Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter November 19, 2013

Assessing the Causal Effect of Policies: An Example Using Stochastic Interventions

  • Iván Díaz EMAIL logo and Mark J. van der Laan

Abstract

Assessing the causal effect of an exposure often involves the definition of counterfactual outcomes in a hypothetical world in which the stochastic nature of the exposure is modified. Although stochastic interventions are a powerful tool to measure the causal effect of a realistic intervention that intends to alter the population distribution of an exposure, their importance to answer questions about plausible policy interventions has been obscured by the generalized use of deterministic interventions. In this article, we follow the approach described in Díaz and van der Laan (2012) to define and estimate the effect of an intervention that is expected to cause a truncation in the population distribution of the exposure. The observed data parameter that identifies the causal parameter of interest is established, as well as its efficient influence function under the non-parametric model. Inverse probability of treatment weighted (IPTW), augmented IPTW and targeted minimum loss-based estimators (TMLE) are proposed, their consistency and efficiency properties are determined. An extension to longitudinal data structures is presented and its use is demonstrated with a real data example.

References

1. PearlJ. Causality: models, reasoning, and inference. Cambridge: Cambridge University Press, 2000.Search in Google Scholar

2. PearlJ. Causal inference in statistics: an overview. Stat Surv2009;350:96146.10.1214/09-SS057Search in Google Scholar

3. RubinD. Bayesian inference for causal effects: the role of randomization. Ann Stat1978;6:3458.10.1214/aos/1176344064Search in Google Scholar

4. RubinDB. Estimating causal effects of treatments in randomized and nonrandomized studies. J Educ Psychol1974;66(5):688. Available at: http://www.eric.ed.gov/ERICWebPortal/detail?accno=EJ118470.10.1037/h0037350Search in Google Scholar

5. KorbK, HopeL, NicholsonA, AxnickK. Varieties of causal intervention. In: ZhangC, GuesgenHW, YeapW-K, editors. PRICAI 2004: trends in artificial intelligence, Lecture notes in computer science, vol. 3157. Berlin/Heidelberg: Springer, 2004:32231.Search in Google Scholar

6. McAlisterAL. Population behavior change: a theory-based approach. J Public Health Policy1991;12:34561. Available at: http://www.jstor.org/stable/3342846.10.2307/3342846Search in Google Scholar

7. NeugebauerR, van der LaanM. Nonparametric causal effects based on marginal structural models. J Stat Plann Inference2007;137:41934. Available at: http://www.sciencedirect.com/science/article/pii/S0378375806000334.10.1016/j.jspi.2005.12.008Search in Google Scholar

8. DidelezV, DawidAP, GenelettiS. Direct and indirect effects of sequential treatments. In UAI, AUAI Press, 2006.Search in Google Scholar

9. DawidAP, DidelezV. Identifying the consequences of dynamic treatment strategies: a decision-theoretic overview. CoRR, abs/2010;abs/1010.3425.10.1214/10-SS081Search in Google Scholar

10. TianJ. Identifying dynamic sequential plans. In Proceedings of the Twenty-Fourth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-08). Corvallis, OR: AUAI Press, 2008:55461.Search in Google Scholar

11. HafemanDM, VanderWeeleTJ. Alternative assumptions for the identification of direct and indirect effects. Epidemiology2011;22(6):75364.10.1097/EDE.0b013e3181c311b2Search in Google Scholar PubMed

12. PearlJ. Direct and indirect effects. In Proceedings of the 17th Conference in Uncertainty in Artificial Intelligence, UAI ‘01, San Francisco, CA: Morgan Kaufmann Publishers, 2001:41120. Available at: http://dl.acm.org/citation.cfm?id=647235.720084.Search in Google Scholar

13. RobinsJ, GreenlandS. Identifiability and exchangeability for direct and indirect effects. Epidemiology1992;3:14355.10.1097/00001648-199203000-00013Search in Google Scholar PubMed

14. ZhengW, van der LaanMJ. Targeted maximum likelihood estimation of natural direct effect. Working Paper 288 http://www.bepress.com/ucbbiostat/paper288, U.C. Berkeley Division of Biostatistics Working Paper Series, 2011.Search in Google Scholar

15. RobinsJ, RichardsonT. Alternative graphical causal models and the identification of direct effects. Working Paper 100, Harvard School of Public Health, 2010.10.1093/oso/9780199754649.003.0011Search in Google Scholar

16. TaubmanSL, RobinsJM, MittlemanMA, HernánMA. Intervening on risk factors for coronary heart disease: an application of the parametric g-formula. Int J Epidemiol2009;38:1599611. Available at: http://ije.oxfordjournals.org/content/38/6/1599.abstract.10.1093/ije/dyp192Search in Google Scholar PubMed PubMed Central

17. StitelmanOM, HubbardAE, JewellNP. The impact of coarsening the explanatory variable of interest in making causal inferences: implicit assumptions behind dichotomizing variables, 2010. Available at: http://www.bepress.com/ucbbiostat/paper264.Search in Google Scholar

