[ Citation in PubMed ] [ Related Articles ]
Most models of how science works propose that competition between ideas contributes to the advancement of knowledge. Criticism of scientific work plays a part in facilitating such competition by exposing the strengths and weaknesses of rival explanations, encouraging debate, and suggesting alternatives. Nevertheless, not all criticism has equal value to the scientific process.
In a review of epidemiologic studies on fine particles and mortality that appeared recently in EHP, Gamble (1) charged that the two major studies on this topic (2,3) may have been compromised by bias, yet he offered no serious effort to evaluate the alleged errors with the same standards of rigor demanded of the original studies.
For example, in Gamble's (1) claim that the study findings are compromised by the ecologic fallacy, he failed to address two important issues. First, the major prospective studies of fine particles and mortality are not classical ecologic designs because only air pollution exposures are measured on the aggregate level; the outcome and potential confounders are based on individual-level measurements. Thus the biases stemming from the ecologic fallacy do not apply to these studies. Instead, these should be viewed as individual level studies in which exposure is measured with error (4). The calculations in Gamble's Table 2 (1) are erroneously portrayed as demonstrating the ecologic fallacy. Instead they appear to present strong evidence of a supralinear dose-response relationship between particles and mortality. This type of dose response would be expected if there existed a subset with much greater susceptibility, e.g., a bimodal distrubution of susceptibility. Second, the inference that the U.S. Environmental Protection Agency (EPA) and others have drawn from the studies' results is logically consistent with evidence based on exposure measured at the group level. The correct inference is that individuals living in communities with high air pollution levels have a higher risk of dying than people living in communities with low pollution levels and, therefore, that lowering community-wide air pollution levels should reduce community mortality rates. Such a policy is a logical and efficient means of minimizing the health impact of a widespread exposure (5).
Even when the criticism focuses on specific factors that might explain the results of the studies, it does not consistently address the potential magnitude and direction of alleged biases. For example, Gamble (1) concluded that lung function is a probable confounder of the observed relative risk for PM2.5 because the average lung function of the Six Cities cohort differs by city and because reduced lung function is a risk factor for mortality. However, neither the text nor Gamble's Figure 3 (1) explicitly identify possible ranges of confounding that this variable may have produced in the relative risk estimate of 1.26 derived in the original Six Cities Study. Gamble's Figure 3C shows that the average forced expiratory volume in 1 sec (FEV1) differs by approximately 0.1 L between the cohort members in the dirtiest and cleanest cities. In light of the cited 1.52 relative risk estimate for total mortality associated with a 1-L decrease in FEV1, it is difficult to imagine how adjustment for a FEV1 difference one order of magnitude smaller could explain the observed association (6). Perhaps more importantly, reduced lung function could be on the causal pathway between chronic exposure to particles and mortality, such that any adjustment for FEV1 would introduce bias. Sedentary lifestyle was posited as another potential confounder; however, Gamble (1) presented no evidence to suggest that sufficient differences in sedentary lifestyle among the six cities could account for the observed particle/mortality association.
Gamble (1) raised further doubts about cohort studies of mortality and fine particles by reviewing lists of criteria for epidemiologic studies (7,8). Hill never intended that his standards be used to exclude evidence. In his famous 1965 paper (7), Hill wrote,
I do not believe ... that we can usefully lay down some hard-and-fast rules of evidence that must be obeyed before we accept cause and effect.
Hill emphasized that guidelines can be helpful at the margins of epidemiologic interpretation, but the contribution of any particular study must be evaluated through careful assessment of the individual facts. He wrote (7),
None of my nine viewpoints ... can be required as a sine qua non. What they can do, with greater or less strength, is help us to make up our minds on the fundamental question...."
Hill's criteria appropriately would require a careful review not of one study design, but rather of the entire body of literature pertaining to the hypothesis of a causal association between air pollution exposure and health. This literature includes experimental studies in animals and clinical and epidemiologic studies in humans, examining outcomes ranging from direct measures of lung function to self-reported symptoms to hospital admissions and death.
