Document X7dr3YX9e8p5akm2gVozVdD6G

FILE NAME Brakes BRK DATE 2004 DOC BRK149 DOCUMENT DESCRIPTION Letter to the Editor American Journal of Industrial Medicine RE Asbestos in Brakes Exposure and Risk of Disease Included in 143 AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 315-318 2004 Letter to the Editor car RE Asbestos in Brakes Exposure and Risk of Disease To the Editor nine areas of consideration strength of association tempor- Dr. Teta has claimed that my review of the issues surrounding the ability of asbestos to cause disease in persons exposed to asbestos from brakes does not meet her standards of equivocal or evidence Dr. Teta's comments are related only to mesothelioma a sentinel disease for asbestos exposure Mullan and Murthy 1991 while my paper ality biologic gradient consistence specificity biologic plausibility coherence experimental evidence and analogy As Hill noted one of my nine view points can bring indisputable evidence for or against the cause and effect hypothesis and none can be required as a sine qua non Instead he emphasized that the responsibility for making causal judgments rested with a scientific evaluation of the discusses all asbestos diseases totality of the data He further acknowledged that decisions Dr. Teta asks if consistency in epidemiological studies on causation must be made in the absence of perfect data Dr. that suffer from imprecision justify a conclusion that the Teta has chosen to base her criticisms of my paper without a evidence is equivocal As Ahlborm et al 1990 report through review of other factors as suggested by Prof. Hill e It should however be of equal importance to evaluate those Let us look at these nine criteria individually in deter- studies which do not suggest any effects of a particular mining whether the data are equivocal and whether the ee of exposure on the risk of disease i.e. studies in which the effect weight the evidence is lacking in supporting the contention measured is estimated to be near the null value In my paper of no relationship with exposure to brakes containing ed I have evaluated such studies and have pointed out why we asbestos and the probability of asbestos disease onl yn cannot rely solely on them for determination of cause and the fel ot First strength of association is a reflection ofthe strength a thy reason why the epidemiology is equivocal of effect in a given study The epidemiological studies stanbomie a evidence recognizes much more than just conducted to date have suffered from variety of impressions ie epidemiology as defined by the Agency for Toxic Substances and confounders as I have pointed out and thus cannot be and Disease Registries ATSDR in making causal used by themselves to justify a effect conclusion associations I do point out the inherent limitations of human observational data and place it in context with the experi- mental data as well as the available exposure data As pointed Second temporality requires that the cause precede the effect In each ofthe studies referenced by myself and Dr. Teta this has been true out by Dr. Sven Hernberg Editor Emeritus of the Scandinavian Journal Work Environment and Health A Third biologic gradient is the existence of a doseresponse relationship for the proposed cause effect combina- Fie truly negative study must i be large ii be sensitive and iii tion Experimental studies and epidemiological studies of Be have documented exposure data Hernberg 1981 asbestos have demonstrated such a response relation- A One of the most widely acclaimed methodologies for ship Specifically the studies of brake workers that have determining cause and effect for general causation was proposed by Sir Austin Bradford Hill in 1965 Hill 1965 In using this methodology researchers are asked to evaluate examined this issue have shown this response effect for example Iwatsubo et al 1998 clearly came to this conclusion Fourth consistency requires that a proposed effect be Dr. Lemen is retired Assistant Surgeon General USPHS and retired Deputy Director and Acting Director NIOSH and has testified as a plaintiff's expert in brake exposure cases observed repeatedly under different circumstances While some of the epidemiology studies have been consistent in GA Correspondence to Richard A Lemen 241 Rose Ridge Court Canton 30115 mail richard ralemen org a Accepted 26 May 2004 atk DOI 10 1002 ajim.20061 Published online in Wiley InterScience Ho Ate wwwinterscience wwwinterscience wiley.com 1 evidence- evidence analysis involves