Search

Observational data have been actively used to estimate treatment effect, driven by the growing availability of electronic health records (EHRs). However, EHRs typically consist of longitudinal records, often introducing time-dependent confoundings that hinder the unbiased estimation of treatment effect. Inverse probability of treatment weighting (IPTW) is a widely used propensity score method since it provides unbiased treatment effect estimation and its derivation is straightforward. In this study, we aim to utilize IPTW to estimate treatment effect in the presence of time-dependent confounding using claims records. Previous studies have utilized propensity score methods with features derived from claims records through feature processing, which generally requires domain knowledge and additional resources to extract information to accurately estimate propensity scores. Deep sequence models, particularly recurrent neural networks and self-attention-based architectures, have demonstrated good performance in modeling EHRs for various downstream tasks. We propose that these deep sequence models can provide accurate IPTW estimation of treatment effect by directly estimating the propensity scores from claims records without the need for feature processing. We empirically demonstrate this by conducting comprehensive evaluations using synthetic and semi-synthetic datasets.

Tailoring treatment for individual patients is crucial yet challenging in order to achieve optimal healthcare outcomes. Recent advances in reinforcement learning offer promising personalized treatment recommendations; however, they rely solely on current patient observations (vital signs, demographics) as the patient's state, which may not accurately represent the true health status of the patient. This limitation hampers policy learning and evaluation, ultimately limiting treatment effectiveness. In this study, we propose the Deep Attention Q-Network for personalized treatment recommendations, utilizing the Transformer architecture within a deep reinforcement learning framework to efficiently incorporate all past patient observations. We evaluated the model on real-world sepsis and acute hypotension cohorts, demonstrating its superiority to state-of-the-art models. The source code for our model is available at https://github.com/stevenmsm/RL-ICU-DAQN.

This paper is about how to partition decision variables while decomposing a large-scale optimization problem for the best performance of distributed solution methods. Solving a large-scale optimization problem sequen- tially can be computationally challenging. One classic approach is to decompose the problem into smaller sub-problems and solve them in a distributed fashion. However, there is little discussion in the literature on which variables should be grouped together to form the sub-problems, especially when the optimization formulation involves complex constraints. We focus on one of the most popular distributed approaches, dual decomposition and distributed sub-gradient methods. Based on a theoretical guarantee on its convergence rate, we explain that a partition of variables can critically affect the speed of convergence and highlight the importance of the number of dualized constraints. Then, we introduce a novel approach to find a partition that reduces the number of dualized constraints by utilizing a community detection algorithm from physics literature. Roughly speaking, the proposed method groups decision variables that appear together in con- straints and solves the resulting sub-problems with blocks of variables in parallel. Empirical experiments on a real application show that the proposed method significantly accelerates the convergence of the distributed sub-gradient method. The advantage of our approach becomes more significant as the size of the problem increases and each constraint involves more variables.

The objective of this study is to introduce methodology for studying longitudinal claims data observed at the patient level, with inference on the heterogeneity of healthcare utilization behaviors within large healthcare systems such as Medicaid. The proposed approach is model-based, allowing for visualization of longitudinal utilization behaviors using simple stochastic graphical networks. The approach is general, providing a framework for the study of other chronic conditions wherever longitudinal healthcare utilization data are available. Our methods are inspired by and applied to patient-level Medicaid claims for asthma-diagnosed children diagnosed observed over a period of five years, with a comparison of two neighboring states, Georgia and North Carolina.

Rejoinder of "Spatial accessibility of pediatric primary healthcare: Measurement and inference" by Mallory Nobles, Nicoleta Serban and Julie Swann [arXiv:1501.03626]., Comment: Published in at http://dx.doi.org/10.1214/14-AOAS728R the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Although improving financial access is in the spotlight of the current U.S. health policy agenda, this alone does not address universal and comprehensive healthcare. Affordability is one barrier to healthcare, but others such as availability and accessibility, together defined as spatial accessibility, are equally important. In this paper, we develop a measurement and modeling framework that can be used to infer the impact of policy changes on disparities in spatial accessibility within and across different population groups. The underlying model for measuring spatial accessibility is optimization-based and accounts for constraints in the healthcare delivery system. The measurement method is complemented by statistical modeling and inference on the impact of various potential contributing factors to disparities in spatial accessibility. The emphasis of this study is on children's accessibility to primary care pediatricians, piloted for the state of Georgia. We focus on disparities in accessibility between and within two populations: children insured by Medicaid and other children. We find that disparities in spatial accessibility to pediatric primary care in Georgia are significant, and resistant to many policy interventions, suggesting the need for major changes to the structure of Georgia's pediatric healthcare provider network., Comment: Published in at http://dx.doi.org/10.1214/14-AOAS728 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Degradation modeling has traditionally relied on historical signals to estimate the behavior of the underlying degradation process. Many models assume that these historical signals are acquired under the same environmental conditions and can be observed along the entire lifespan of a component. In this paper, we relax these assumptions and present a more general statistical framework for modeling degradation signals that may have been collected under different types of environmental conditions. In addition, we consider applications where the historical signals are not necessarily observed continuously, that is, historical signals are sparse or fragmented. We consider the case where historical degradation signals are collected under known environmental states and another case where the environmental conditions are unknown during the acquisition of these historical data. For the first case, we use a classification algorithm to identify the environmental state of the units operating in the field. In the second case, a clustering step is required for clustering the historical degradation signals. The proposed model can provide accurate predictions of the lifetime or residual life distributions of engineering components that are still operated in the field. This is demonstrated by using simulated degradation signals as well as vibration-based degradation signals acquired from a rotating machinery setup., Comment: Published in at http://dx.doi.org/10.1214/14-AOAS749 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Research in examining the equity of service accessibility has emerged as economic and social equity advocates recognized that where people live influences their opportunities for economic development, access to quality health care and political participation. In this research paper service accessibility equity is concerned with where and when services have been and are accessed by different groups of people, identified by location or underlying socioeconomic variables. Using new statistical methods for modeling spatial-temporal data, this paper estimates demographic association patterns to financial service accessibility varying over a large geographic area (Georgia) and over a period of 13 years. The underlying model is a space--time varying coefficient model including both separable space and time varying coefficients and space--time interaction terms. The model is extended to a multilevel response where the varying coefficients account for both the within- and between-variability. We introduce an inference procedure for assessing the shape of the varying regression coefficients using confidence bands., Comment: Published in at http://dx.doi.org/10.1214/11-AOAS473 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Sensor-based degradation signals measure the accumulation of damage of an engineering system using sensor technology. Degradation signals can be used to estimate, for example, the distribution of the remaining life of partially degraded systems and/or their components. In this paper we present a nonparametric degradation modeling framework for making inference on the evolution of degradation signals that are observed sparsely or over short intervals of times. Furthermore, an empirical Bayes approach is used to update the stochastic parameters of the degradation model in real-time using training degradation signals for online monitoring of components operating in the field. The primary application of this Bayesian framework is updating the residual lifetime up to a degradation threshold of partially degraded components. We validate our degradation modeling approach using a real-world crack growth data set as well as a case study of simulated degradation signals., Comment: Published in at http://dx.doi.org/10.1214/10-AOAS448 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org)