scholarly journals Comparison of Decision Modeling Approaches for Health Technology and Policy Evaluation

2020 ◽  
Author(s):  
John Graves ◽  
Shawn Garbett ◽  
Zilu Zhou ◽  
Jonathan S. Schildcrout ◽  
Josh Peterson
Keyword(s):  
Discrete Time ◽  
Stochastic Models ◽  
State Transition ◽  
Markov Models ◽  
Transition Probabilities ◽  
Discrete Event ◽  
Computation Time ◽  
Quality Adjusted Life Year ◽  
Simulated Patients ◽  
Optimal Decisions

ABSTRACTWe discuss tradeoffs and errors associated with approaches to modeling health economic decisions. Through an application in pharmacogenomic (PGx) testing to guide drug selection for individuals with a genetic variant, we assessed model accuracy, optimal decisions and computation time for an identical decision scenario modeled four ways: using (1) coupled-time differential equations [DEQ]; (2) a cohort-based discrete-time state transition model [MARKOV]; (3) an individual discrete-time state transition microsimulation model [MICROSIM]; and (4) discrete event simulation [DES]. Relative to DEQ, the Net Monetary Benefit for PGx testing (vs. a reference strategy of no testing) based on MARKOV with rate-to-probability conversions using commonly used formulas resulted in different optimal decisions. MARKOV was nearly identical to DEQ when transition probabilities were embedded using a transition intensity matrix. Among stochastic models, DES model outputs converged to DEQ with substantially fewer simulated patients (1 million) vs. MICROSIM (1 billion). Overall, properly embedded Markov models provided the most favorable mix of accuracy and run-time, but introduced additional complexity for calculating cost and quality-adjusted life year outcomes due to the inclusion of “jumpover” states after proper embedding of transition probabilities. Among stochastic models, DES offered the most favorable mix of accuracy, reliability, and speed.

scholarly journals Comparison of Decision Modeling Approaches for Health Technology and Policy Evaluation

2021 ◽  
Vol 41 (4) ◽  
pp. 453-464
Author(s):  
John Graves ◽  
Shawn Garbett ◽  
Zilu Zhou ◽  
Jonathan S. Schildcrout ◽  
Josh Peterson
Keyword(s):  
Discrete Time ◽  
Stochastic Models ◽  
State Transition ◽  
Markov Models ◽  
Transition Probabilities ◽  
Discrete Event ◽  
Computation Time ◽  
Quality Adjusted Life Year ◽  
Simulated Patients ◽  
Optimal Decisions

We discuss tradeoffs and errors associated with approaches to modeling health economic decisions. Through an application in pharmacogenomic (PGx) testing to guide drug selection for individuals with a genetic variant, we assessed model accuracy, optimal decisions, and computation time for an identical decision scenario modeled 4 ways: using 1) coupled-time differential equations (DEQ), 2) a cohort-based discrete-time state transition model (MARKOV), 3) an individual discrete-time state transition microsimulation model (MICROSIM), and 4) discrete event simulation (DES). Relative to DEQ, the net monetary benefit for PGx testing (v. a reference strategy of no testing) based on MARKOV with rate-to-probability conversions using commonly used formulas resulted in different optimal decisions. MARKOV was nearly identical to DEQ when transition probabilities were embedded using a transition intensity matrix. Among stochastic models, DES model outputs converged to DEQ with substantially fewer simulated patients (1 million) v. MICROSIM (1 billion). Overall, properly embedded Markov models provided the most favorable mix of accuracy and runtime but introduced additional complexity for calculating cost and quality-adjusted life year outcomes due to the inclusion of “jumpover” states after proper embedding of transition probabilities. Among stochastic models, DES offered the most favorable mix of accuracy, reliability, and speed.

scholarly journals A Hidden Semi-Markov Model with Duration-Dependent State Transition Probabilities for Prognostics

2014 ◽  
Vol 2014 ◽  
pp. 1-10 ◽  
Author(s):  
Ning Wang ◽  
Shu-dong Sun ◽  
Zhi-qiang Cai ◽  
Shuai Zhang ◽  
Can Saygin
Keyword(s):  
Markov Model ◽  
State Transition ◽  
Markov Models ◽  
Transition Probabilities ◽  
Independence Assumption ◽  
Modeling And Analysis ◽  
Maintenance Systems ◽  
Dependent State ◽  
Real World Problems

Realistic prognostic tools are essential for effective condition-based maintenance systems. In this paper, a Duration-Dependent Hidden Semi-Markov Model (DD-HSMM) is proposed, which overcomes the shortcomings of traditional Hidden Markov Models (HMM), including the Hidden Semi-Markov Model (HSMM): (1) it allows explicit modeling of state transition probabilities between the states; (2) it relaxes observations’ independence assumption by accommodating a connection between consecutive observations; and (3) it does not follow the unrealistic Markov chain’s memoryless assumption and therefore it provides a more powerful modeling and analysis capability for real world problems. To facilitate the computation of the proposed DD-HSMM methodology, new forward-backward algorithm is developed. The demonstration and evaluation of the proposed methodology is carried out through a case study. The experimental results show that the DD-HSMM methodology is effective for equipment health monitoring and management.