the balanced review and integration of relevant exposure toxicologic epidemiologic medical and health outcome data to help determine whether exposures to contaminant levels under specific condi- tions might result in harmful effects ATSDR 2004 a '2004 Wiley Inc. 316 Lemen their lack of significance they have also been consistent in their lack of strength to detect an effect either due to size confounders or lack of sufficient work histories Fifth specificity requires that each cause have a single effect This is rarely a useful criterion when dealing with environmental carcinogens because most of them cause more than one type of cancer Monson 1996 Such is the case with exposure to asbestos Sixth biologic plausibility seeks to determine if the theory of causation fits known mechanisms of injury causation While it is impossible to have a complete understanding of the mechanisms of cancer causation the biologic facts known about the various asbestos fibers are clear that they cause both malignant and malignant report excesses in lung cancer and bronchus ICD for all age groups except ages 20-29 above 1.0 in the Automobile Mechanics and Repair Workers category 472 They also report seven cases of mesothelioma of the pleura with six of them occurring in the age group that could potentially have the longest latency age group 70-79 PMR = 184 As indicated in the Technical Notes associated with the Occupational Mortality Database of the Washington State Department of Health from which Petersen and Milham 1980 and Milham and Ossiander 2001 data come from PMRs are of greatest value in studies where mortality rates cannot be calculated due to a lack of information on the population at risk particularly exposure history www3.doh.wa.gov/OCCMORT/HTML/OMtechnotes.htm diseases in experimental systems animals and humans 3/16/2004 Seventh coherence asks if the causal theory is not If Dr. Teta is referring to not including these papers into then I thank her since the inconsistent with what is already known of the injury or my count of mesotheliomas disease There is no biologic argument that explains why omitted articles add at least another 49 cases to the list of asbestos is incapable of causing asbestosis lung cancer or mesotheliomas reported among persons potentially exposed mesothelioma in one person while not in another to asbestos from brakes Olsen and Jensen 1987 Coggon evidence can consist of laboratory et al 1995 Milham and Ossiander 2001 Eighth experimental studies animal studies controlled clinical trials or observa- The Coggon et al 1995 paper relies only on the & studies The data in my paper support such occupation as listed on the death certificate which can be tional pathology misleading since a thorough examination of the decedents evidence Ninth analogy asks if epidemiological and other studies entire work history and their potential for exposure through home related brake work must be considered Also it must be have established that an environmental exposure analogous considered cause diseases similar kept in mind that PMR analysis is not equivalent to o2o: ttoo tthheoseexproespuorrteebdefionrg the exposuremabeying considered Lemen establishing a relative risk RR or an odds ratio OR and oat 2004 clearly supports this analogy that they only measure relative frequency of a particular control Dr. Teta states that Two other negative cause among all causes of death in the data set analyzed studies were not discussed both of which examined the risk Colton and Clapp 2000 While this study reports PMRS of in vehicle mechanics excluding men who 46 12 deaths from pleural cancer 88 3 deaths from of mesothelioma also worked in occupations known to pose a mesothelioma peritoneal cancer and 80 2 deaths from asbestosis the the deaths from all causes in this risk due to asbestos exposure McDonald and McDonald authors have not given 1980 Agudo et al 2000 The McDonald and McDonald category of motor mechanics as they have for the other paper simply stated that There were small excesses of some groups they list in their job group table making interpretation difficult Important findings of this study occupations but none in the larger notably garage of their findings the author's maintenance and transport work This paper does not which I did not include in my paper are contribute to our knowledge of brake exposed workers since conclusion that when assuming a linear exposure the no idea of whether exposures even existed to that pleural mesothelioma may be underestimated