Hertz-Picciotto's (8) criteria on the use of epidemiology in quantitative risk assessment are also cited by Gamble (1), with emphasis on their application for setting air quality standards. In fact, Hertz-Picciotto's framework was developed for the specific purpose of classifying "individual epidemiological studies as to their adequacy for use in dose-response extrapolation" (8), not for assessing weight of evidence regarding causation. Gamble (1) applied these criteria to a study design rather than individual studies; because he characterized them as having an ecologic design, he wrongly concluded that the alleged design weaknesses threaten the evidential base of the EPA's new PM2.5 standard. This use of Hertz-Picciotto's criteria (8) does not reflect the spirit in which they were intended, that is, "to have a reliable process for making the best use of available data ..." for dose-response extrapolation.
It is appropriate to criticize epidemiologic findings, especially when they have major implications for public policy, as the studies of fine particles do. However, it is much more helpful to evaluate the evidence that a given type of bias did (or did not) occur and to quantify its direction and magnitude than merely to suggest it is a possibility. Sensitivity analysis is an excellent tool for quantitative exploration of potential biases that can be used to gauge which of the many biases that can be envisioned are plausible explanations for a set of results and which are not. The magnitude of confounding can be calculated from several parameters related to the univariate distrubution of the postulated confounder and bivariate distributions involving the confounder, the exposure of interest, and the outcome (9). The kind of detailed analysis that these techniques encourage is fundamental to serious scientific criticism. Without these elements, Gamble's review (1) is largely an expression of opinion, more appropriately published as a commentary.
Dana Loomis
Irva Hertz-Picciotto
Marie O'Neill
School of Public Health
University of North Carolina-Chapel Hill
Chapel Hill, North Carolina
E-mail: dana.loomis@unc.edu
References and Notes
1. Gamble JF. PM2.5 and mortality in long-term prospective cohort studies: cause-effect or statistical association? Environ Health Perspect 106:535-549 (1998).
2. Dockery DW, Pope CA III, Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG Jr, Speizer FE. An association between air pollution and mortality in six US cities. N Engl J Med 329:1753-1759 (1993).
3. Pope CA III, Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, Heath CW Jr. Particulate air pollution as a predictor of mortality in a prospective study of US adults. Am J Respir Crit Care Med 151:669-674 (1995).
4. Morgenstern H. Ecologic studies. In: Modern Epidemiology (Rothman KJ, Greenland S, eds). 2nd ed. Philadelphia, PA:Lippincott Raven Publishers, 1998;459-480.
5. Rose G. Sick individuals and sick populations. Int J Epidemiol 14(1):32-38 (1985).
6. Cornfield J, Haenszel W, Hammond EC, Lilienfeld AM, Sumkin MB, Winder EL. Smoking and lung cancer: recent evidence and a discussion of some questions. J Natl Cancer Inst 22:173-203 (1959).
7. Hill AB. The environment and disease: association or causation? Proc R Soc Med 58:295-300 (1965).
8. Hertz-Picciotto I. Epidemiology and quantitative risk assessment: a bridge from science to policy. Am J Public Health 85:484-491 (1995)
9. Flanders DW, Khoury MJ. Indirect assessment of confounding: graphic description and limits on effect of adjusting for covariates. Epidemiology 1:239-246 (1990).
I would like to address comments of Loomis et al. about inferences drawn from studies using group-level exposure variables, the use of the tobacco analogy, the application of Hill's criteria for causality (1), and the use of the Hertz-Picciotto criteria for evaluating studies (2). Whether the hybrid studies under discussion (3-5) are considered partly ecological (6,7) or individual level with exposure misclassification, bias (ecological or otherwise) is possible and should be checked. I hope that these discussions will lead to more considerations of the interplay between outcomes and confounders measured at the individual level and exposure measured at the group level.
