Document 50b31xnZwvK4O6bV5Vqbjq0Qz

SELECTED BIBLIOGRAPHY ON RISK ASSESSMENT All mar., W.F. 1 9 85. "We Have Nothing to Fear". Science 35, October, 1 935. pp. 38-41. Allman, W.F. 1985. "Staving Alive m the 20th Century", Science 95, October 1985, pp. 31-37. American Industrial Health Council Interdisciplinary Panel on Carcinogenicity 1984. "Criteria for Evidence of Chemical Carcinogenicity", Science 225: 682-687. Clayson, D.B., Krewski, D., and Munro, I., eds. 1985. Toxicological Risk Assessment, Volume I: Biological and Statistical Criteria, CRC Press, Inc., Boca Raton, Florida. Clayson, D.B., Krewski, 0., and Munro, I., Toxicological Risk Assessment, Volume Criteria and Case Studies, CRC Press, Raton, Florida. eds. 1985. II: General Inc., Boca Food Safety Council, 1982. A Proposed Food Safety Evaluation Process: Final Report of Board of Trustees, The Nutrition Foundation, Inc., Washington. Hanes, 3. and Wede1, T. 1985. "A Selected Review of Risk Models: One Hit, Multihit, Multistage, Probit, Weibull, and Pharmacox metic," J. Am. College Toxicol. Krewski, D. and 3rown, 7. 1931. "Carcinogenic Risk Assessment: A Slide to the Literature," Biometrics 3': 353-365. Krewski, D., Brown, C. and Murdoch, D. 1984. "Determining 'Safe' Levels of Exposure: Safety Factors or Mathe matical Models". Fund. Apd. Toxicol. 4: 5 3 8 3 -- 5 394 . National Academy of Sciences 1983. Risk Assessment m the Federal Government: Managing the Process, National Academy Press, 1983. Office of Science and Technology Policy, 1985. Carcinogens- "A Review of the Science and Associated Principles, Federal Register, 12372-10442. "Chemical Its Vol 50, LLA 000467 Samuels, 3.W. and Adamson, . H. 1 93 5. "Quantitative Risk Assessment: Report of the Subcommittee on Environmental Carcinogenesis, National Cancer Advisory Board." JNCI '4: 945-951. 5hodei1, M. 1985. "Risky Business," Science 85, October 1985, pp. 43-37. Stava, J.F. and Erdreich, L.S. 1935. "Advances in Health Risk Assessment for Systemic Toxicants and Chemical Mixtures," Toxicology and Industrial Health, Vo 1.1, Princeton Scientific Publishing Co., Inc., Princeton, N. J. U.S. Environmental Protection Agency, 1982. Pesticide Assessment Guidelines, Subdivision F, Hazard Evalua tion, Human and Domestic Animals, U.S. EPA, Washington, D.C . U.S. Environmental Protection Agency, 1984. Proposed Guidelines for Carcinogen Risk Assessment. Federal Register 49: 46294-46301. U.S. Environmental Protection Agency, 1984. Proposed Guidelines for Exposure Assessment. Federal Register 49: 46 304-46 3 1 2 / U.S. Food and Drug Administration, 1982. Toxicological Principles for the Safety Assessment of Direct Food Additives and Color Additives Used in Food. LLA 000468 RISK ASSESSMENT .OS LOW LE\ E_S OF ASSESTnS EXPOSURE -- Reg r.ad 1 hr to talk about the techniques by which we generate data and I have only one half that time to talk about interpret data. Unfortunately, I think that the proportion of time at this conference deio'ed to interpreting asbestos sampling results is much greater than the proportion of time at: effort devoted to interpretation of actual sampling results. We collect mountains of data, >et little effort is spent in learning what it all means :r communicating the results in an understandable fashion to th.se -no are exposed, or to those wn-. are responsible for exposures. -- This : s a deplorable d 1 ' J 3 t ion. As I . H . s we may be cal led upon to ;nte r p r e t s a m p 1 i n g da` a. Indeed some legal au: hone: tell us that we have a i uty to warn those who are at r: sk . -- Thus, the goal of my presen tation is to convey to you some information that you may f : n d helpful if need to assess and commun icate the risk of asbes* os exposure. i r i r s t I'll p r e s e r. \ rt Tj - j f the basic information on as fees tos risk. 2. Then, look at an example of risk assessment and communication of hazard. 