Abstract
When interest lies in causal analysis of the effects of multiple exposures on an outcome, one may be interested in investigating the interaction between the exposures. In such settings, causal analysis requires modeling the joint distribution of exposures given pertinent confounding variables. In the most general setting, this may require modeling the effect of confounding variables on the association between exposures via a second-order regression model. We consider joint modeling of exposures for causal analysis via regression adjustment and inverse weighting. In both frameworks, we also investigate the asymptotic bias of estimators when the dependence model for the generalized propensity score incorrectly assumes conditional independence of exposures or is based on a naive dependence model which does not accommodate the effect of confounders on the conditional association of exposures. We also consider the problem of a semi-continuous bivariate exposure and propose a two-stage estimation technique to study the effects of prenatal alcohol exposure, and the effects of drinking frequency and intensity on childhood cognition.
Similar content being viewed by others
References
Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1):41–55
Imbens GW (2000) The role of the propensity score in estimating dose-response functions. Biometrika 87(3):706–710
Hirano K, Imbens GW (2004) The propensity score with continuous treatments. Appl Bayesian Model Causal Inference Incomplete-Data Perspect 226164:73–84
Dominici F, Peng RD, Barr CD, Bell ML (2010) Protecting human health from air pollution: shifting from a single-pollutant to a multi-pollutant approach. Epidemiology 21(2):187
Duell EJ (2012) Epidemiology and potential mechanisms of tobacco smoking and heavy alcohol consumption in pancreatic cancer. Mol Carcinog 51(1):40–52
Suzuki T, Wakai K, Matsuo K, Hirose K, Ito H, Kuriki K, Sato S, Ueda R, Hasegawa Y, Tajima K (2006) Effect of dietary antioxidants and risk of oral, pharyngeal and laryngeal squamous cell carcinoma according to smoking and drinking habits. Cancer Sci 97(8):760–767
Beelen R, Hoek G, van den Brandt PA, Goldbohm RA, Fischer P, Schouten LJ, Armstrong B, Brunekreef B (2008) Long-term exposure to traffic-related air pollution and lung cancer risk. Epidemiology 19:702–710
Vegetabile BG, Gillen DL, Stern HS (2020) Optimally balanced gaussian process propensity scores for estimating treatment effects. J R Stat Soc Ser A 183(1):355–377
Williams JR, Crespi CM (2020a) Causal inference for multiple continuous exposures via the multivariate generalized propensity score. arXiv:2008.13767
Chen J, Zhou Y (2023) Causal effect estimation for multivariate continuous treatments. Biometr J. https://doi.org/10.1002/bimj.202200122
Di Credico G, Edefonti V, Polesel J, Pauli F, Torelli N, Serraino D, Negri E, Luce D, Stucker I, Matsuo K et al (2019) Joint effects of intensity and duration of cigarette smoking on the risk of head and neck cancer: a bivariate spline model approach. Oral Oncol 94:47–57
Jacobson SW, Jacobson JL, Sokol RJ, Chiodo LM, Corobana R (2004) Maternal age, alcohol abuse history, and quality of parenting as moderators of the effects of prenatal alcohol exposure on 7.5-year intellectual function. Alcoholism 28(11):1732–1745
Kodituwakku P (2007) Defining the behavioral phenotype in children with fetal alcohol spectrum disorders: a review. Neurosci Biobehav Rev 31(2):192–201
Lewis CE, Thomas KG, Dodge NC, Molteno CD, Meintjes EM, Jacobson JL, Jacobson SW (2015) Verbal learning and memory impairment in children with fetal alcohol spectrum disorders. Alcoholism 39(4):724–732
Jacobson JL, Dodge NC, Burden MJ, Klorman R, Jacobson SW (2011) Number processing in adolescents with prenatal alcohol exposure and ADHD: differences in the neurobehavioral phenotype. Alcoholism 35(3):431–442
Jacobson JL, Jacobson SW, Sokol RJ, Martier SS, Ager JW, Kaplan-Estrin MG (1993) Teratogenic effects of alcohol on infant development. Alcoholism 17(1):174–183
Imai K, Van Dyk DA (2004) Causal inference with general treatment regimes: generalizing the propensity score. J Am Stat Assoc 99(467):854–866