Loomis et al. suggest that the biases stemming from the "ecologic fallacy" do not apply to the PM2.5 air pollution studies because they are individual-level studies where exposure is measured with error. That is, by implication there is one PM exposure variable. But as indicated by Morgenstern (8,9), ecologic bias can arise when the mean of a group-level exposure variable has an effect on the individual-level exposure. By this definition there will be ecologic bias whenever the ecologic exposure variable has an effect, and when there is also an individual-level exposure effect in addition to the ecologic exposure effect. Unmeasured individual-level exposure to PM2.5 from all sources can be several orders of magnitude higher than ambient PM2.5 concentrations (10) because of extensive exposure to unmeasured sources such as tobacco and indoor combustion. These individual-level exposures vary for individuals within the group and contribute to the individual-level risk. The additional effect of ambient exposure provides the group-level component that leads to ecologic bias.
The American Cancer Society (ACS) Study (4) and the Six Cities Study (3) suggest that an increase of about 20 µg/m3 PM2.5 results in a 20-30% increase in total mortality. I sought to test the consistency of these findings by comparing risk estimates based on group-level exposure estimates to those based on individual-level exposure to a similar but more thoroughly studied particulate (i.e., tobacco smoke). Applying the models developed in these studies to tobacco smoke, one can predict that a 20-µg/m3 difference in ambient PM2.5 between cities is too small to result in a measurable difference in overall mortality (6). If this is true, the differences in mortality between cities may be due to causes other than differences in PM. Whether there is ecologic bias, exposure misclassification bias operating at the individual level, or uncontrolled bias from other sources, the tobacco analogy suggests that bias away from the null may be operating in these studies.
Loomis et al. suggest that the tobacco analogy presents "strong evidence of a supralinear dose-response relationship between particles and mortality." In order to fit the data, the degree of supralinearity would have to be enormous. In fact, an increase of 19.6 µg/m3 in ambient PM2.5 and an increase of 16,000 µg/m3 from smoking would have to both result in a similar 20-30% increased risk (Figure 1) It is not plausible that two increases in exposure, which differ by almost three orders of magnitude, would both produce the same response. A more plausible inference is that either the PM2.5 or the smoking risk estimates are in error. However, I would place more credence in the smoking relative risks (RRs) because smoking is measured at the individual rather than the group level, and the smoking RRs are compatible with a large body of literature.
Figure 1. Association of total mortality with group level ambient PM2.5 exposure and individual level tobacco smoke exposure. Data from the Six Cities Study (3).
aApproximately 16,000 µg/m3.
It is not necessarily correct to infer, as Loomis et al. do, that lowering community-wide air pollution below existing levels will reduce community mortality rates. In making this inference, one assumes there is independent evidence for a causal relationship between ambient PM2.5 and mortality. These studies (4,6) showed that there were differences in total mortality, but did not show why mortality was higher in cities with higher PM2.5 concentrations. If PM2.5 is the reason for increased mortality, all important individual risk factors must be taken into account to a reasonable degree. Total mortality has a large number of risk factors. It is speculative therefore to assume, as Loomis et al. do, that lowering PM2.5 concentrations beyond existing levels will provide a "logical and efficient means of minimizing the health impact of a widespread exposure," and that the proposed cure will produce the desired effect. The tobacco analogy provides evidence against such an effect.
Loomis et al. state that the effects of potential confounders are too small to explain the observed associations in the PM studies. In my paper (6) I assessed differences in lung function and sedentary living as two examples of possible confounders because some evidence was available to me. Even in these examples, individual-level data were not available to adequately estimate or adjust for these effects (6). There are undoubtedly many other examples such as personal lifestyle factors or other inadequately controlled variables that are correlated with dirty versus clean cities or with geography. Based on the tobacco analogy, it appears that whatever biases are operating resulted in a large overestimate of the PM2.5 risk.