3. Some of the basic information plus a bibliography has been handed out. Risk assessment-- The determination of the risk to a populations (humans or other organisms) from exposure to a hazardous agent. The scope of risk assessment ca'6. be extremely broad including: --source of agent --distribution m en\ i raiment --routes of exposure --populations exposed --degree of exposure --hazards associated --dose -response !~elat: onship --excess risk to populations --remedial actions LLA 000469 Assessments of this scope are often associated with contaminants in the outdoor environment. Fortunately, m I.H., concern is typically focused on small populations of people whose exposure may be characterized by direct measurement. e.g. workers on an abatement project or children in a school with ACM. objective is ften to estimate the excess risk to population members resulting from exposure and to communica t e that risk sc that informed act o n s may be taken. - -_ 7 h e r p are at least 3 i e u e i s i s k assessmen ! Fundamental Summary and Evaluative numerous papers m the literature on asbestos Applications to specific olrcumstances st'jdues ' Some of the components of our risk assessment Source- usually known Routes of Esposure- in halation Degree of exp.- air sampling results are obvious: Effects: -- We are usually concerned with mortality from asbestos related diseases since the data is available. Much less is known regarding morbidity and subclinical effects. -- Mortality is expressed as some measure of excess risk associated with asbestos exposure. Excess over what0 --often comparison is made with the general population, but workers are healthier than the general population. They show ~80% of the death rate of the gen. population when matched for age and sex. --Measures : Relative Risk, o'SMR -- for rare diseases, such as mesotheliomas, measures reach absurdly high values since the denominator very snal1. ooeo:is LLA 000470 Ar. alternative is Excess Risk--this is the absolute prooability of an outcome. Here 1 w 1 i i usually use deaths per 100.000 persons exposed. Dose-Response This is the ra o s r asbestos. But, it dose is seldom, if measured. complex part of the risk assessment for is something of a misnomer since the true every known-- only exposure levels are Factors affecting dose-response. Smoking Sichoisrr '1583 of It. 5: na i i r. r.: s recent study for EPA cites work by Seiikoff, Hammond, Berry and McDonald in support of the assertion that the effects of smoking and asbestos exposure on lung cancer are multiplicative. A study of over 12,000 asbestos workers showed that both s mp k e csanorT-sm e r s had 5 time the risk of dying from lung ca nferiHo uev , tTie risk is low for non-smokers; therefore, mulitpiyir.g it by five does not result in many cases, although any excess is undeslreable. On the Other hand, smoking itself caused a significant increase (1C X controls!. Thus, that high risk multiplied by 5 gives an immense increase. -- In a dissenting opinion Hughes and Weill cite the same studies, but come to the conclusion that the percentage increase in lung CA risk may be "somewhat less" for smokers than for nonsmokers . -- Smoking behav:or does no seem to affect the relationship between mesothelioma and exposure. Age at which Exposure Occurs Lung CA- (Relative Risk Model) -Relative risk increases after time of first exposure until 3540 years. The shape of these curves is independent of the age at first exposure. - Median time from first exposure to death was 32 years (Seiikoff) -Life time risk seems to oe cniy related to cumulative exposure (nicholson, 1983) LU 000471 -The i.atency period has been sho-n to be function of both exposure and age. The higher the exposure, the shorter the iate.nc; period. The older the person, the shorter the latency period. Bottom Line'. life time risk of Tung CA mortality seems to be dependent only on the degree of exposure. Mesothelioma- [Absolute Risk Model) -Incidence of death is independent of the age of first exposure and increases according to a power of time after a "delay period" of about 10 years Nicholson, 1 98 3) . - Med 1 an time to death after first evpsoure is about 36 years . - The rates of mesothelioma increases - i t n time more rapidly than