Multiply robust estimators of causal effects for survival outcomes. (English) Zbl 1496.62040
Summary: Multiply robust estimators of the longitudinal g-formula have recently been proposed to protect against model misspecification better than the standard augmented inverse probability weighted estimator [A. Rotnitzky, J. Robins and L. Babino, “On the multiply robust estimation of the mean of the g-functional”, Preprint, arXiv:1705.08582; A. R. Luedtke et al., “Sequential double robustness in right-censored longitudinal models”, Preprint, arXiv:1705.02459]. These multiply robust estimators ensure consistency if one of the models for the treatment process or outcome process is correctly specified at each time point. We study the multiply robust estimators of Rotnitzky et al. [loc. cit.] in the context of a survival outcome. Specifically, we compare various estimators of the g-formula for survival outcomes in order to (1) understand how the estimators may be related to one another, (2) understand each estimator’s robustness to model misspecification, and (3) construct estimators that can be more efficient than others in certain model misspecification scenarios. We propose a modification of the multiply robust estimators to gain efficiency under misspecification of the outcome model by using calibrated propensity scores over non-calibrated propensity scores at each time point. Theoretical results are confirmed via simulation studies, and a practical comparison of these estimators is conducted through an application to the US Veterans Aging Cohort Study.
MSC:
| 62D20 | Causal inference from observational studies |
| 62F35 | Robustness and adaptive procedures (parametric inference) |
Keywords:
augmented inverse probability weighting; estimating equations; iterative conditional expectation; local efficiency; multiple robustness; causal inferenceSoftware:
nleqslvReferences:
| [1] | Balzer, L., Ahern, J., Galea, S., & van derLaan, M. (2016). Estimating effects with rare outcomes and high dimensional covariates: Knowledge is power. Epidemiologic Methods, 5, 1-18. · Zbl 1416.92151 |
| [2] | Bang, H., & Robins, J. (2005). Doubly robust estimation in missing data and causal inference models. Biometrics, 61, 962-973. · Zbl 1087.62121 |
| [3] | Benkeser, D., Horvath, K., Reback, C. J., Rusow, J., & Hudgens, M. (2020). Design and analysis considerations for a sequentially randomized HIV prevention trial. Statistics in Biosciences, 12, 446-467. |
| [4] | Cai, W., & van derLaan, M. J. (2019). One‐step targeted maximum likelihood estimation for time‐to‐event outcomes. Biometrics, 76, 722-733. · Zbl 1468.62373 |
| [5] | Cao, W., Tsiatis, A. A., & Davidian, M. (2009). Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data. Biometrika, 96, 723-734. · Zbl 1170.62007 |
| [6] | Deville, J., & Särndal, C. (1992). Calibration estimators in survey sampling. Journal of the American Statistical Association, 87, 376-382. · Zbl 0760.62010 |
| [7] | Grabar, S., Kousignian, I., Sobel, A., Le Bras, P., Gasnault, J., Enel, P., Jung, C., Mahamat, A., Lang, J.‐M., & Costagliola, D. (2004). Immunologic and clinical responses to highly active antiretroviral therapy over 50 years of age. results from the French hospital database on HIV. AIDS, 18, 2029-2038. |
| [8] | Han, P. (2016). Intrinsic efficiency and multiple robustness in longitudinal studies with drop‐out. Biometrika, 103, 683-700. · Zbl 1506.62239 |
| [9] | Hasselman, B. (2018). Package ’nleqslv’. |
| [10] | Hernán, M. A., & Robins, J. M. (2020). Causal inference: What if. Chapman & Hall/CRC Press. |
| [11] | HIV‐CAUSAL Collaboration. (2011). When to initiate combined antiretroviral therapy to reduce mortality and aids‐defining illness in HIV‐infected persons in developed countries: An observational study. Annals of Internal Medicine, 154, 509-515. |
| [12] | King, G., & Zeng, L. (2001). Logistic regression in rare events data. Political Analysis, 9, 137-163. |
| [13] | Kreif, N., Tran, L., Grieve, R., De Stavola, B., Tasker, R., & Petersen, M. (2017). Estimating the comparative effectiveness of feeding interventions in the pediatric intensive care unit: A demonstration of longitudinal targeted maximum likelihood estimation. American Journal of Epidemiology, 186, 1370-1379. |