I agree with Loomis et al. that to apply Hill's (1) criteria appropriately requires a "careful review not of one study design, but rather of the entire body of literature pertaining to the hypothesis of a causal association between [chronic PM] air pollution exposure and health [mortality]." In applying these criteria, I have included experimental studies in animals and epidemiologic studies in humans (6). Rodents exposed to high concentrations of diesel exhaust for life did not show early or increased mortality (6). The available epidemiologic studies (3-5) showed only weak associations across a narrow exposure range. The possible role of PM and lung function in the Six Cities Study suggested that there were differences in lung function between cities, but no measurable effect attributable to PM2.5. In the Seventh Day Adventist Study (5), there appeared to be a coherent relationship between PM and self-reported symptoms, but not between PM and mortality. However, the analyses required to evaluate fully this PM-symptoms association were not reported. Furthermore, short-term exposures and hospital admissions may have little to no relevance for mortality from chronic exposure (6,11). Finally, individual-level measurements of an analogous/surrogate PM2.5 exposure from the same populations and same cities provided a test of the internal consistency and biological coherence of the ambient PM2.5 associations. I know of no other experimental or clinical human studies that can be used as a more appropriate test.
I did not say or mean to imply that all of Hill's (1) criteria must be obeyed before accepting cause and effect. The only criterion that must be met is that exposure must precede the effect (6,12). I agree with Loomis et al. that Hill's (1) guidelines can be "helpful at the margins of epidemiologic interpretations" (as with PM), but also provide a good framework for assessing causality in general. I do not believe that evidence was excluded, as alleged by Loomis et al., although further tests of the plausibility and coherence of these associations with PM may be possible.
Regarding the use of Hertz-Picciotto's criteria (2), my point was to assess whether the EPA was justified in developing quantitative concentration-response information useful in developing an annual PM2.5 standard from these studies. Table 5 in my paper (6) was an attempt to do this; because both studies were of the same design, the criteria were applied to both the design and the two individual studies. I concluded (6) that
none of the Hertz-Picciotto criteria for quantification of risk and setting air quality standards using [these] epidemiology studies are met.
I believe these are useful guidelines and that they do "contribute to a firmer scientific foundation for low-dose risk estimates and the ensuing regulatory actions"(2).
I suggest that the tobacco analogy analysis provides evidence "that a given type of bias did ... occur" and that it did "quantify its direction and magnitude,"as stated by Loomis et al., within the limits of the data available. It was only possible to suggest possible sources of bias. The magnitude of the confounding could not be calculated as suggested by Loomis et al. for several reasons. The univariate distributions were not provided in the reports of these studies. The bivariate distributions based on individual-level data cannot be determined (8) because the exposure variable is based on aggregate data. Sensitivity analyses are valuable, but I suggest that the tobacco analogy provides a useful and fundamentally sound method for what Loomis et al. describe as "quantitative exploration of potential biases."
I suggest that one cannot be sure, with any degree of certainty, that studies using group-level measures of exposure are free from the potential biases afflicting completely ecological designs. This is an area where research using individual-level data is needed to improve the quality of information used to guide regulatory decisions.
John Gamble
Exxon Biomedical Sciences, Inc.
East Millstone, New Jersey
E-mail: jfgambl@fple.erenj.com
References and Notes
1. Hill AB. The environment and disease: association or causation? Proc R Soc Med 58:295-300 (1965).
2. Hertz-Picciotto I. Epidemiology and quantitative risk assessment: a bridge from science to policy. Am J Public Health 85:484-491 (1995).
3. Dockery DW, Pope CA III, Xu, S, Spengler JD, Ware JH, Fay ME, Ferris BG Jr, Speizer FE. An association between air pollution and mortality in six U.S. cities. N Engl J Med 329:1753-1759 (1993).
4. Pope CA III, Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, Heath CW Jr. Particulate air pollution as a predictor of mortality in a prospective study of U.S. adults. Am J Respir Crit Care Med 151:669-674 (1995).
5. Abbey DE, Hwang BL, Burchette RJ, Vancuren T, Mills PK. Estimated long-term ambient concentrations of PM10 and development of respiratory symptoms in a nonsmoking population. Arch Environ Health 50:139-152 (1995).
6. Gamble JF. PM2.5 and mortality in long-term prospective cohort studies: cause-effect or statistical associations? Environ Health Perspect 106:535-549 (1998).
7. Künzli N, Tager IB. Comments on "PM2.5 and mortality in long-term prospective cohort studies: cause-effect or statistical associations?" [letter]. Environ Health Perspect 107: A234-A235 (1999).
8. Morgenstern H. Ecologic studies in modern epidemiology. In: Modern Epidemiology (Rothman K, Greenland S, eds). 2nd ed. Lippincott Raven Publishers, 1998;459-480.
9. Morgenstern H. Uses of ecologic analysis in epidemiologic research. Am J Public Health 72:1336-1344 (1982) .
10. U.S. EPA. Air Quality Criteria for Particulate Matter, EPA-600-AP-95-001C. Washington, DC:U.S. Environmental Protection Agency, 1996.
11. Gamble JF. Reply to Künzli and Tager regarding causality in PM2.5 cohort studies [letter].Environ Health Perspect 107:A234-A235 (1999).
12. Rothman KJ. Modern Epidemiology. 1st ed. Boston, MA:Little, Brown, and Company, 1986.
The Focus article "Double Exposure" [EHP 107:A196-A201 (1999)] is, for the most part, nothing more than the continued complaining of those who don't realize that living is a hazard. They would place a risk on everything and then ignore those they don't like. I would like to challenge Manuel to provide a laboratory analysis from any reliable source [the American Cancer Society (ACS), etc.] that can identify even half of the stated 4,500 components he cites. The ACS stated that 
2,000 components were present in smoke, but when asked to list them, they could not. The California Air Resources Board cites < 50 components at current detectable levels. Only the tars and benzo[a]pyrene have even been proven harmful to rats, and this has been in massive doses. As a retired chemical engineer with an extensive background in chromatography and mass spectrography, I can say with certainty that if 4,500 components exist in tobacco smoke, then the clean air we breathe must contain 6,000 or more. At 66 years of age and a smoker of over two packs of cigarettes a day for 47 years, I was recently rejected for a Veterans' Administration lung health study because my lungs were "too healthy." Three guesses what their study results will show.
William C. Briggs
Chemical Engineer, retired
Concord, California
E-mail: mrbill2b@silcon.com
The figure of more than 4,500 compounds in tobacco smoke was obtained from an article titled "Assessment of Exposure to Environmental Tobacco Smoke" (1). The U.S. Environmental Protection Agency (2) cites a figure of "more than 4,000 compounds" in its 1992 report. A table of many of the constituents is listed on pages 3-5 through 3-9 of that report.
John Manuel
Durham, North Carolina
E-mail: JManuel782@aol.com
References and Notes
1. Jaakola MS, Jaakola JJK. Assessment of exposure to environmental tobacco smoke. Eur Respir J 10:2386 (1997).
2. U.S. EPA. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. Washington, DC:U.S. Environmental Protection Agency, 1992.
The Focus article "Double Exposure" [EHP 107:A196-A201 (1999)] mistakenly reported that a meta-analysis of epidemiology studies examining the relationship between exposure to environmental tobacco smoke (ETS) and human cancer was prepared for the review of ETS in consideration for listing in the National Toxicology Program (NTP) Report on Carcinogens. We would like to point out that the results of several published meta-analyses were considered during the ETS review, but that the NTP did not, in fact, perform an additional meta-analysis of these data.
