×

zbMATH — the first resource for mathematics

Interpretable dynamic treatment regimes. (English) Zbl 1409.62231
Summary: Precision medicine is currently a topic of great interest in clinical and intervention science. A key component of precision medicine is that it is evidence-based, that is, data-driven, and consequently there has been tremendous interest in estimation of precision medicine strategies using observational or randomized study data. One way to formalize precision medicine is through a treatment regime, which is a sequence of decision rules, one per stage of clinical intervention, that map up-to-date patient information to a recommended treatment. An optimal treatment regime is defined as maximizing the mean of some cumulative clinical outcome if applied to a population of interest. It is well-known that even under simple generative models an optimal treatment regime can be a highly nonlinear function of patient information. Consequently, a focal point of recent methodological research has been the development of flexible models for estimating optimal treatment regimes. However, in many settings, estimation of an optimal treatment regime is an exploratory analysis intended to generate new hypotheses for subsequent research and not to directly dictate treatment to new patients. In such settings, an estimated treatment regime that is interpretable in a domain context may be of greater value than an unintelligible treatment regime built using “black-box” estimation methods. We propose an estimator of an optimal treatment regime composed of a sequence of decision rules, each expressible as a list of “if-then” statements that can be presented as either a paragraph or as a simple flowchart that is immediately interpretable to domain experts. The discreteness of these lists precludes smooth, that is, gradient-based, methods of estimation and leads to nonstandard asymptotics. Nevertheless, we provide a computationally efficient estimation algorithm, prove consistency of the proposed estimator, and derive rates of convergence. We illustrate the proposed methods using a series of simulation examples and application to data from a sequential clinical trial on bipolar disorder.

