The structure of optimal statin initiation policies for patients with Type 2 diabetes

Kurt, Murat; Denton, Brian T.; Schaefer, Andrew J.; Shah, Nilay D.; Smith, Steven A.

doi:10.1080/19488300.2010.550180

Cited by 37 publications

(14 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…, T }. The model is based primarily on the work Hsih [2010] (with a few extensions) and also exhibits similarities to the ones described in Kurt et al [2011], Mason et al [2012], and Mason et al [2014]. Because diabetes is a chronic disease, its patients require long-term, personalized treatment plans that take into account a variety of factors, such as measures of blood glucose, negative side effects, and medical costs.…”

Section: Numerical Resultsmentioning

confidence: 99%

See 1 more Smart Citation

An Approximate Dynamic Programming Algorithm for Monotone Value Functions

Jiang

Powell

2015

Operations Research

View full text Add to dashboard Cite

Many sequential decision problems can be formulated as Markov Decision Processes (MDPs) where the optimal value function (or cost-to-go function) can be shown to satisfy a monotone structure in some or all of its dimensions. When the state space becomes large, traditional techniques, such as the backward dynamic programming algorithm (i.e., backward induction or value iteration), may no longer be effective in finding a solution within a reasonable time frame, and thus we are forced to consider other approaches, such as approximate dynamic programming (ADP). We propose a provably convergent ADP algorithm called Monotone-ADP that exploits the monotonicity of the value functions in order to increase the rate of convergence. In this paper, we describe a general finite-horizon problem setting where the optimal value function is monotone, present a convergence proof for Monotone-ADP under various technical assumptions, and show numerical results for three application domains: optimal stopping, energy storage/allocation, and glycemic control for diabetes patients. The empirical results indicate that by taking advantage of monotonicity, we can attain high quality solutions within a relatively small number of iterations, using up to two orders of magnitude less computation than is needed to compute the optimal solution exactly.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

“…The optimal time for statin initiation, however, is a difficult medical problem due to the competing forces of health benefits and side effects. Kurt et al [2011] models the problem as an MDP with a value function monotone in a risk factor known as the lipid-ratio (LR). Finance…”

Section: Operations Researchmentioning

confidence: 99%

An Approximate Dynamic Programming Algorithm for Monotone Value Functions

Jiang

Powell

2015

Operations Research

View full text Add to dashboard Cite

show abstract

“…For certain problems, the IFR property together with some additional (and generally nonrestrictive) conditions guarantee an optimal threshold policy (see chapter 4 of Puterman [59] for a thorough discussion of this topic). These conditions have been used, for example, in the context of HIV (Shechter et al [65]), liver disease (Alagoz et al [2]), and type 2 diabetes (Kurt [44]) to prove the existence of an optimal control-limit policy under an ordering of states. A control-limit policy is one in which one action is optimal for all states below a certain threshold state (e.g., wait to transplant if the MELD score is below 25) and another action is optimal for all states at or above a certain value (e.g., transplant if the MELD score is at or above 25).…”

Section: Mdp and Pomdp Solution Methodsmentioning

confidence: 99%

Optimization of Sequential Decision Making for Chronic Diseases: From Data to Decisions

Denton

2018

Recent Advances in Optimization and Modeling of Contemporary Problems

Self Cite

View full text Add to dashboard Cite

Rapid advances in healthcare for chronic diseases such as cardiovascular disease, cancer, and diabetes have made it possible to detect diseases at early stages and tailor treatment based on individual patient risk factors including demographic factors and disease-specific biomarkers. However, a large number of relevant risk factors, combined with uncertainty in future health outcomes and the side effects of health interventions, makes clinical management of diseases challenging for physicians and patients. Data-driven operations research methods have the potential to help improve medical decision making by using observational data that are now routinely collected in many health systems. Optimization methods in particular, such as Markov decision processes and partially observable Markov decision processes, have the potential to improve the protracted sequential decisionmaking process that is common to many chronic diseases. This tutorial provides an introduction to some of the most commonly used methods for building and solving models to optimize sequential decision making. The context of chronic diseases is emphasized, but the methods apply broadly to sequential decision making under uncertainty. We pay special attention to the challenges associated with using observational data and the influence of model parameter uncertainty and ambiguity. Keywords stochastic dynamic programming • Markov decision process • hidden Markov model • chronic disease • data analytics

show abstract

“…Although screening for disease is one area of active research ("secondary prevention," as described in §3 above), there are many other areas of potential research that relate to chronic disease management. Specific examples include identification of high-risk geographical areas of disease (discussed below), optimizing treatment for cancer (Cao and Lim [13]), managing treatment for diabetic patients (Kurt et al [68], Mason et al [76], Shah et al [98]), evaluating treatment for patients with Alzheimer's (Craft et al [30]), preventing disease through early interventions, allocating resources between prevention and treatment (Alistar and Brandeau [2]), or allocating capacity for chronic care (Deo et al [36]). We describe work on two specific chronic diseases below.…”

Section: Chronic Diseasesmentioning

confidence: 99%

Research in Public Health for Efficient, Effective, and Equitable Outcomes

Ayer¹,

Keskinocak²,

Swann³

2014

Bridging Data and Decisions

View full text Add to dashboard Cite

Public health focuses on preventing disease, prolonging life, and promoting health in a population, and there are many ways in which operations research and management science (OR/MS) researchers can contribute to improved decision making in this area. In this tutorial, we discuss examples of work in several areas of public health, including access and equity, disease screening, chronic diseases, and infectious diseases. In each of these areas, we give a brief overview of challenging issues and related literature, give a specific example of a research topic, and point to potential future directions for research. We describe several trends that could impact future public health research including population ageing and an increased use of electronic records. The tutorial is intended to introduce OR/MS researchers to the many opportunities for having a positive impact on specific populations and society overall, through OR/MS research related to public health.

show abstract

The structure of optimal statin initiation policies for patients with Type 2 diabetes

Cited by 37 publications

References 47 publications

An Approximate Dynamic Programming Algorithm for Monotone Value Functions

An Approximate Dynamic Programming Algorithm for Monotone Value Functions

Optimization of Sequential Decision Making for Chronic Diseases: From Data to Decisions

Research in Public Health for Efficient, Effective, and Equitable Outcomes

Contact Info

Product

Resources

About