John R. Bucher
C.W. Jameson
NIEHS
Research Triangle Park, North Carolina
E-mail: bucher@niehs.nih.gov
Work in multiple laboratories on paraoxonase polymorphism has shown that this mammal enzyme protects against chlorpyrifos and other organophosphates differentially, depending on which inherited forms of the enzyme predominate. This variable ability to process pesticides exists in humans and is trackable ethnically (1), and very recent articles have suggested the polymorphism may be responsible for host susceptibility to Gulf War syndrome (2,3). Because my own interest is in multiple chemical sensitivity (MCS) mechanisms, I immediately searched Medline (National Library of Medicine, Bethesda, MD) for publications that included paraoxonase and MCS, fibromyalgia, or chronic fatigue syndrome (CFS), which have been suggested as being overlapping or identical with each other and with Gulf War syndrome (4,5). I found no other references at all.
I believe that there is a significant possibility that paraoxonase is important in these conditions. A meta-model generating six examples of MCS mechanisms that could arise during situations of chemical damage or rapid contextual change has previously been shown (6). The first of these models, and apparently the most likely on the basis of existing clinical evidence, is pesticide damage to the central nervous system, in which exposures to chlorpyrifos and other organophosphates figure prominently in the supporting data. In addition, another of the models explores how host genetic variability in immune response could lead, through conditioning, to the heightened reactions to low levels of other toxicants that is characteristic of MCS, CFS, fibromyalgia (4), and Gulf War syndrome (5). Both of these multiple chemical sensitivity models could be satisfied simultaneously by the genetic variation of an enzyme that controls the ability to resist damage from pesticides--an enzyme such as paraoxonase.
This suggests that it may be worthwhile to test statistically significant sizes of MCS, CFS, fibromyalgia, and Gulf War syndrome patient populations to determine what paraoxonase polymorphisms they carry. If it were determined in any one of these populations that the condition is genetically determined by reduced paraoxonase levels or activity, this would solve the debate about psychogenic or biogenic origins of that condition. Additionally, it would allow research funding to be confidently spent following up the mechanism, i.e., determining exactly how the inability to process these insecticides results in the condition.
A new study correlates significantly lower frequency and activity of PON1-Q paraoxonase polymorphism, which processes several organophosphates including some nerve gases, with a symptomatic group of Gulf War syndrome veterans, as compared to well controls (7). This study has in effect carried out my suggestion for one of the groups, and its result strongly implies the need to do the same for the others: MCS, fibromyalgia, and CFS.
Steven C. Rowat
Grantham's Landing
British Columbia, Canada
E-mail: Steven_Rowat@sunshine.net
References and Notes
1. Diepgen TL, Geldmacher-von Mallinckrodt M. Interethnic differences in the detoxification of organophosphates: the human serum paraoxonase polymorphism. Arch Toxicol Suppl 9:154-158 (1986).
2. Mackness B, Mackness MI, Arrol S, Turkie W, Durrington PN. Effect of the molecular polymorphisms of human paraoxonase (PON1) on the rate of hydrolysis of paraoxon. Br J Pharmacol 122(2):265-268 (1997).
3. Mackness B, Durrington PN, Mackness MI. Human serum paraoxonase. Gen Pharmacol 31(3):329-336 (1998).
4. Ziem G, McTamney J. Profile of patients with chemical injury and sensitivity. Environ Health Perspect 105(suppl 2):417-436 (1997).
5. Miller CS. Chemical sensitivity: symptom, syndrome or mechanism for disease? Toxicology 111:69-86 (1996).
6. Rowat SC. Integrated defense system overlaps as a disease model: with examples for multiple chemical sensitivity. Environ Health Perspect 106(suppl 1):85-109 (1998).
7. Haley RW, Billecke S, La Du BN. Association of low PON1 type Q (type A) arylesterase activity with neurologic symptom complexes in Gulf War veterans. Toxicol Appl Pharmacol 157(3):227-233 (1999).
Last Updated: July 19, 1999