Reliability modeling for a discrete time multi-state system with random and dependent transition probabilities

2019 ◽  
Vol 233 (5) ◽  
pp. 747-760 ◽  
Author(s):  
Linmin Hu ◽  
Rui Peng
Keyword(s):  
Discrete Time ◽  
Power Distribution ◽  
State Transition ◽  
Transition Probabilities ◽  
Survival Function ◽  
Copula Function ◽  
State System ◽  
Dependence Structure ◽  
Reliability Modeling ◽  
Dynamic Reliability

In a random environment, state transition probabilities of a multi-state system can change as the environment changes. Thus, a dynamic reliability model with random and dependent transition probabilities is developed for non-repairable discrete-time multi-state system in this article. The dependence among the random state transition probabilities of the system is modeled by a copula function. By probability argument and random process theory, we obtain explicit expressions of some reliability characteristics and joint survival function of random time spent by the system in all working states (partially and completely working states). A special case is considered when the state transition probabilities are dependent random variables with power distribution, and the dependence structure is modeled by Farlie–Gumbel–Morgenstern copula. Numerical examples are also presented to demonstrate the developed model and perform a comparison for the models with random and fixed transition probabilities.

scholarly journals Worth the effort? A practical examination of random effects in hidden Markov models for animal telemetry data

2020 ◽  
Author(s):  
Brett T. McClintock
Keyword(s):  
Hidden Markov Models ◽  
Random Effects ◽  
State Transition ◽  
Markov Models ◽  
Transition Probabilities ◽  
Hidden Markov ◽  
Random Effect ◽  
Costs And Benefits ◽  
State Process ◽  
Random Effect Models

AbstractHidden Markov models (HMMs) that include individual-level random effects have recently been promoted for inferring animal movement behaviour from biotelemetry data. These “mixed HMMs” come at significant cost in terms of implementation and computation, and discrete random effects have been advocated as a practical alternative to more computationally-intensive continuous random effects. However, the performance of mixed HMMs has not yet been sufficiently explored to justify their widespread adoption, and there is currently little guidance for practitioners weighing the costs and benefits of mixed HMMs for a particular research objective.I performed an extensive simulation study comparing the performance of a suite of fixed and random effect models for individual heterogeneity in the hidden state process of a 2-state HMM. I focused on sampling scenarios more typical of telemetry studies, which often consist of relatively long time series (30 – 250 observations per animal) for relatively few individuals (5 – 100 animals).I generally found mixed HMMs did not improve state assignment relative to standard HMMs. Reliable estimation of random effects required larger sample sizes than are often feasible in telemetry studies. Continuous random effect models performed reasonably well with data generated under discrete random effects, but not vice versa. Random effects accounting for unexplained individual variation can improve estimation of state transition probabilities and measurable covariate effects, but discrete random effects can be a relatively poor (and potentially misleading) approximation for continuous variation.When weighing the costs and benefits of mixed HMMs, three important considerations are study objectives, sample size, and model complexity. HMM applications often focus on state assignment with little emphasis on heterogeneity in state transition probabilities, in which case random effects in the hidden state process simply may not be worth the additional effort. However, if explaining variation in state transition probabilities is a primary objective and sufficient explanatory covariates are not available, then random effects are worth pursuing as a more parsimonious alternative to individual fixed effects.To help put my findings in context and illustrate some potential challenges that practitioners may encounter when applying mixed HMMs, I revisit a previous analysis of long-finned pilot whale biotelemetry data.

AFSIM’s pseudo-realtime hybrid simulation software design

2021 ◽  
pp. 154851292098526
Author(s):  
J Scott Thompson ◽  
Douglas D Hodson
Keyword(s):  
Air Force ◽  
Discrete Time ◽  
Software Design ◽  
Research Laboratory ◽  
Discrete Event ◽  
Hybrid Simulation ◽  
Simulation Software ◽  
Human Interaction ