the summation, job by job, of dust concentrations multiplied by the number of years in the job -- allowed a study of the shape of ex posure-response relationships; that for lung cancer, in particular, was effectively linear.3,7,3 This finding and the results from other studies have provided strong evidence against a threshold or "safe" exposure, despite the hope expressed in an earlier review* that was cited, but not endorsed, by Gloag.1" There have been so few studies in which asbestos exposure has been assessed in more refined terms than duration of exposure that it is impossible to estimate exposure-response relationships for each fibre type and each response. However, we can evaluate, using certain reasonable assumptions, the fibre gradient for certain responses. Thus, the fibre gradient is almost certainly steeper for mesothelioma than for other responses.11'11 With lung cancer the gradient remains quite definite; in chrysotile production the risk is only mildly in creased;113 in chrysotile processing the risk may be higher, although the few small studies are difficult to' interpret; in pure crocidolite exposure the risk is always much higher;1'1'1' and working with amphibole-rich mix tures carries some of the highest risks.17,18 The association of asbestos exposure with gastroin testinal cancer is uncertain; some other etiologic factor may also be involved.11 Evidence concerning laryngeal cancer is even more equivocal,11,111 but this cancer is so rare that even if asbestos exposure enhanced the risk the effect on total mortality would undoubtedly be small. Diagnosis of asbestosis is far from standard, if only because no clinical signs are specific and some history of asbestos exposure is required. However, there is some support for the usual "fibre gradient", although rather shallow.1 Among asbestos workers cigarette smoking increases the risks of lung and laryngeal cancer but not meso thelioma or gastrointestinal cancer and perhaps not asbestosis. In Quebec the relative risk of lung cancer appeared to depend on smoking habits, the slope of the asbestos exposure-response line being steeper for nonsmokers than for smokers;3 however, in other stu dies the slopes seemed to differ only slightly.30,31 Either way it is clear that exposure to chrysotile in mining and milling at the current control limits is equivalent in car cinogenic potency to smoking about three or four ciaarettes a week. CMA JOURNAL/AUGUST 1, 1981/VOL. 125 237 ft FMSI 05518 In all asbestos-related diseases there is a long in terval between the first exposure and the onset of symptoms. Today's cases are attributable to working conditions 30 to 40 years ago. Over the next decade or so more cases will undoubtedly appear, perhaps decades after cessation of exposure. However, since conditions have generally been improving since about 1950 we may have already seen the worst of the pathologic effects of asbestos, despite some question able forecasts widely circulated in the United States.*2 Cases arising from the current levels of exposure will not be seen until well into the 21st century. Some of "today's cases" are discussed by Finkelstein and coworkers in this issue of the Journal (pages 259 to 262). They explain that most of their 172 former work ers certified as having asbestosis had first been exposed "when hygiene conditions were considerably worse than they are today"; their general findings support earlier results."'** However, specific interpretation is not easy because the 20 cases in which compensation was awarded between 1942 and 1969 may well have differed from the others for the following reasons: change in the criteria of certification during the three decades; a wide range of ages at the time of award; different types of exposure, particularly to amphibole; and accepted interactions of these factors. All these factors make Finkelstein and coworkers' discussion especially difficult to generalize from, particularly as, for most of the subjects, the period of observation following the award was rather short. Furthermore, their Fig. 1 is just another way of presenting the data for "all causes" that appear in their Table III; while there is undoubtedly a "horror story" from the past it is one told twice over and is not two such stories. Ne vertheless, one can only agree with the sentiment in their final remark: that hygiene must be maintained so that the future risk of asbestos-related disease is elim inated. No control limit can guarantee absolute safety; limits for industrial exposure are usually derived from find ings in workers with low exposures (by the standards of the 1950s). At a factory in Rochdale, Lancashire, England (where chrysotile was used mainly, although crocidolite was probably used substantially, even into the 1960s) no relation was found, among workers who had entered the industry after 1951, between lung cancer and asbestos exposure up to the equivalent of eight respirable fibres per millilitre of air (well above most countries' control limits) for 50 years." In Quebec no excess exposure was detected in those who had worked all their lives with asbestos at concentrations below about 20 fibres per millilitre.' Crocidolite should be subject to much more stringent regulations, including special care in any essential processing or handling. Because amosite is an amphibole and evidence against it is accumulating, it seems sensible for it to be sub ject to similarly stringent regulations. Asbestos exposure has occurred in nonoccupational settings,1 but it is unlikely that the concentrations can have been high enough for long enough" to constitute a serious hazard (except in instances of household contact with asbestos workers). Nonoccupational ex posure may occur around demolition sites or when certain building materials are damaged; if the materials contain amphibole strict precautionary measures must be taken during repair or replacement of building mate rials. On the other hand, the hazards during such pro cedures must not be allowed to be worse than those of leaving the asbestos in place. No ill effects have yet been demonstrated from the presence of asbestos in drinking water, food or beverages, or in the general atmosphere. Indeed, the air pollution around Thetford Mines, PQ was quite severe for many decades, but there was no evidence that excess asbestos-related mortality or morbidity in the local population could be attributed to anything other than occupational exposure. Finally, some adequate substitutes for asbestos do not resemble asbestos; however, of the ones that do, the better they mimic the properties of asbestos the more similar the effects on health are likely to be. Indeed, in experiments in animals glass fibres with the same dimensions as crocidolite produced biologic effects at least as serious as those of the amphibole.*7 At least 30 years' exposure of a large group of workers would be required to test the effects of any substitute; even then its safety could not be demonstrated. Decisions on control limits, on whether to repair or replace asbestos after it has been damaged and on sub stitutes for asbestos, must be based not on emotions stirred by the results of past exposure to clearly exces sive doses of asbestos but on the best available scien tific evidence. References 1. Acheson ED, Gardner MJ: The ill-effects of asbestos on health. In Health and Safety Commission: Asbestos. Final Report of the Advisory Committee on Asbestos, vol 2, HMSO, London, Engl, 1979: 7-83 2. McDonald JC, Liddell FDK, Gibbs GW. Eyssen GE, McDonald AD: Dust exposure and mortality in chrysotile mining, 19KP75. Br J Ind Med 1980;. 37: 11-24 3. Wagner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, lntl Agency for Research on Cancer, Lyon, France, 1980 4. Becklake MR: Asbestos-related diseases of the lung and other organs: their epidemiology and implications for clinical practice. Am Rev Respir Dis 1976; 114: 187-227 5. Rowlands N, Gibbs GW, McDonald AD: Asbestos fibres in the lungs of chrysotile miners and milters: a preliminary report. In Walton WH. Critchlow A (eds): Inhaled Particles V, Pergamon. Oxford (in press) 6. Health and Safety Commission: Asbestos. Final Report of the Advisory Committee on Asbestos, vol 1, HMSO, London, Engl, 1979 7. Liddell FDK, McDonald JC,Thomas DC: Methods of cohort analysis: appraisal by application to asbestos mining. J R Star Soc 1977; 140 (series A): 469-491 8. Berry G: Dose-response in case-control studies. J Epidemiol Com- miut Med 1980: 34 : 217-222 9. Zielhuis RL: Public Health Risks of Exposure to Asbestos. Report of a Working Group of Experts Prepared for the Commission of the European Communities, Director-General for Social Affairs. Health and Safety Directorate, Pergamon, Oxford, 1977 10. Gloag D: Asbestos -- can it be used safely? Br Med J 1981; 282: 551-553 . 11. McDonald JC: Asbestos-relateddisease: an epidemiological review. In Wagner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, Inti Agency for Research on Cancer, Lyon, France, 1980: 587-601 12. McDonald AD, McDonald JC, Pooley FD: Mineral fibre content of mesothelial tumours in North America. In Walton WH, Critch low a (eds): Inhaled Particles V, Pergamon, Oxford (m press) 13. Rubino GF, Piolatto G, Newhouse ML, Scansetti G, Arisini GA. Murray R: Mortality of chrysotile asbestos workers at the Balangero Mine. Northern Italy. Br J Ind Med 1979; 36: 187-194 14. Jones JSP. Pooley FD, Sawle GV, Madeley RJ, Haggerwal A, Smith PG. Berry G, Wignall BK: The consequences of exposure to asbestos dust in a wartime gas-mask factory. In Wagner JC <ed): Biological Effects of Mineral Fibres. IARC sci pub no 30, Inti Agency for Research on Cancer, Lyon. France, 1980: 637-653 15. McDonald AD, McDonald JC: Mesothelioma after crocidolite ex posure during gas mask manufacture. Environ Res 1978; 17: 340-346 238 CM A JOURNAL/AUGUST 1, 1981/VOL. 125 FMSI 05519 i > ( 16. Hobbs MST. Woodward ST, Murphy B, Musk AW, Elder JE: The incidence of pneumoconiosis, mesothelioma and other respiratory cancer in men engaged in mining and milling crooidolite in Western Australia. In Wagner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, Inti Agency for Research on Cancer, Lyon, France, 1980: 615-625 17. Newhouse ML: Asbestos in the workplace and the community. Ann Occup Hyg 1973; 16: 97-107 18. Seljkoff IJ, Hammond EC: Multiple risk factors in environmental cancer. In Fraumeni JF Jr (ed): Persons at High Risk of Cancer: an Approach to Cancer Etiology and Control. Proceedings of a Conference, Key Biscayne, Florida, December 10-12, 1974, Acad Pr, London, Engl, 1975 : 467 19. Newhouse ML, Gregory MM, Shannon H: Etiology of carcinoma of the larynx. In Wagner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, Inti Agency for Research on Cancer. Lyon, France. 1980: 687-695 20. Saracci R: Asbestos and lung cancer: an analysis of the epidemio logical evidence on the asbestos-smoking interaction. Inti 1 Cancer 1977; 20: 323-331 21. Hammond EC, Selikoff D, Seidman H: Asbestos exposure, cigarette smoking and death rates. Ann NY Acad Sci 1979; 330 : 473-490 22. Peto R: Distorting the epidemiology of cancer: the need for a more balanced view. Nature 1980; 284 : 297-300 23. Liddell FD. McDonald JC: Radiological findings as predictors of mortality in Quebec asbestos workers. Br 3 Ind Med 1980; 37: 257 267 24. Berry G: The prognosis following certification with asbestosis in the United Kingdom. In Waoner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, Inti Agency for Research on Cancer. Lyon, France, 1980: 603-608 25. Peto J: Lung cancer mortality in relation to measured levels in an asbestos textile factory. Ibid: 829-836 26. Byron JC, Hodgson AA, Holmes S: A dust survey carried out in buildings incorporating asbestos-based materials in their con struction. Ann Occup Hyg 1969; 12: 141-145 27. Miller K: The in vivo effects of glass fibres on alveolar macro phage membrane characteristics. In Wagner JC (ed): Biological Effects of Mineral Fibres, IARC sci publ no 30, Inti Agency for Research on Cancer, Lyon, France, 1980: 459-465 College of Family Physicians of Canada's position on family medicine certification Donald I. Rice,* md I am hesitant to comment, on the views expressed in the editorial by Conn and OTjleagues, of the Canadian Federation of Medical StudeSte (CFMS), on family medicine certification,1 given tnV risk of further ag gravating the serious level of emodonalism that has already developed over the decision cw the College of Family Physicians of Canada (CFPC^to enforce its long-standing policy of limiting practice >hgibility to sit its certification examination in family medicine to family physicians who graduated prior to 1981. As one might have anticipated, the undergraduate student body as represented bv the CFMS is particularly seS^i- tive to this issue. \ I would be less than responsible, however, if I per mitted a number of the statements made in Conn and colleagues' editorial to go unchallenged. Since the CFPC and myself, as the college's chief executive of ficer, appear to be the authors' main focus for atten tion, 1 want to share with the readers of the Journal several observations, in the hope that a more balanced perspective may be provided as to what the policy of the CFPC on practice eligibility is all about. The following claims represent the major thrust, as I interpret it, of their editorial. Claim 1. The College of General Practice of Can ada, now the College of Family Physicians of Canada, was established by the Canadian Medical Association (CMA) in 1954, and was provided with a mandate as set out in the original objectives of the college. Response: True. The CFPC, like the Royal College of Physicians and Surgeons of Canada, the Canadian Executive director. College of Family Physicians of Canada Reprint requests to: Dr. Donald I. Rice, Executive director. College of Family Physicians of Canada, 4000 Leslie St.. Willowdale. Ont. M2K 2R9 Cancer Society, the Canadian Council on Hospital Accreditation and a number of other national associa tions, was established by the CMA. The CFPC is proud of this heritage and continues its close associa tion with the CMA as an affiliate member. Like other national associations established by the CMA the CFPC is an independent, autonomous body, incorporated by an act of Parliament, that is free to establish its pol icies within the framework of its act of incorporation. Claim 2. Over a period of 25 years the CFPC has digressed from its original objectives, which, in the words used by Conn and colleagues, represents a "clear violation of the second proviso set out by the VMA in 1954". They provide two examples: (a) the piSmotion of graduate (residency) training in family medniuie and (b) involvement in economics and politics. They pfcnt out in particular that the CFPC has some how performed a disservice to Canadian medicine and to the Canaan public by creating a two-tier system -- certificated ami noncertificated family physicians. Response: Tree and false. The CFPC takes pride in the knowledge thaKits policy has changed significantly since its inception Zi years ago, in response to the changing needs of fanuli physicians and the Canadian public for the services prmnded by family doctors. The decision to establish residency training programs in family medicine was made b)kjDractising family phy sicians who, on the basis of tbeir own experience, acknowledged that a period of urniergraduate training and an internship were inadequatAj>reparation for family practice. The CFPC is concerned and involvec^With large-E economics and large-P politics within theYtamework of doing those things necessary to maintain abstract ive practice climate for its members. This inter^. has stopped short of involvement in the nitty gritty ofbiw and how much a family physician should be paid amL CMA JOURNAL/AUGUST 1. 1981/VOL. 125 239 FMSI 05520 V- lNST'Ti'TP. OPOCCUPAli' AND Eh'" '-\'MSNTAL HE.-V k.-JNTREAI* CANA.;,- M Thorax, 1981, 36, 241-244 IOEH 5511 Code 0 1 9 90 81 Editorial Asbestos and public health The highly complex health effects of asbestos have been the subject of many symposia and publications over the last two decades. The Health and Safety Commission's Advisory Committee on Asbestos produced its final report1 late in 1979; in a second volume (ofcommissioned papers), there is an excellent review by Acheson and Gardner of the 111-Effects of Asbestos on Health.1 More recently, McDonald's team has published the latest report of mortaii'y in a very large cohort of Quebec miners and millers.1 and the proceedings of a symposium held at the (WHO) International Agency for Research on Cancer, in Lyon, in September 1979, have now appeared.4 At the September 1980 British Occupational Hygiene Society's Symposium on Inhaled Panicles, all but one of the asbestos papers fitted and helped to fill out the pattern. The findings from the last were completely out of line with anything that has gone before; so much so that reasonable scientists must await careful evaluation before allowing this paper4 to influence judgement based on scores of well-authenticated reports. With this single exception, matters have become sufficiently clear that reasonable synthesis seems possible. This editorial draws freely on many sources but, to keep within reasonable limits, it concentrates largely on mortality in relation to asbestos exposure, with especial reference to death from lung cancer. Morbidity is given less weight because of the greater difficulties of diagnosis and attribution, particularly bearing in mind the inter actions with smoking. In view of the recent editorial,4 mesotheliomas are reviewed only cursorily. The bulk of the Western World's asbestos is "chrysolite." a magnesium silicate mined mainly in Quebec; the remainder consists of the so-called "amphiboles," ferrous and ferric silicates--almost entirely crocidolite and amosite--both now produced only in South Africa. Russia produces almost as much chrysottle as the rest of the world, mainly for "home" consumption, but no amphibole. It is now thought that the dimensions of retained fibres in the lung-are more important in carcino genesis than their chemical composition, while the processes of inhalation, elimination, and retention also seem to depend on physical characteristics. It is Addms for reorim rvoimia: Professor O Liddell. Department of Eoidemioioty end Health. McGill University, 3773 University Street, Montreal. Quebec, Caiuuia H3a 284. possible that chrysotile may not penetrate deeply and that those fibres which are initially retained become susceptible to the