18. CainLE, RobinsJM, LanoyE, LoganR, CostagliolaD, HernánMA. When to start treatment? A systematic approach to the comparison of dynamic regimes using observational data. Int J Biostat2010;6. Available at: http://www.bepress.com/ijb/vol6/iss2/18.10.2202/1557-4679.1212Search in Google Scholar

19. YoungJ, CainL, RobinsJ, O’ReillyE, HernánM. Comparative effectiveness of dynamic treatment regimes: an application of the parametric g-formula. Stat Biosci2011;3:11943. DOI:10.1007/s12561-011-9040-7. Available at: http://dx.doi.org/10.1007/s12561-011-9040-7.10.1007/s12561-011-9040-7Search in Google Scholar

20. DíazI, van der LaanM. Population intervention causal effects based on stochastic interventions. Biometrics2012;68:5419.10.1111/j.1541-0420.2011.01685.xSearch in Google Scholar

21. BickelP, KlaassenC, RitovY, WellnerJ. Efficient and adaptive estimation for semiparametric models. Baltimore, MD, USA: Springer-Verlag, 1997.Search in Google Scholar

22. van der LaanM, RobinsJ. Unified methods for censored longitudinal data and causality. New York: Springer, 2003.10.1007/978-0-387-21700-0Search in Google Scholar

23. RoseS, van der LaanM. Targeted learning: causal inference for observational and experimental data. New York: Springer, 2011.Search in Google Scholar

24. TsiatisA. Information based monitoring of clinical trials. Stat Med2006.10.1002/sim.2625Search in Google Scholar

25. RobinsJ. A new approach to causal inference in mortality studies with sustained exposure periods – application to control of the healthy worker survivor effect. Math Model1986;7(9):1393512.10.1016/0270-0255(86)90088-6Search in Google Scholar

26. PetersenML, PorterKE, GruberS, WangY, van der LaanMJ. Diagnosing and responding to violations in the positivity assumption. Stat Methods Med Res2010;21(1):3154.10.1177/0962280210386207Search in Google Scholar PubMed PubMed Central

27. BrotmanRM, KlebanoffMA, NanselTR, AndrewsWW, SchwebkeJR, ZhangJ, et al. A longitudinal study of vaginal douching and bacterial vaginosis marginal structural modeling analysis. Am J Epidemiol2008;168:18896. Available at: http://aje.oxfordjournals.org/content/168/2/188.abstract.10.1093/aje/kwn103Search in Google Scholar PubMed PubMed Central

28. BryanJ, YuZ, van der LaanMJ. Analysis of longitudinal marginal structural models. Biostatistics2004;5:36180. Available at: http://biostatistics.oxfordjournals.org/content/5/3/361.abstract.10.1093/biostatistics/kxg041Search in Google Scholar

29. JoffeMM, HaveTR, FeldmanHI, KimmelSE. Model selection, confounder control, and marginal structural models: Review and new applications. Am Stat2004;58:2729. Available at: http://www.jstor.org/stable/27643582.10.1198/000313004X5824Search in Google Scholar

30. TagerIB, HaightT, SternfeldB, YuZ, van der LaanM. Effects of physical activity and body composition on functional limitation in the elderly: application of the marginal structural model. Epidemiology2004;15:47993. Available at: http://www.jstor.org/stable/20485932.10.1097/01.ede.0000128401.55545.c6Search in Google Scholar PubMed

31. PorterKE, GruberS, van der LaanMJ, SekhonJS. The relative performance of targeted maximum likelihood estimators. Int J Biostat2011;7:134.10.2202/1557-4679.1308Search in Google Scholar PubMed PubMed Central

32. KangJ, SchaferJ. Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data (with discussion). Stat Sci2007;22:52339.10.1214/07-STS227Search in Google Scholar PubMed PubMed Central

33. van der LaanM, PolleyE, HubbardA. Super learner. Stat Appl Genet Mol Biol2007;6(1):121.10.2202/1544-6115.1309Search in Google Scholar PubMed

34. van der LaanM, DudoitS, KelesS. Asymptotic optimality of likelihood-based cross-validation. Stat Appl Genet Mol Biol2004;3(1):125.10.2202/1544-6115.1036Search in Google Scholar PubMed

35. van der LaanM, DudoitS. Unified cross-validation methodology for selection among estimators and a general cross-validated adaptive epsilon-net estimator: finite sample oracle inequalities and examples. Technical report, Division of Biostatistics, University of California, Berkeley, CA, 2003.Search in Google Scholar

36. van der LaanM, RubinD. Targeted maximum likelihood learning. Int J Biostat2006;2(1):140.10.2202/1557-4679.1043Search in Google Scholar

37. MannJK, BalmesJR, BrucknerTA, MortimerKM, MargolisHG, PrattB, et al. Short-term effects of air pollution on wheeze in asthmatic children in Fresno, CA. Environ Health Perspect2010;118(10):14971502. Available at: http://dx.doi.org/10.1289/ehp.0901292.10.1289/ehp.0901292Search in Google Scholar PubMed PubMed Central

Published Online: 2013-11-19

©2013 by Walter de Gruyter Berlin / Boston

Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/ijb-2013-0014/html
Scroll to top button