This paper investigates different approaches for causal estimation under multiple concurrent medications. Our parameter of interest is the marginal mean counterfactual outcome under different combinations of medications. We explore parametric and non-parametric methods to estimate the generalized propensity score. We then apply three causal estimation approaches (inverse probability of treatment weighting, propensity score adjustment, and targeted maximum likelihood estimation) to estimate the causal parameter of interest. Focusing on the estimation of the expected outcome under the most prevalent regimens, we compare the results obtained using these methods in a simulation study with four potentially concurrent medications. We perform a second simulation study in which some combinations of medications may occur rarely or not occur at all in the dataset. Finally, we apply the methods explored to contrast the probability of patient treatment success for the most prevalent regimens of antimicrobial agents for patients with multidrug-resistant pulmonary tuberculosis.
World Health Organization. WHO treatment guidelines for drug-resistant tuberculosis 2016 update, Geneva: World Health Organization, 2016.
5.
AhujaSDAshkinDAvendanoM, et al.Multidrug resistant pulmonary tuberculosis treatment regimens and patient outcomes: an individual patient data meta-analysis of 9,153 patients. PLoS Med2012; 9(9): e1001300.
6.
ImbensGW. The role of the propensity score in estimating dose-response functions. Biometrika2000; 87(3): 706–710.
7.
ImaiKVan DykDA. Causal inference with general treatment regimes: generalizing the propensity score. J Am Stat Assoc2004; 99(467): 854–866.
8.
Lopez MJ and Gutman R. Estimation of causal effects with multiple treatments: a review and new ideas. Statistical Science32(3), 432–454.
9.
McCaffreyDFGriffinBAAlmirallD, et al.A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med2013; 32(19): 3388–3414.
10.
HorvitzDGThompsonDJ. A generalization of sampling without replacement from a finite universe. J Am Stat Assoc1952; 47(260): 663–685.
11.
VansteelandtSDanielRM. On regression adjustment for the propensity score. Stat Med2014; 33(23): 4053–4072.
12.
ScharfsteinDORotnitzkyARobinsJM. Adjusting for nonignorable dropout using semiparametric nonresponse models (with discussion and rejoinder). J Am Stat Assoc1999; 94(448): 1121–1146.
13.
Van der LaanMJRubinD. Targeted maximum likelihood learning. Int J Biostat2006; 2: 1557–4679.
14.
RobinsJ. A new approach to causal inference in mortality studies with a sustained exposure periodapplication to control of the healthy worker survivor effect. Math Model1986; 7(9–12): 1393–1512.
15.
RosenbaumPRRubinDB. The central role of the propensity score in observational studies for causal effects. Biometrika1983; 70(1): 41–55.
16.
HastieTTibshiraniRFriedmanJ. The elements of statistical learning, 2nd ed. New York, NY: Springer, 2009.
17.
AhujaYYadavSK. Multiclass classification and support vector machine. Global J Comput Sci Technol2012; 12(11): 14–20.
HsuCWLinCJ. A comparison of methods for multiclass support vector machines. IEEE Transact Neural Network2002; 13(2): 415–425.
21.
KaratzoglouAMeyerDHornikK. Support vector machines in R. J Stat Software2006; 15(9): 1–28.
22.
WuTFLinCJWengRC. Probability estimates for multi-class classification by pairwise coupling. J Mach Learn Res2004; 5: 975–1005.
23.
Hosmer JrDWLemeshowSSturdivantRX. Applied logistic regression. Wiley Series in probability and statistics2013; Vol. 3, Hoboken, New Jersey: John Wiley & Sons.
Ridgeway G, McCaffrey D, Morral A, et al. Twang: Toolkit for weighting and analysis of nonequivalent groups. R package version 1.0–1. 2006. https://CRAN.R-project.org/package=twang.
26.
SnowdenJMRoseSMortimerKM. Implementation of G-computation on a simulated data set: demonstration of a causal inference technique. Am J Epidemiol2011; 173(7): 731–738.
27.
ZeileisA. Object-oriented computation of sandwich estimators. J Stat Software2006; 16(6): 1–16.
28.
Van der LaanMJRoseS. Targeted learning: causal inference for observational and experimental data, New York: Springer-Verlag, 2011.
29.
TsiatisA. Semiparametric theory and missing data, New York: Springer-Verlag, 2007.
30.
Kennedy EH. Semiparametric theory and empirical processes in causal inference. Technical report, Cornell University Library, https://arxiv.org/abs/1510.04740 (2016).
31.
FriedmanJHastieTTibshiraniR. Regularization paths for generalized linear models via coordinate descent. J Stat Software2010; 33: 1.
32.
ColeSRHernánMA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol2008; 168(6): 656–664.
33.
World Health Organization. Guidelines for the programmatic management of drug-resistant tuberculosis, Geneva: World Health Organization, 2006.
34.
MunirSMahmoodNShahidS, et al.Molecular detection of isoniazid, rifampin and ethambutol resistance to M. tuberculosis and M. bovis in multidrug resistant tuberculosis (MDR-TB) patients in Pakistan. Microbial Pathogenesis2017; 110: 262–274.
35.
AllanaSShashkinaEMathemaB, et al.pncA gene mutations associated with pyrazinamide resistance in drug-resistant tuberculosis, South Africa and Georgia. Emerg Infect Dis2017; 23(3): 491–495.
36.
US Department of Health and Human Services. A profile of older Americans: 2016. Technical report, 2016.
37.
QatoDMWilderJSchummLP, et al.Changes in prescription and over-the-counter medication and dietary supplement use among older adults in the United States, 2005 vs 2011. JAMA Intern Med2016; 176(4): 473–482.
38.
ChaoCTTsaiHBWuCY, et al.Cumulative cardiovascular polypharmacy is associated with the risk of acute kidney injury in elderly patients. Medicine2015; 94(31): e1251.
39.
BenkeserDCaroneMVan der LaanMJ, et al.Doubly robust nonparametric inference on the average treatment effect. Biometrika2017; 104(4): 863–880.
40.
GruberSvan der LaanMJ. C-TMLE of an additive point treatment effect, New York, NY: Springer New York, 2011, pp. 301–321.
41.
SchnitzerMEVan der LaanMJMoodieEEM, et al.Effect of breastfeeding on gastrointestinal infection in infants: a targeted maximum likelihood approach for clustered longitudinal data. Ann Appl Stat2014; 8(2): 703–725.
42.
Ju C, Schwab J and van der Laan MJ. On adaptive propensity score truncation in causal inference. Stat Meth Med Res 2018 (in press).
43.
Ip S and Xue J. A multivariate regression view of multi-label classification. Technical report, University College London, http://www.ucl.ac.uk/zcapg66/CSML/report.pdf (accessed 17 July, 2017).
44.
Cheng W, Hüllermeier E and Dembczynski KJ. Bayes optimal multilabel classification via probabilistic classifier chains. In: Proceedings of the 27th international conference on machine learning (ICML-10), Haifa, Israel, 21–24 June 2010, pp.279–286.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