Lipsitz SR, Laird NM, Harrington DP (1991) Generalized estimating equations for correlated binary data: using the odds ratio as a measure of association. Biometrika 78(1):153–160
Hernán MA, Robins JM (2010) Causal inference
White H (1982) Maximum likelihood estimation of misspecified models. Econometrica 1–25
Rotnitzky A, Wypij D (1994) A note on the bias of estimators with missing data. Biometrics 1163–1170
Yule GU (1912) On the methods of measuring association between two attributes. J R Stat Soc 75(6):579–652
Williams JR, Crespi CM (2020b) Causal inference for multiple continuous exposures via the multivariate generalized propensity score. arXiv:2008.13767
Hocagil TA, Cook RJ, Jacobson SW, Jacobson JL, Ryan LM (2021) Propensity score analysis for a semi-continuous exposure variable: a study of gestational alcohol exposure and childhood cognition. J R Stat Soc 184:1390–1413
Robins JM (2000) Marginal structural models versus structural nested models as tools for causal inference. In: Statistical models in epidemiology, the environment, and clinical trials. Springer, New York, pp 95–133
Zhao LP, Prentice RL (1990) Correlated binary regression using a quadratic exponential model. Biometrika 77(3):642–648
Kalema G, Molenberghs G, Kassahun W (2016) Second-order generalized estimating equations for correlated count data. Comput Stat 31:749–770
Scharfstein DO, Rotnitzky A, Robins JM (1999) Adjusting for nonignorable drop-out using semiparametric nonresponse models. J Am Stat Assoc 94(448):1096–1120
Snowden JM, Rose S, Mortimer KM (2011) Implementation of g-computation on a simulated data set: demonstration of a causal inference technique. Am J Epidemiol 173(7):731–738
Williamson EJ, Forbes A, Wolfe R (2012) Doubly robust estimators of causal exposure effects with missing data in the outcome, exposure or a confounder. Stat Med 31(30):4382–4400
Jacobson SW, Chiodo LM, Sokol RJ, Jacobson JL (2002) Validity of maternal report of prenatal alcohol, cocaine, and smoking in relation to neurobehavioral outcome. Pediatrics 109(5):815–825
Jacobson JL (2022) Effects of prenatal alcohol exposure on cognitive development: a dose-response analysis. Under review
Austin PC (2019) Assessing covariate balance when using the generalized propensity score with quantitative or continuous exposures. Stat Methods Med Res 28(5):1365–1377
Wu X, Mealli F, Kioumourtzoglou MA, Dominici F, Braun D (2022) Matching on generalized propensity scores with continuous exposures. J Am Stat Assoc 1–29
Splawa-Neyman J, Dabrowska DM, Speed TP (1990) On the application of probability theory to agricultural experiments. Essay on principles. Section 9. Stat Sci 465–472
Rubin DB (1974) Estimating causal effects of treatments in randomized and nonrandomized studies. J Educ Psychol 66(5):688
Prentice RL, Zhao LP (1991) Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses. Biometrics 825–839
Acknowledgements
We thank Neil Dodge, Ph.D., for his assistance with data management support: This research was funded by grants to Sandra W. Jacobson and Joseph L. Jacobson from the National Institutes of Health/National Institute on Alcohol Abuse and Alcoholism (NIH/NIAAA; R01-AA025905) and the Lycaki-Young Fund from the State of Michigan. Richard J. Cook was supported by the Natural Sciences and Engineering Research Council of Canada through grants RGPIN 155849 and RGPIN 04207. Louise Ryan was supported by the Australian Research Council Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS) CE140100049. Data collection for the Detroit Longitudinal Study was supported by grants from NIH/NIAAA (R01-AA06966, R01-AA09524, and P50-AA07606) and NIH/National Institute on Drug Abuse (R21- DA021034).
Author information
Authors and Affiliations
Corresponding author
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, K., Akkaya-Hocagil, T., Cook, R.J. et al. Use of Generalized Propensity Scores for Assessing Effects of Multiple Exposures. Stat Biosci (2023). https://doi.org/10.1007/s12561-023-09403-8
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12561-023-09403-8