system of natural elimination including dissolution. Amphiboles, on the other hand, may penetrate quite deeply, and their relative indestructibility seems to lead to retention more or less indefinitely. Whatever the mechanisms, it seems that, on average and where exposures are similar, considerably less chrysotile than amphibole is found in the lung post mortem.7 5 If these beliefs are wellfounded, it is clear that, compared with chrysotile, the amphiboles have substantially greater potential for causing ill effects. Also important is that, with all asbestos-related disease, there is a delay, usually of several decades, between first exposure to respirable fibres and the onset of symptoms. Thus, today's cases are not attributable to present conditions, but to those of 40 or more years ago. Meanwhile new cases can arise several decades after cessation of exposure. It is also clear that it could take up to half a century before there would be any serious possibility of discovering whether control had been satisfactory and perhaps a like period before the appearance of a new hazard. In no study in man is it possible to measure the dose --that is. the amount of dust reaching or better, retained in, the target organ. At best, measurements can be made of the fibre concentrations in the air close to that inspired by individual workers. Personal samplers are now in use, but in the past only assess ments in the general area of the work place were possible, and even these have been available in very few studies. With such measurements, linked to work histories, it is possible to assess roughly the exposure experienced by each worker in certain periods, and this may be a reasonably satisfactory index of dose. Without such measurements, the only available surrogate is duration of exposure; this has been used in most studies, but is obviously less than satisfactory and. indeed, of dubious validity. If only for the difficulties just mentioned, there can be no ideal epidemiological study. The best have been in occupational settings and that of Quebec chrysotile production workers4 has many advantages over most. A large birth cohort of 10 939 males and 440 females who worked at least a month in the industry, some as early as 1904, was followed to the end of 1975; only 2?o of those known to be alive in 1936 were untraced "0 FMS1 05521 i 242 and there had been nearly 4500 deaths. Estimates of been reported, and their findings are all difficult to respirable dust concentrations, job by job. were interpret; they are not incompatible with risks higher obtained to cover the relevant periods of exposure, than in mining and milling, but much less than with and smoking histories were obtained for almost all the amphiboles (except in reference 5). The risk of men alive in 1950. The data have been analysed by lung cancer in pure crocidolite exposure was much different methods.* and the results have been consist higher in Nottingham.17 Eastern Canada,10 and ent.10 The inclusion of those with very short employ Western Australia.1* and some of the highest risks ment led to a very wide range of accumulated dust were in amphibole-rich mixtures.30 31 exposure--the summation, job by job. of dust Gastrointestinal cancers appear to be asbestos- concentration multiplied by years in the job. This associated only in certain circumstances, so that some overcame some of the problems of selection and other aetiological factor may also be involved.14 survival that arise when a study group is defined in Cancer of the larynx was clearly unrelated to asbestos terms of long employment, or employment at a exposure in Quebec3 and in London,33 but has been particular time, as has been usual in other studies. associated with asbestos, probably amphibole. in a The advantage of having such a wide range of few other studies.1 Fortunately, the tumour is rare exposures is that it allows study of the shape of the and even if there is an enhanced risk from asbestos exposure-response relationship; in Quebec for lung exposure, the absolute effect on total mortality is cancer it was effectively linear.3*11 Results from undoubtedly small. Diagnosis of asbestosis in life is other studies have tended to support such a relation so difficult (because even the radiological signs are not ship11 or one that is possibly "sub-linear"--that is, specific and some history of asbestos exposure is steeper for short periods of exposure and more required) that reliable comparisons between fibre shallow for longer periods.13 Either way. there seems- type can only be made where the diagnoses have been no evidence for a threshold or "safe" dose. The made by the same team. Diagnosis of asbestosis at implications of this are important. death is inevitably related to awareness and to It is difficult to make tiuantitative comparisons compensation procedures. However, what evidence between the health effects of the different types of there is suggests the same gradient, if perhaps even asbestos fibre. For many purposes, mixtures of less steep, by fibre type.3 amphibole and chrysotile have been found satis Cigarette smoking is an important factor in lung factory from a commercial point of view, and there cancer and cancer of the larynx, but does not seem to are very few large groups of workers (other than affect the risk of mesothelioma or of gastrointestinal miners and millers) who have been exposed to a cancer, and perhaps not of asbestosis. In the latest single fibre type. Even were single-fibre' studies Quebec data on lung cancer.3 the slope of the asbestos possible, differences in selection and management exposure-response line appeared to depend on policies, problems over reference populations, and smoking habits, being steeper for non-smokers than the almost certain lack of information on dust for definite smokers; other data seem to fit the concentrations, all militate against reliability of multiplicative model better.33 34 However, from the comparison. In mixed-fibre studies, where groups of Quebec data it is clear, whatever the model, that workers have been distinguished by exposure to a today's control limits for occupational exposure to single fibre, the above factors should be standardised. chrysotile arc equivalent in carcinogenic potency to However, the classification by fibre type may have very light smoking--that is, only three or four been imperfect (or there may have been contami cigarettes smoked each week. nation), and numbers have tended to be rather small. Bearing in mind the lag period, it is important to The findings from most studies to date appear to consider findings in men with low exposures to support the hypothesis of a "fibre gradient", such chrysotile. In analysis of the Quebec data on lung that crocidolite has much the most severe health cancer up to the end of 1973, it was reported* that no effects, and chrysotile the least, with amosite some excess was detectable (at anything approaching a where in between. This is compatible with the beliefs conventional level of statistical significance) where expressed in the third paragraph of this editorial, and exposure was less than a certain amount, based on the the gradient seems to exist with all health hazards. integration over time of a concentration of respirable For mesothelioma, the gradient is undoubted, and is asbestos dust. This is equivalent to saying that ton almost certainly steeper for this outcome than for any the linear hypothesis and with a conservative fibre/dust other.14 13 conversion ratio) in a 50-year working lifetime, in As to lung cancer, the gradient still exists although concentrations below about 20 respirable fibres per it is probably not quite so steep. Chrysotile production millilitre of air(ie 20 f/ml), no excess could have been has yielded a comparatively mild excess.3 10 Only a detected with any confidence. The proposed control handful of small studies in chrysotile processing have limit1 for occupational exposure to asbestos is of a FMSI 05522 243 concentration of I f/ml. or one-twentieth that indicated in parentheses above. At Rochdale (where the fibre was mainly chrvsotile but where there was probably significant use of crocidolite even into the 1960s). no relationship was found--in those who had entered the industry after 1951 --between lung cancer and exposure up to the equivalent of 8 f/'ml for 50 years.35 Other low exposures have occurred in nonoccupational settings3; it is difficult to see how concentrations can have been severe enough for long enough for them to have been a serious hazard in the past, except where there was domestic contact with asbestos workers. Oyer the next decade or longer, more cases of asbestos-related disease will undoubtedly appear-- but they will be the result of working conditions 30--tO years before their appearance. However, conditions have generally been improving for at least two decades, and it is possible that we are already over the worst of the pathological effects, despite some questionable forecasts informally, but wideiy, circu lated in the USA.2* Should any cases arise from today's levels of exposure they will not be seen until well into the twenty-first century. The latest UK government regulations limiting occupational exposure to chrysotiie to I f/ml from 1981 do not. of course, guarantee absolute safety. Even if it were assumed that there had been no crocidolite at the Rochdale factory, and putting a very gloomy interpretation on the findings there. 