Prioritizing Hepatitis C Treatment in U.S. Prisons

Abstract

Hepatitis C virus (HCV) prevalence in prison systems is about 10 times higher than in the community. As such, prison systems offer a unique opportunity to control the HCV epidemic. New HCV-treatment drugs are very effective, but providing treatment to all inmates is prohibitively expensive unless prices fall. Current practice is to prioritize treatment based on disease severity and puts less emphasis on other factors such as the remaining sentence length and injection drug use behavior. In “Prioritizing Hepatitis C Treatment in U.S. Prisons,” T. Ayer, C. Zhang, A. Bonifonte, A. Spaulding, and J. Chhatwal analyze optimal approaches for treatment prioritization under resource constraints by developing a restless bandit modeling framework. They present an easy-to-implement closed-form index policy to support hepatitis C treatment prioritization decisions in U.S. prisons. They also test their proposed policy using a detailed, realistic agent-based simulation model and shed light on several controversial health policy decisions related to hepatitis C treatment prioritization.

OM Forum—Pandemics/Epidemics: Challenges and Opportunities for Operations Management Research

“…Their findings show that intervention via regulating and enforcing tattoo parlors is cost-effective for HCV control. Ayer et al (2019) investigate how prisoners suffering from hepatitis C virus (HCV) infection need to be prioritized for treatments. The appropriateness of prioritization is identified based on criteria such as liver disease and remaining sentence length (short or long).…”

A Fast-Pivoting Algorithm for Whittle’s Restless Bandit Index

“…Three approaches can be distinguished in the literature: deriving the Whittle index in closed form, iterative index approximation, and exact numerical computation. The first approach is to derive the Whittle index in closed form, as, for example, in [5,13,[15][16][17][18]. This is to be preferred whenever possible, as the resulting analytical expressions for the Whittle index, besides facilitating its numerical evaluation, may provide valuable insight on the index dependence on model parameters.…”

“…Besides its intrinsic interest for solving the aforementioned single-project parametric problem collection P, Whittle proposed in [2] to use the index λ * i as the basis of a widely popular heuristic for the multi-armed restless bandit problem (MARBP), in which M out of N > M restless bandit projects must be selected to be engaged at each time to maximize the value (under a discounted or long-run average criterion) earned from the N projects over an infinite horizon. For a sample of recent applications, see, for example, [5,[7][8][9][10][11][12][13][14][15][16][17][18]. While the MARBP is computationally intractable (PSPACE-hard; see [19]), the Whittle index policy is an intuitively appealing heuristic where, at each time, M projects with the highest current indices are engaged, so the Whittle index plays the role of a priority index for a project to be engaged.…”

Multi-Gear Bandits, Partial Conservation Laws, and Indexability

“…The latter has huge modeling power but is computationally intractable, and Whittle's index policy has proven effective in an ever-increasing variety of models for multifarious applications. Thus, e.g., to name a few, scheduling multi-class make-to-stock queues [9], scheduling multi-class queues with finite buffers [10], admission control and routing to parallel queues with reneging [11], obsolescence mitigation strategies [12], sensor scheduling and dynamic channel selection [13][14][15][16], group maintenance [17], multi-target tracking with Kalman filter dynamics [18,19], scheduling multi-armed bandits with switching costs [20] or switching delays [21], the dynamic prioritization of medical treatments or interventions [22,23], and resource allocation with varying requests and with resources shared by multiple requests [24].…”