An individualized treatment rule (ITR) is a decision rule that recommends treatments for patients based on their individual feature variables. In many practices, the ideal ITR for the primary outcome is also expected to cause minimal harm to other secondary outcomes. Therefore, our objective is to learn an ITR that not only maximizes the value function for the primary outcome, but also approximates the optimal rule for the secondary outcomes as closely as possible. To achieve this goal, we introduce a fusion penalty to encourage the ITRs based on different outcomes to yield similar recommendations. Two algorithms are proposed to estimate the ITR using surrogate loss functions. We prove that the agreement rate between the estimated ITR of the primary outcome and the optimal ITRs of the secondary outcomes converges to the true agreement rate faster than if the secondary outcomes are not taken into consideration. Furthermore, we derive the non-asymptotic properties of the value function and misclassification rate for the proposed method. Finally, simulation studies and a real data example are used to demonstrate the finite-sample performance of the proposed method.

Dynamic treatment regimens (DTRs) aim at tailoring individualized sequential treatment rules that maximize cumulative beneficial outcomes by accommodating patients' heterogeneity in decision-making. For many chronic diseases including type 2 diabetes mellitus (T2D), treatments are usually multifaceted in the sense that aggressive treatments with a higher expected reward are also likely to elevate the risk of acute adverse events. In this paper, we propose a new weighted learning framework, namely benefit-risk dynamic treatment regimens (BR-DTRs), to address the benefit-risk trade-off. The new framework relies on a backward learning procedure by restricting the induced risk of the treatment rule to be no larger than a pre-specified risk constraint at each treatment stage. Computationally, the estimated treatment rule solves a weighted support vector machine problem with a modified smooth constraint. Theoretically, we show that the proposed DTRs are Fisher consistent, and we further obtain the convergence rates for both the value and risk functions. Finally, the performance of the proposed method is demonstrated via extensive simulation studies and application to a real study for T2D patients.

COVID-19 pandemic has caused unprecedented negative impacts on our society, including further exposing inequity and disparity in public health. To study the impact of socioeconomic factors on COVID transmission, we first propose a spatial-temporal model to examine the socioeconomic heterogeneity and spatial correlation of COVID-19 transmission at the community level. Second, to assess the individual risk of severe COVID-19 outcomes after a positive diagnosis, we propose a dynamic, varying-coefficient model that integrates individual-level risk factors from electronic health records (EHRs) with community-level risk factors. The underlying neighborhood prevalence of infections (both symptomatic and pre-symptomatic) predicted from the previous spatial-temporal model is included in the individual risk assessment so as to better capture the background risk of virus exposure for each individual. We design a weighting scheme to mitigate multiple selection biases inherited in EHRs of COVID patients. We analyze COVID transmission data in New York City (NYC, the epicenter of the first surge in the United States) and EHRs from NYC hospitals, where time-varying effects of community risk factors and significant interactions between individual- and community-level risk factors are detected. By examining the socioeconomic disparity of infection risks and interaction among the risk factors, our methods can assist public health decision-making and facilitate better clinical management of COVID patients.

For mental disorders, patients' underlying mental states are non-observed latent constructs which have to be inferred from observed multi-domain measurements such as diagnostic symptoms and patient functioning scores. Additionally, substantial heterogeneity in the disease diagnosis between patients needs to be addressed for optimizing individualized treatment policy in order to achieve precision medicine. To address these challenges, we propose an integrated learning framework that can simultaneously learn patients' underlying mental states and recommend optimal treatments for each individual. This learning framework is based on the measurement theory in psychiatry for modeling multiple disease diagnostic measures as arising from the underlying causes (true mental states). It allows incorporation of the multivariate pre- and post-treatment outcomes as well as biological measures while preserving the invariant structure for representing patients' latent mental states. A multi-layer neural network is used to allow complex treatment effect heterogeneity. Optimal treatment policy can be inferred for future patients by comparing their potential mental states under different treatments given the observed multi-domain pre-treatment measurements. Experiments on simulated data and a real-world clinical trial data show that the learned treatment polices compare favorably to alternative methods on heterogeneous treatment effects, and have broad utilities which lead to better patient outcomes on multiple domains.