| [14] | Lodi, S., Costagliola, D., Sabin, C., Del, J. A., Logan, R., Abgrall, S., Reiss, P., Jose, S., Blanco, J., & Hernando, V. (2017). Effect of immediate initiation of antiretroviral treatment in HIV‐positive individuals aged 50 years or older. Journal of Acquired Immune Deficiency Syndromes, 76, 311-318. |
| [15] | Lodi, S., Phillips, A., Logan, R., Olson, A., Costagliola, D., Abgrall, S., vanSighem, A., Reiss, P., Miró, J. M., & Ferrer, E. (2015). Comparative effectiveness of immediate antiretroviral therapy versus cd4‐based initiation in HIV‐positive individuals in high‐income countries: Observational cohort study. The Lancet HIV, 2, e335-e343. |
| [16] | Luedtke, A. R., Sofrygin, O., van derLaan, M. J., & Carone, M. (2018). Sequential double robustness in right‐censored longitudinal models. arXiv preprint arXiv:1705.02459. |
| [17] | Molina, J., Rotnitzky, A., Sued, M., & Robins, J. (2017). Multiple robustness in factorized likelihood models. Biometrika, 104, 561-581. · Zbl 07072228 |
| [18] | Neugebauer, R., Schmittdiel, J., & van derLaan, M. (2014). Targeted learning in real‐world comparative effectiveness research with time‐varying interventions. Statistics in Medicine, 33, 2480-2520. |
| [19] | Neugebauer, R., Schmittdiel, J. A., & van derLaan, M. J. (2016). A case study of the impact of data‐adaptive versus model‐based estimation of the propensity scores on causal inferences from three inverse probability weighting estimators. The International Journal of Biostatistics, 12, 131-155. |
| [20] | Papke, L. E., & Wooldridge, J. M. (2008). Panel data methods for fractional response variables with an application to test pass rates. Journal of Econometrics, 145, 121-133. · Zbl 1418.62527 |
| [21] | Peduzzi, P., Concato, J., Kemper, E., Holford, T. R., & Feinstein, A. R. (1996). A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology, 49, 1373-1379. |
| [22] | Petersen, M., Schwab, J., Gruber, S., Blaser, N., Schomaker, M., & van derLaan, M. (2014). Targeted maximum likelihood estimation for dynamic and static longitudinal marginal structural working models. Journal of Causal Inference, 2, 147-185. |
| [23] | Richardson, T. S., & Rotnitzky, A. (2014). Causal etiology of the research of James M. Robins. Statistical Science, 29, 459-484. · Zbl 1331.00045 |
| [24] | Robins, J. M., Rotnitzky, A., & Zhao, L. P. (1994). Estimation of regression coefficients when some regressors are not always observed. Journal of American Statistical Association, 89, 846-866. · Zbl 0815.62043 |
| [25] | Robins, J., Sued, M., Lei‐Gomez, Q., & Rotnitzky, A. (2007). Comment: Performance of double‐robust estimators when “inverse probability” weights are highly variable. Statistical Science, 22, 544-559. · Zbl 1246.62076 |
| [26] | Robins, J. M. (1986). A new approach to causal inference in mortality studies with a sustained exposure period: Application to control of the healthy worker survivor effect. Mathematical Modelling, 7, 1393-1512. · Zbl 0614.62136 |
| [27] | Robins, J. M. (1997). Causal inference from complex longitudinal data. In M.Berkane (ed.) (Ed.), Latent variable modeling and applications to causality (pp. 69-117). Springer. · Zbl 0969.62072 |
| [28] | Robins, J. M. (2000) Robust estimation in sequentially ignorable missing data and causal inference models. In Proceedings of the American statistical association Section on Bayesian Statistical Science (pp. 6‐10, Annual meeting of the American Statistical Association). The Association, 2000. |
| [29] | Robins, J. M., Hernan, M. A., & Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11, 550-560. |
| [30] | Robins, J. M., & Rotnitzky, A. (2001). Comment on the Bickel and Kwon article, “inference for semiparametric models: Some questions and an answer”. Statistica Sinica, 11, 920-936. |
| [31] | Rotnitzky, A., Lei, Q., Sued, M., & Robins, J. M. (2012). Improved double‐robust estimation in missing data and causal inference models. Biometrika, 99, 439-456. · Zbl 1239.62071 |
| [32] | Rotnitzky, A., Robins, J., & Babino, L. (2017). On the multiply robust estimation of the mean of the g‐functional. arXiv preprint arXiv:1705.08582. |