50 years' continuous exposure to the upper limit allowed by the regulations would lead to I '25% excess mortality from lung cancer, or about 0125% overall excess.3 Crocidolite is, of course, subject to much more stringent regulations--no more imports and especial care in any necessary processing or handling. As amosite is an amphibole, and as evidence against it is accumulating, it would seem sensible to treat it on the same lines as crocidolite. One reason is that chrystotile can. perhaps with some ingenuity, often be used instead, although there will remain specific problems such as the production of large-diameter pressure pipes. Non-occupational exposures to asbestos may continue because of demolition or damage to certain buiiding materials, particularly those used for insulation where the materials often contained amphiboles. There is no doubt that any exposure to respirable asbestos is to be avoided wherever possible, and strict precautionary measures during repair or replacement are clearly indicated. Neverthe less. great care must be exercised that the hazards during such operations are no worse than those arising "naturally''--that is. from leaving the asbestos in situ, it is also important to note that the peak figure identified in a survey27 of buildings containing asbestos materials in the UK was only 0 03 f/ml (one-twelfth of the proposed control limit for occupational exposure); however, it is agreed that more information is required "about asbestos levels in new and old buildings in relation to type and usage of asbestos-containing materials, particularly insulation materials."2 No ill effects have yet been demonstrated of asbestos in drinking water, in food and beverages or in the general atmosphere. Indeed, the pollution of the air around Thetford Mines, by far the dustier of the two mining areas in Quebec, was for many decades quite severe, but there was no evidence that excess asbestos-related mortality or morbidity in the general population could not be attributed to occupational exposure. A final point must be made about substitutes for asbestos. Some adequate substitutes exist which do not resemble asbestos; where they do. it should be borne in mind that the better a substitute mimics the physical and chemical properties of asbestos the closer the ill effects on health are likely to be. Indeed, there is animal evidence2* that glass fibre of the same dimensions as crocidolite has biological effects at least as serious as those of the amphibole. It must also be emphasised that at least 30 years' exposure of a large group of workers would be required to "test" any substitute, and even then it would not be possible to demonstrate that it was "safe." DOUGLAS LIDOELL Department of Epidemiology and Health McCill University, Montreal, Quebec, Canada References 1 Health and Safety Commission. Asbestos Volume I: final report of the advisory committee. London: HMSO, 1979. 2 Acheson ED, Gardner MJ. The ill-effects of asbestos on health. In; Health and Safety Commission. Asbestos Volume 2: papers prepared for the advisory committee. London: HMSO 1979: 7-83. 3 McDonald JC. Liddell FDK. Gibbs GW, Eyssen GE. McDonald AD. Dust exposure and mortality in chrysotiie mining, 1910-75. Br J htd Med 1980; 37: 11-24. 4 Wagner JC Ed. Biological Effects of Mineral Fibres: Proceedings of a Symposium organised by lARC. the French National Institute of Health and Medical Research and the Medical Research Council, Penarih, UK, held at the International Agency far Research on Cancer. Lyon. France, 25-27 September 1979. IARC Scientific Publication no 30. Lyon: IARC, 1980. 3 Dement JM. Harris RL. Jr, Symons MJ, Shy C. Estimates of dosc-rcsponsc for respiratory cancer among chrysotiie asbestos textile workers. In: FMSI 05523 244 ' Walton WH, Crilchlow A eds. Inhaled Particles V. Oxford: Pergamon Press, 1981. 6 Elmes PC. Mesotheliomas, minerals, and man-made mineral fibres. Tltorax 1980; 35: 561-3. 7 Gibbs GW. Personal communication. 1980. 8 Rowlands N, Gibbs GW, McDonald AD. Asbestos fibres in the lungs of chrysotile miners and millers: a preliminary report. In: Walton WH, Critchiow A eds. Inhaled Particles V. Oxford: Pergamon Press, 1981. 9 Liddell FDK. McDonald JC Thomas DC. Methods of cohort analysis: appraisal by application to asbestos mining. J R Slat Soc 1977; Series A 140: 469-91. 10 McDonald JC, Liddell FDK. Mortality in Canadian miners and milters exposed to chrysolite. Ann NY AcadSci I979;330: 1-10. 