MSC:
62P10 Applications of statistics to biology and medical sciences; meta analysis
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Ashley, E. A., The Precision Medicine Initiative: A New National Effort, Journal of the American Statistical Association, 313, 2119-2120, (2015)
[2] Breiman, L.; Friedman, J. H.; Olshen, R. A.; Stone, C. J., Classification and Regression Trees, (1984), New York: CRC Press, New York · Zbl 0541.62042
[3] Collins, F. S.; Varmus, H., A New Initiative on Precision Medicine, New England Journal of Medicine, 372, 793-795, (2015)
[4] Cormen, T. H.; Leiserson, C. E.; Rivest, R. L.; Stein, C., Introduction to Algorithms, (2009), New York: McGraw-Hill · Zbl 1187.68679
[5] Doove, L.; Dusseldorp, E.; Van Deun, K.; Van Mechelen, I., A Novel Method for Estimating Optimal Tree-Based Treatment Regimes in Randomized Clinical Trials, (2015)
[6] Eberts, M.; Steinwart, I., Optimal Regression Rates for SVMs using Gaussian Kernels, Electronic Journal of Statistics, 7, 1-42, (2013) · Zbl 1337.62073
[7] Friedman, J.; Hastie, T.; Tibshirani, R., Regularization Paths for Generalized Linear Models via Coordinate Descent, Journal of Statistical Software, 33, 1-22, (2010)
[8] Jameson, J. L.; Longo, D. L., Precision Medicine Personalized, Problematic, and Promising, New England Journal of Medicine, 372, 2229-2234, (2015)
[9] Kang, C.; Janes, H.; Huang, Y., Combining Biomarkers to Optimize Patient Treatment Recommendations, Biometrics, 70, 695-707, (2014) · Zbl 1299.62125
[10] Kim, D.; Sra, S.; Dhillon, I. S., Tackling Box-Constrained Optimization via a New Projected Quasi-Newton Approach, SIAM Journal on Scientific Computing, 32, 3548-3563, (2010) · Zbl 1220.93085
[11] Kim, J.; Pollard, D., Cube Root Asymptotics, The Annals of Statistics, 18, 191-219, (1990) · Zbl 0703.62063
[12] Kimeldorf, G.; Wahba, G., Some Results on Tchebycheffian Spline Functions, Journal of Mathematical Analysis and Applications, 33, 82-95, (1971) · Zbl 0201.39702
[13] Krumholz, H. M., Big Data and New Knowledge in Medicine: The Thinking, Training, and Tools Needed for a Learning Health System, Health Affairs, 33, 1163-1170, (2014)
[14] Laber, E.; Zhao, Y.; Regh, T.; Davidian, M.; Tsiatis, A. A.; Stanford, J. B.; Zeng, D.; Kosorok, M. R., Sizing a Phase II Trial to Find a Nearly Optimal Personalized Treatment Strategy, Statistics in Medicine, 35, 1245-1256, (2016)
[15] Laber, E. B.; Linn, K. A.; Stefanski, L. A., Interactive Model Building for Q-Learning, Biometrika, 101, 831-847, (2014) · Zbl 1306.62235
[16] Laber, E. B.; Zhao, Y. Q., Tree-Based Methods for Individualized Treatment Regimes, Biometrika, 102, 501-514, (2015) · Zbl 06519989
[17] Lei, H.; Nahum-Shani, I.; Lynch, K.; Oslin, D.; Murphy, S. A., A ‘SMART’ Design for Building Individualized Treatment Sequences, Annual Review of Clinical Psychology, 8, 21-48, (2012)
[18] Letham, B.; Rudin, C.; McCormick, T. H.; Madigan, D., Building Interpretable Classifiers with Rules using Bayesian Analysis, (2012)
[19] Liaw, A.; Wiener, M., Classification and Regression by randomForest, R News, 2, 18-22, (2002)
[20] Marchand, M.; Sokolova, M., Learning with Decision Lists of Data-Dependent Features, Journal of Machine Learning Research, 6, 427-451, (2005) · Zbl 1222.68257
[21] Moodie, E. E. M.; Dean, N.; Sun, Y. R., Q-Learning: Flexible Learning About Useful Utilities, Statistics in Biosciences, 6, 1-21, (2013)
[22] Murphy, S. A., Optimal Dynamic Treatment Regimes, Journal of the Royal Statistical Society: Series B, 65, 331-355, (2003) · Zbl 1065.62006
[23] An experimental Design for the Development of Adaptive Treatment Strategies, Statistics in Medicine, 24, 1455-1481, (2005)
[24] Qian, M.; Murphy, S. A., Performance Guarantees for Individualized Treatment Rules, Annals of Statistics, 39, 1180-1210, (2011) · Zbl 1216.62178
[25] Rivest, R. L., Learning Decision Lists, Machine Learning, 2, 229-246, (1987)
[26] Robins, J. M., Optimal Structural Nested Models for Optimal Sequential Decisions, Proceedings of the Second Seattle Symposium in Biostatistics, 189-326, (2004), New York: Springer, New York · Zbl 1279.62024
[27] Sachs, G. S.; Nierenberg, A. A.; Calabrese, J. R.; Marangell, L. B.; Wisniewski, S. R.; Gyulai, L.; Friedman, E. S.; Bowden, C. L.; Fossey, M. D.; Ostacher, M. J.; Ketter, T. A.; Patel, J.; Hauser, P.; Rapport, D.; Martinez, J. M.; Allen, M. H.; Miklowitz, D. J.; Otto, M. W.; Dennehy, E. B.; Thase, M. E., Effectiveness of Adjunctive Antidepressant Treatment for Bipolar Depression, New England Journal of Medicine, 356, 1711-1722, (2007)
[28] Sachs, G. S.; Thase, M. E.; Otto, M. W.; Bauer, M.; Miklowitz, D.; Wisniewski, S. R.; Lavori, P.; Lebowitz, B.; Rudorfer, M.; Frank, E.; Nierenberg, A. A.; Fava, M.; Bowden, C.; Ketter, T.; Marangell, L.; Calabrese, J.; Kupfer, D.; Rosenbaum, J. F., Rationale, Design, and Methods of the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD), Biological Psychiatry, 53, 1028-1042, (2003)
[29] Schulte, P. J.; Tsiatis, A. A.; Laber, E. B.; Davidian, M., Q- and A-Learning Methods for Estimating Optimal Dynamic Treatment Regimes, Statistical Science, 29, 640-661, (2014) · Zbl 1331.62437
[30] Stone, C. J., Optimal Global Rates of Convergence for Nonparametric Regression, The Annals of Statistics, 10, 1040-1053, (1982) · Zbl 0511.62048
[31] Taylor, J. M. G.; Cheng, W.; Foster, J. C., Reader Reaction to “A Robust Method for Estimating Optimal Treatment Regimes” by Zhang et al. (2012), Biometrics, 71, 267-273, (2015) · Zbl 1329.62438
[32] Wang, F.; Rudin, C., Falling Rule Lists, Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, 1013-1022, (2015)
[33] Xu, Y.; Müller, P.; Wahed, A. S.; Thall, P. F., Bayesian Nonparametric Estimation for Dynamic Treatment Regimes with Sequential Transition Times, Journal of the American Statistical Association, 111, 921-950, (2015)
[34] Xu, Y.; Yu, M.; Zhao, Y.-Q.; Li, Q.; Wang, S.; Shao, J., Regularized Outcome Weighted Subgroup Identification for Differential Treatment Effects, Biometrics, 71, 645-653, (2015) · Zbl 1419.62483
[35] Zhang, B.; Tsiatis, A. A.; Davidian, M.; Zhang, M.; Laber, E., Estimating Optimal Treatment Regimes from a Classification Perspective, Stat, 1, 103-114, (2012)
[36] Zhang, B.; Tsiatis, A. A.; Laber, E. B.; Davidian, M., Robust Estimation of Optimal Dynamic Treatment Regimes for Sequential Treatment Decisions, Biometrika, 100, 681-694, (2013) · Zbl 1284.62508
[37] Zhang, Y.; Laber, E. B.; Tsiatis, A.; Davidian, M., Using Decision Lists to Construct Interpretable and Parsimonious Treatment Regimes, Biometrics, 71, 895-904, (2015) · Zbl 1419.62490
[38] Zhao, Y.; Zeng, D.; Laber, E. B.; Kosorok, M. R., New Statistical Learning Methods for Estimating Optimal Dynamic Treatment Regimes, Journal of the American Statistical Association, 110, 583-598, (2015) · Zbl 1373.62557
[39] Zhao, Y.; Zeng, D.; Rush, A. J.; Kosorok, M. R., Estimating Individualized Treatment Rules Using Outcome Weighted Learning, Journal of the American Statistical Association, 107, 1106-1118, (2012) · Zbl 1443.62396
[40] Zhao, Y.; Zeng, D.; Socinski, M. A.; Kosorok, M. R., Reinforcement Learning Strategies for Clinical Trials in Nonsmall Cell Lung Cancer, Biometrics, 67, 1422-1433, (2011) · Zbl 1274.62922
[41] Zhao, Y. Q.; Zeng, D.; Laber, E. B.; Song, R.; Yuan, M.; Kosorok, M. R., Doubly Robust Learning for Estimating Individualized Treatment with Censored Data, Biometrika, 102, 151-168, (2015) · Zbl 1345.62092
[42] Zhou, X.; Kosorok, M. R., Nearest Neighbor Rules for Optimal Treatment Regimes, (2016)
[43] Zhou, X.; Mayer-Hamblett, N.; Khan, U.; Kosorok, M. R., Residual Weighted Learning for Estimating Individualized Treatment Rules, Journal of the American Statistical Association, 112, 169-187, (2015)
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.