truss of lung T A and r.ay equal those of Tu n g u A 45 years or so after first, exposure. - Thus, age of exposure is of considerable consequence for estimating mesothelioma risk. iloet`fn--Hi i~5--rs--u f t n n--n-e-fe- T-d>. t>e counted--,-oj:--eypl toi tl-y--rrr--r'l b'k--gnciimewti Nicholson's <~-&6rk for EPA has look at this in detail. Sex - Does not affect dose-respoose for mesothelioma, but males seem to have 2.5 to 3 time the excess risk of women for death by lung CA from asbestos exposure. Types of Asbestos - All forms of asbestos are capable of producing cancer. i However, there is controversy regarding whether crocidolite or amphiboles in general are more carcinogenic than chyr3otile. Standards in other countries, such as Britain and Sweden take the type of asbestos into account, while U.S. Standards do not. The epidemiological data are quite contradictory, possible because good particle 3ize distribution data were not available for the study exposures. LLA 000472 4 {-*!&---- ACM composition OA~ Proces s Gradients in lung CA rates have been observed in the progressive processing of asbestos. Lowest rates of lung CA are associated with raining and friction product production, and the highest with asbestos textile production. It seems likely that these gradients could be explained by differences in particles size distributions. -- PCM microscopy yields an index of exposure; it is a surrogate for the more complete characterization of exposure and thus may yield annomalies in dose-response relationships. -- There is in suffice n t e'-idenre "o determine if there are gradients in mesothelioma rates oy industrial process after accounting ror the type of asbestos used. Dose-Response Lung Cancer Two hypotheses seem epidemiologists who asbestos. to be widely accepted by toxicologist have examined the dose-response for and 1. After an# appropriate latency period, the relative risk of lung CA is independnent of the elapsed time between exposure and observation. 2. The relative risk of lung CA increases linearly with cumulative exposure. Cor re 1laries : a. We can use linear extrapolation from the doseresponse of industrial workers exposed at high levels years past to determine the risk of exposures several orders of magnitude less. in b. Cumulative dose may be represented by the product of the average concentration and the duration of exposure Although is is doubtful that the dose-response relationship for low exposures can ever be determined by direct observation, these assumptions are consistent with observations at higher levels of exposure and is suoported by the mathematical theoc of multistage carcinogenesis Also, there is general agreement chat it is unlikely that one risk increases more rapidly chan linearly at low levels. Thu3, it is believed that these assumptions lead to conservative estimates of risk. LLA 000473 Mesothelioma- In exposed workers, the incidence rate increases rapidly with age and duration of exposure, at least until the age at which other causes of death become important. Relatively little quantitative dose-response data is available. Thus, mathematical models have play a large role in estimating tisk mesothelioma mortal it;. . -f Nevertheless, Schneiderman i 19 8 1 ) observed that two industrial studies which investigated exposure gradients found response rates per unit dose which were of the same order of magnitude. Thus, he assumed a linear dose-response relationship. In ligr.t of the nature of mesothelioma risk, the use of his approach may r, o* be valid for assessing the risk of childhood exposures. Summary Dose-Response Data -Lifetime excess risk of death for exposure to 1 fiberyea r / c c . -Excess risks are in relatively good agreement compared with risk assessments for other agents. Summary of Asbestos Concentrations PCM-TEd Conversion -Adds an additional degree of uncertainty. -dost risk assessments deal with concentrations in fibers/cc. If TEM is used, ng/cu.d are converted to fibers/cc. VAlidity is dublOUS. Excess Cancer Deaths at Various Exposure Levels Case Study LLA 000474