| [33] | Rotnitzky, A., & Robins, J. M. (1995). Semiparametric efficiency in multivariate regression models with missing data. Biometrika, 90, 122-129. · Zbl 0818.62043 |
| [34] | Sabin, C., Smith, C., d Arminio Monforte, A., Battegay, M., Gabiano, C., Galli, L., Geelen, S., Gibb, D., Guiguet, M., & Judd, A. (2008). Collaboration of observational HIV epidemiological research Europe (COHERE) study group: Response to combination antiretroviral therapy: Variation by age. AIDS, 22, 1463-1473. |
| [35] | Scharfstein, D., Rotnitzky, A., & Robins, J. (1999). Adjusting for nonignorable drop‐out using semiparametric nonresponse models. Journal of the American Statistical Association, 94, 1096-1120. · Zbl 1072.62644 |
| [36] | Schnitzer, M. E., Lok, J. J., & Bosch, R. J. (2016). Double robust and efficient estimation of a prognostic model for events in the presence of dependent censoring. Biostatistics, 17, 165-177. |
| [37] | Schnitzer, M. E., Moodie, E., van derLaan, M., Platt, R., & Klein, M. (2014). Modeling the impact of hepatitis c viral clearance on end‐stage liver disease in an HIV co‐infected cohort with targeted maximum likelihood estimation. Biometrics, 70, 144-152. · Zbl 1419.62435 |
| [38] | Sofrygin, O., Zhu, Z., Schmittdiel, J. A., Adams, A., Grant, R., van derLaan, M., & Neugebauer, R. (2017). Targeted learning with daily ehr data. arXiv:1705.09874. |
| [39] | Tan, Z. (2006). A distributional approach for causal inference using propensity scores. Journal of the American Statistical Association, 101, 1619-1637. · Zbl 1171.62320 |
| [40] | Tan, Z. (2010). Bounded, efficient and doubly robust estimation with inverse weighting. Biometrika, 97, 661-682. · Zbl 1195.62037 |
| [41] | Tan, Z. (2020). Regularized calibrated estimation of propensity scores with model misspecification and high‐dimensional data. Biometrika, 107, 137-158. · Zbl 1435.62274 |
| [42] | Tchetgen Tchetgen, E. (2009). A commentary on G. Molenberghs’s review of missing data methods. Drug Information Journal, 43, 433-435. |
| [43] | Tran, L., Yiannoutsos, C., Wools‐Kaloustian, K., Siika, A., van derLaan, M., & Petersen, M. (2019). Double robust efficient estimators of longitudinal treatment effects: Comparative performance in simulations and a case study. The International Journal of Biostatistics, 15, 1-27. |
| [44] | Tsiatis, A. (2006). Semiparametric theory and missing data (1st ed.). Springer. · Zbl 1105.62002 |
| [45] | van derLaan, M. J., & Gruber, S. (2011). Targeted minimum loss based estimation of an intervention specific mean outcome. |
| [46] | Van der Vaart, A. W. (2000). Asymptotic statistics (Vol. 3). Cambridge University Press. · Zbl 0910.62001 |
| [47] | Vansteelandt, S., Bekaert, M., & Claeskens, G. (2012). On model selection and model misspecification in causal inference. Statistical Methods in Medical Research, 21, 7-30. · Zbl 1365.62431 |
| [48] | Vermeulen, K., & Vansteelandt, S. (2015). Bias‐reduced doubly robust estimation. Journal of the American Statistical Association, 110, 1024-1036. · Zbl 1373.62218 |
| [49] | Wang, L., Rotnitzky, A., Lin, X., Millikan, R. E., & Thall, P. F. (2012). Evaluation of viable dynamic treatment regimes in a sequentially randomized trial of advanced prostate cancer. Journal of the American Statistical Association, 107, 493-508. · Zbl 1328.62601 |
| [50] | Wen, L., Robins, J., Young, J., & Hernán, M. (2021). Parametric g‐formula implementations for causal survival analyses. Biometrics, 77, 740-753. · Zbl 1520.62368 |
| [51] | Young, J. G., Cain, L. E., Robins, J. M., O’Reilly, E. J., & Hernán, M. A. (2011). Comparative effectiveness of dynamic treatment regimes: An application of the parametric g‐formula. Statistics in Biosciences, 3, 119-143. |
| [52] | Young, J. G., Hernán, M. A., & Robins, J. M. (2014). Identification, estimation and approximation of risk under interventions that depend on the natural value of treatment using observational data. Epidemiologic Methods, 3, 1-19. · Zbl 1343.92252 |
| [53] | Young, J. G., Vatsa, R., Murray, E. J., & Hernán, M. A. (2019). Interval‐cohort designs and bias in the estimation of per‐protocol effects: A simulation study. Trials, 20, 1-9. |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