paper gave brakes in the three categories mentioned such an assumption Olsen and Jensen 1987 report on 93,810 cancer cases is pure speculation without giving detailed exposure data by occupation in Denmark for the years 1970-1979 While cancers for brake repair Agudo al 2000 is included in my paper and as the authors this study does not report specific of mesothelioma and the mechanics motor vehicles category does have workers the data do support an excess suggest risk of asbestos exposure however the numbers were quite lung cancer in occupations that may have included brake specifically small three cases from which to draw causal relationships repair workers however there is no way to and for this reason I included them only in my list of total identify brake repair exposures For example out of a total of 426 cases of mesotheliomas the authors found that the SPIR cases reported When Dr. Teta uses PMR less than 1.0 is not clear to what diseases or studies she is referring Take for example Petersen and Milham 1980 who report an increase for the automobile and repair category 472 for lung cancer a disease causally associated with asbestos exposure In later analysis of the same data base Milham and Ossiander 2001 for transport and communications to have 27 observed mesotheliomas when only 18.1 were expected Excesses were also found when looking at subgroups of transport and storage Most of the subgroups experienced elevated risks of mesothelioma with 21 cases observed and 13.5 expected SPIR 155 95 CI 101 23F8or the category of repair of ath Letter to the Editor 317 motor vehicles and motorcycles 34 lung cancers were observed but were not significant and no mesotheliomas were observed The question remains how many performed brake work versus how many did not Further analysis needs to be done to identify specific exposures to asbestos from brakes before this study can contribute to our knowledge concerning brake repair workers and potential diseases from asbestos While mesothelioma accounts for only 0.3 of the total cancers during this period in Denmark the authors say a substantial increase of mesothelioma in incidence has taken place since the inception of the Danish Cancer Registry in 1942. association exists between brake exposures from asbestos As stated by Phil Alderson Associate Director of the UK Cochrane Centre We need to create a culture that is comfortable with estimating and discussing uncertainty Alderson 2004 Therefore the conclusions of my paper can not be ignored by the serious researcher in determining the causal association between asbestos exposures to brake workers and disease Richard A. Lemen PhD MSPH REFERENCES Hodgson et al 1997 like the other surveillance data base studies is quite useful for evaluating overall general Agudo A Gonz^lezCA Bleda MJ Ramirez J Hern^ndezS L^pezF. Calleja A Panad^sR Turuguet D Escolar A Baltr^nM Gonz^lez- categories of cancer mortality trends but not specifically useful when looking at very specific occupational categories Moya JE 2000. Occupation and risk of malignant pleural mesothe- lioma A control study in Spain Am J Ind Med 159 168 or where the risks may be small The authors report PMRs for 27 groups of males 16-74 and have a table of 25 job groups where the PMRs exceeded 100. Motor mechanics were not Ahlborm A Axelson O Hansen ES Hogstedt C Jensen UJ Olsen J 1990. Interpretation of negative studies in occupational epidemiol- ogy Scand J Work Environ Health 153 157 included in this list and puts them into the lower category The authors point out that the confidence interval for motor mechanics is wide It should be noted that Where risks are small P values may well mislead confidence intervals are likely to be wide indicating considerable uncertainty Altman and Bland 1995 The lack of detailed occupational Alderson P. 2004. Absence of evidence is not evidence of absence We need to report uncertain results and do it clearly BMJ 476-477 Altman DG Bland JM 1995 Absence of evidence is not evidence of absence BMJ 311.485 ATSDR evidence update 2004. Chapter 8 Health effects evaluation Weight analysis In Public health assessment guidance manual Atlanta GA Agency for Toxic Substances and Disease histories again is a problem with this study Does this mean no Research United States Public Health Service Department of Health carve and Human Services risk Dr. Teta states that she and her colleagues obtained data from the Spirtas et al 1985 study using the SEER database and after analysis found no associations with brake work citing a paper in press by Hessel et al 2004 I would caution that without detailed occupational histories many specific Coggon D Inskip H Winter P Pannett B. 1995. Differences in occupational mortality from pleural cancer peritoneal cancer and asbestosis Occup Environ Med 52 775-777 Colton T Clapp RW 2000. Proportional Mortality Study In Benichou J Armitage P editors Encyclopedia of epidemiological methods Wiley reference series in biostatistics UK John Wiley & Sons Ltd occupational exposures can be overlooked Take for example the analysis by Pinto et al 1995 where nine cases of Hernberg S. 1981. Negative results in cohort studies to recognize fallacies Scand J Work Environ Health Suppl 4 121 126 eialtual mesothelioma diagnosed at their Institute which would have Hessel PA Teta MJ Goodman M 2004 Mesothelioma among brake atol si gone into the idiopathic category had not diligent mechanics An expanded analysis of control study Risk Analysis interviews with the patients and their relatives by a trained in press medical doctor been conducted Does the Hessel et al 2004 Hill B. 1965. The Environment and disease Association or causation study meet such through an examination Without being able Proc R Soc Med 295-299 to examine a paper in press or Letter to MJ Teta from JT Hodgson JT Peto J Jones JR Matthews FE 1997. Mesothelioma Walker of NIOSH an answer to this question is impossible and leaves me unable to make further comment Walker mortality in Britain Patterns by birth cohort and occupation Ann Occup Hyg 129 133 2002 While I agree with Dr. Teta that epidemiological studies provide the best data for causation which she contends have been done however it is not adequate to rely upon them when they suffer from inadequacies She points out " ... ignoring or misrepresenting the available evidence from epidemiological studies cannot be defended When one IPCS International Program on Chemical Safety 1998 Env Health Criteria 203 Chrysotile Asbestos XXX World Health Organization p 107 Iwatsubo Y Pairon JC Boutin C Menard O Mass~-nN Caillaud D. Orlowski E Galateau F Bignon J Brochard P. 1998. Pleural mesothelioma response relation at low levels of asbestos exposure in a French population control study Am J Epidemiol 133 142 looks at the true evidence as I have in my paper the five million people whom Nicholson et al 1984 estimated to have been vehicle mechanics in the US at some time in their careers deserve no less than a through review of the all the evidence before determining that no causal Lemen R. 2004. Asbestos in brakes Exposure anc risk of disease Am J Ind Med 229 23E7rrata In the text on page 234 left column last paragraph the citation of Anderson et al 1973 should read Henderson 2000 McDonald AD McDonald JC 1980. Malignant mesothelioma in North America Cancer .1650-1656 .1650-1656 318 Lemen Milham S Ossiander E. 2001 Occupational mortality in Washington State 1950-1999 Epidemiology Office Washington State Department of Health Monson RR 1996. Occupation 18. In Cancer epidemiology and prevention 2nd edn Schottenfeld D Fraumeni JF Jr. editors New York & Oxford Oxford University Press pp 373-405 Mullan RJ Murthy LM 1991. Occupational sentinel health events An dated list for physician recognition and public health surveillance Am J Ind Med 775-799 Nicholson WJ Daum SM Lorimer WV Velez H Lilis R Selikoff IJ Miller A Anderson HA Fischbein SA Holstein EC Rom WN Rosenman K Todaro JD Cheng W Li V Tarr DT 1984. Investigation of health hazards in brake lining repair and maintenance workers occupationally exposed to asbestos National Institute for Occupational Safety and Health NIOSH prepared by Mt. Sinai School of Medicine CUNY New York August Olsen JH Jensen 1987. Occupation and risk of cancer in Denmark An analysis of 93,810 cancer cases 1970-1979 Scand J Work Environ Health Suppl 1 91 Petersen GR Milham S. 1980. Occupational mortality in the state of California 1959-1961 NIOSH Research Report 33353 Pinto C Soffritti M Maltoni C. 1995. Ignored occupational risks of asbestos mesotheliomas Med Lav 484 489 Spirtas R Keehn R Wright W Stark A Beebe G Dickson E 1985 Mesothelioma risk related to occupational or other asbestos exposure Preliminary results from a control study Am J Epidemiol 518 Walker JT 2002 Letter to MJ Teta from JT Walker regarding inquiry about PMRs for auto mechanics with attached data sheets Department of Health and Human Services DHHS Public Health Service National Institute for Occupational Safety and Health