To address substantial heterogeneity in patient response to treatment of chronic disorders and achieve the promise of precision medicine, individualized treatment rules (ITRs) are estimated to tailor treatments according to patient-specific characteristics. Randomized controlled trials (RCTs) provide gold standard data for learning ITRs not subject to confounding bias. However, RCTs are often conducted under stringent inclusion/exclusion criteria, and participants in RCTs may not reflect the general patient population. Thus, ITRs learned from RCTs lack generalizability to the broader real world patient population. Real world databases such as electronic health records (EHRs) provide new resources as complements to RCTs to facilitate evidence-based research for personalized medicine. However, to ensure the validity of ITRs learned from EHRs, a number of challenges including confounding bias and selection bias must be addressed. In this work, we propose a matching-based machine learning method to estimate optimal individualized treatment rules from EHRs using interpretable features extracted from EHR documentation of medications and ICD diagnoses codes. We use a latent Dirichlet allocation (LDA) model to extract latent topics and weights as features for learning ITRs. Our method achieves confounding reduction in observational studies through matching treated and untreated individuals and improves treatment optimization by augmenting feature space with clinically meaningful LDA-based features. We apply the method to EHR data collected at New York Presbyterian Hospital clinical data warehouse in studying optimal second-line treatment for type 2 diabetes (T2D) patients. We use cross validation to show that ITRs outperforms uniform treatment strategies (i.e., assigning same treatment to all individuals), and including topic modeling features leads to more reduction of post-treatment complications.

To achieve personalized medicine, an individualized treatment strategy assigning treatment based on an individual's characteristics that leads to the largest benefit can be considered. Recently, a machine learning approach, O-learning, has been proposed to estimate an optimal individualized treatment rule (ITR), but it is developed to make binary decisions and thus limited to compare two treatments. When many treatment options are available, existing methods need to be adapted by transforming a multiple treatment selection problem into multiple binary treatment selections, for example, via one-vs-one or one-vs-all comparisons. However, combining multiple binary treatment selection rules into a single decision rule requires careful consideration, because it is known in the multicategory learning literature that some approaches may lead to ambiguous decision rules. In this work, we propose a novel and efficient method to generalize outcome-weighted learning for binary treatment to multi-treatment settings. We solve a multiple treatment selection problem via sequential weighted support vector machines. We prove that the resulting ITR is Fisher consistent and obtain the convergence rate of the estimated value function to the true optimal value, i.e., the estimated treatment rule leads to the maximal benefit when the data size goes to infinity. We conduct simulations to demonstrate that the proposed method has superior performance in terms of lower mis-allocation rates and improved expected values. An application to a three-arm randomized trial of major depressive disorder shows that an ITR tailored to individual patient's expectancy of treatment efficacy, their baseline depression severity and other characteristics reduces depressive symptoms more than non-personalized treatment strategies (e.g., treating all patients with combined pharmacotherapy and psychotherapy).

Learning risk scores to predict dichotomous or continuous outcomes using machine learning approaches has been studied extensively. However, how to learn risk scores for time-to-event outcomes subject to right censoring has received little attention until recently. Existing approaches rely on inverse probability weighting or rank-based regression, which may be inefficient. In this paper, we develop a new support vector hazards machine (SVHM) approach to predict censored outcomes. Our method is based on predicting the counting process associated with the time-to-event outcomes among subjects at risk via a series of support vector machines. Introducing counting processes to represent time-to-event data leads to a connection between support vector machines in supervised learning and hazards regression in standard survival analysis. To account for different at risk populations at observed event times, a time-varying offset is used in estimating risk scores. The resulting optimization is a convex quadratic programming problem that can easily incorporate non-linearity using kernel trick. We demonstrate an interesting link from the profiled empirical risk function of SVHM to the Cox partial likelihood. We then formally show that SVHM is optimal in discriminating covariate-specific hazard function from population average hazard function, and establish the consistency and learning rate of the predicted risk using the estimated risk scores. Simulation studies show improved prediction accuracy of the event times using SVHM compared to existing machine learning methods and standard conventional approaches. Finally, we analyze two real world biomedical study data where we use clinical markers and neuroimaging biomarkers to predict age-at- onset of a disease, and demonstrate superiority of SVHM in distinguishing high risk versus low risk subjects.