A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation
Tóm tắt
Epidemiologic research is often devoted to etiologic investigation, and so techniques that may facilitate mechanistic inferences are attractive. Some of these techniques rely on rigid and/or unrealistic assumptions, making the biologic inferences tenuous. The methodology investigated here is effect decomposition: the contrast between effect measures estimated with and without adjustment for one or more variables hypothesized to lie on the pathway through which the exposure exerts its effect. This contrast is typically used to distinguish the exposure's indirect effect, through the specified intermediate variables, from its direct effect, transmitted via pathways that do not involve the specified intermediates. We apply a causal framework based on latent potential response types to describe the limitations inherent in effect decomposition analysis. For simplicity, we assume three measured binary variables with monotonic effects and randomized exposure, and use difference contrasts as measures of causal effect. Previous authors showed that confounding between intermediate and the outcome threatens the validity of the decomposition strategy, even if exposure is randomized. We define exchangeability conditions for absence of confounding of causal effects of exposure and intermediate, and generate two example populations in which the no-confounding conditions are satisfied. In one population we impose an additional prohibition against unit-level interaction (synergism). We evaluate the performance of the decomposition strategy against true values of the causal effects, as defined by the proportions of latent potential response types in the two populations. We demonstrate that even when there is no confounding, partition of the total effect into direct and indirect effects is not reliably valid. Decomposition is valid only with the additional restriction that the population contain no units in which exposure and intermediate interact to cause the outcome. This restriction implies homogeneity of causal effects across strata of the intermediate. Reliable effect decomposition requires not only absence of confounding, but also absence of unit-level interaction and use of linear contrasts as measures of causal effect. Epidemiologists should be wary of etiologic inference based on adjusting for intermediates, especially when using ratio effect measures or when absence of interacting potential response types cannot be confidently asserted.
Tài liệu tham khảo
Koopman JS, Weed DL: Epigenesis theory: a mathematical model relating causal concepts of pathogenesis in individuals to disease patterns in populations. Am J Epidemiol 1990, 132:366–390.
Thompson WD: Effect modification and the limits of biological inference from epidemiologic data. J Clin Epidemiol 1991, 44:221–232.
Greenland S, Poole C: Invariants and noninvariants in the concept of interdependent effects. Scand J Work Environ Health 1988, 14:125–129.
Greenland S, Rothman KJ: Concepts of Interaction. Modern Epidemiology 2 Edition (Edited by: Rothman KJ, Greenland S). Philadelphia, Pa.: Lippincott-Raven 1998, 329–342.
MacLehose RF, Kaufman S, Kaufman JS, Poole C: Bounding causal effects under uncontrolled confounding using counterfactuals. Epidemiology, in press.
Susser M: Causal Thinking in the Health Sciences: Concepts and Strategies in Epidemiology Oxford University Press: New York 1973.
Szklo M, Nieto FJ: Epidemiology: Beyond the Basics Aspen Publishers: Gaithersburg, MD 2000.
Heck KE, Pamuk ER: Explaining the relation between education and postmenopausal breast cancer. Am J Epidemiol 1997, 145:366–372.
Lim U, Cassano PA: Homocysteine and blood pressure in the Third National Health and Nutrition Examination Survey, 1988–1994. Am J Epidemiol 2002, 156:1105–1113.
Everson SA, Kauhanen J, Kaplan GA, Goldberg DE, Julkunen J, Tuomilehto J, Salonen JT: Hostility and increased risk of mortality and acute myocardial infarction: the mediating role of behavioral risk factors. Am J Epidemiol 1997, 146:142–152.
Taylor JM, Wang Y, Ahdieh L, Chmiel JS, Detels R, Giorgi JV, Kaslow R, Kingsley L, Margolick J: Causal pathways for CCR5 genotype and HIV progression. J Acquir Immune Defic Syndr 2000, 23:160–171.
Lantz PM, House JS, Lepkowski JM, Williams DR, Mero RP, Chen J: Socioeconomic factors, health behaviors, and mortality: results from a nationally representative prospective study of US adults. JAMA 1998, 279:1703–1708.
Robins JM, Greenland S: Identifiability and exchangeability for direct and indirect effects. Epidemiology 1992, 3:143–155.
Poole C, Kaufman JS: What does the standard adjustment for downstream mediators tell us about social effect pathways? Am J Epidemiol 2000, 151:S52.
Cole SR, Hernan MA: Fallibility in estimating direct effects. Int J Epidemiol 2002, 31:163–165.
Kaufman S, Kaufman JS, MacLehose RF, Greenland S, Poole C: Improved estimation of controlled direct effects in the presence of unmeasured confounding of intermediate variables. Stat Med, in press.
Baron RM, Kenny DA: The moderator-mediator variable distinction in social pychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 1986, 51:1173–1182.
MacKinnon DP: Mediating variable. International Encyclopedia of the Social and Behavioral sciences (Edited by: Smelser NJ, Baltes PB). New York: Elsevier 2002, 9503–9507.
Wright S: The method of path coefficients. Annals of Mathematical Statistics 1934, 5:161–215.
Holland PW: Causal inference, path analysis, and recursive structural equations models. Sociological Methodology (Edited by: Clogg CC). American Sociological Association: Washington, DC 1988, 449–484.
Pearl J: Causality: Models, Reasoning and Inference Cambridge University Press: Cambridge 2000.
Rubin DB: Formal modes of statistical inference for causal effects. J Statist Plann Inference 1990, 25:279–292.
Greenland S, Robins JM: Identifiability, exchangeability, and epidemiological confounding. Int J Epidemiol 1986, 15:433–439.
The Lipid Research Clinics Coronary Primary Prevention Trial results: I. Reduction in incidence of coronary heart disease. JAMA 1984, 251:351–364.
Bollen KA: Total, direct and indirect effects in structural equation models. Sociological Methodology (Edited by: Clogg CC). American Sociological Association: Washington, DC 1987, 37–69.
Stolzenberg RM: The measurement and decomposition of casual effects in nonlinear and nonadditive models. Sociological Methodology (Edited by: Schuessler KF). Jossey-Bass: San Francisco 1980, 459–488.
Terry MB, Neugut AI, Schwartz S, Susser E: Risk factors for a causal intermediate and an endpoint: reconciling differences. Am J Epidemiol 2000, 151:339–345.
Greenland S, Robins JM, Pearl J: Confounding and collapsibility in causal inference. Statistical Science 1999, 14:29–46.
Joffe MM, Colditz GA: Restriction as a method for reducing bias in the estimation of direct effects. Stat Med 1998, 17:2233–2249.
Robins JM: A new approach to causal inference in mortality studies with sustained exposure periods – Application to control of the healthy worker survivor effect. Mathematical Modeling 1986, 7:1393–1512.
Sobel ME: Effect analysis and causation in linear structural equation models. Psychometrika 1990, 55:495–515.
Winship C, Morgan SL: The estimation of causal effects from observational data. Annual Review of Sociology 1999, 25:659–707.
Bollen KA: Structural Equations with Latent Variables John Wiley and Sons: New York, NY 1989.
Robins JM: Semantics of causal DAG models and the identification of direct and indirect effects. Highly Structured Stochastic Systems (Edited by: Green P, Hjort N, Richardson S). Oxford University Press: London 2003, 70–81.
Pearl J: Direct and Indirect Effects. Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence San Francisco, CA: Morgan Kaufmann 2001, 411–420.