11 Berry G. Dose-response in - case-control studies. 1 Epidemiol Common Med 1980; 34:217-12. 12. Henderson VL, Enterline PE. Asbestos exposure: factors associated with excess cancer and respiratory disease mortality. Ann NY Acad Sci 1979; 330: 117-26. 13 Seidman H, Lilis R, SeiikolT 1J. Short-term asbestos exposure and delayed cancer risk. In: Niebergs HEed. Proceedings of Third International Symposium on Detection and Prevention of Cancer, Part l, Vol 7. New York: Marcd Dekker, 1979: 943-60. 14 McDonald JC. Asbestos-related disease: an epidemiological review. In: Wagner JC ed. Biological Effects ofMineral Fibres: Proceedings ofa Symposium organised by IARC, the French National Institute of Health and Medical Research and the Medical Research Council, Ptnarth, UK. held at the Inter national Agencyfor Research on Cancer, Lyon, France, 23-27 September 1979, IARC Scientific Publication no 30. Lyon: IARC 1980:2:587-601. 15 McDonald AD. McDonald JC, Pooley FD. Mineral fibre content of mesothelial tumours in North America. In: Walton WH, Critchiow A eds. Inhaled Particles Y. Oxford: Pergamon Press, 1981. 16 Rubino GF, Piolatto G. Newhouse M L. Scansetti G. Aresini GA Murray R. Mortality of chrysolite asbestos workers at the Baiangcro Mine, Northern Italy. Br J Ind1 Med 1979: 36: 187-94. 17 Jones JSP, Smith PG, Pooley FD et al. The con sequences of exposure to asbestos dust in a wartime gas-mask factory. In: Wagner JC ed. Biological Effects of Mineral Fibres: Proceedings of a Sym posium organised by IARC. the French National Institute of Health and Medical Research and the Medical Research Council, Penarth, UK, held at the International Agency for Research on Cancer, Lyon, France, 23-27September 1979. IARC Scientific Publication no 30. Lyon: IARC, 1980'; 2:637-53. 18 McDonald AD. McDonald JC. Mesothelioma after crocidolite exposure during gas mask manufacture. Environ Res 1978: 17: 340-6. 19 Hobbs MST, Woodward SD. Murphy B, Musk AW, Elder JE. The incidence of pneumanoconiosis. mesothelioma and other respiratory cancer in men engaged in mining and milling crocidolite in Western Australia. In: Wagner JC ed. Biological Effects of Mineral Fibres: Proceedings of a Symposium organised by IARC, the French National Institute of Health and Medical Research and the Medical Research Council, Penarth, UK, held al the Inter national Agency for Research qn Cancer, Lyon, France, 25- 27 September 1979. IARC Scientific Publication no 30. Lyon: [ARC 1980; 2:615-25. 20 Newhouse ML Asbestos in the workplace and the community. Ann Oecup Hyg 1973; 16:97-107. 21 SelikoflT IJ, Hammond EC Multiple risk factors in environmental cancer. In: Fraumeni JF ed. Persons at high risk ofcancer: an approach to cancer aetiology and control. London: Academic Press, 1975: 467. 22 Newhouse ML. Gregory MM, Shannon H. Etiology of carcinoma of the larynx. In: Wagner JC ed. Biological Effects of Mineral Fibres: Proceedings of a Symposium organised by IARC, the French National Institute of Health and Medical Research and the Medical Research Council, Penarth, UK, held at the International Agency for Research on Cancer, Lyon, France, 23-27 September 1979. IARC Scientific Publication no 30. Lyon: IARC 1980; 2:687-95. 23 Saracd R. Asbestos and lung cancer: an analysis of the epidemiological evidence on the asbestos smoking interaction. IntJ Cancer 1977; 20:323-31. 24 Hammond EC SeiikolT IJ, Seidmann H. Asbestos exposure, cigarette smoking and death rates. Ann NY AcadSci 1979:330: 473-90. 25 Peto J. Lung cancer mortality in relation to measured dust levels in an asbestos textile factory. In: Wagner JC ed. Biological Effects of Mineral Fibres: Pro ceedings of a Symposium organised by IARC, the French National Institute of Health and Medical Research and the Medical Research Council. Penarth, UK, held at the International Agencyfor Research on Cancer, Lyon, France. 23-27 September 1979. IARC Scientific Publication no 30. Lyon: IARC 1980; 2:829-36. 26 Peto R. Distorting the epidemiology of cancer: the need for a more balanced view. Nature 1980; 284: 297-300. 27 Byron JC. Hodgson AA, Holmes S. A dust survey carried out in buildings incorporating asbestos-based materials. Ann Oecup Hyg 1969: 12: 141-5. 28 Miller K. The in viva effects of glass fibres on alveolar macrophage membrane characteristics. In: Wagner JC cd. Biological Effects of Mineral Fibres Pro ceedings of a Symposium organised by IA RC, die French National Institute of Health ami Medical Research and the Medical Research Council. Penarth, UK, held at the International Agency for Research on Cancer, Lyon. France, 25-27 September 1979. IARC Scientific Publication no 30. Lyon: IARC 1980; 1:459-65. FMSI 05524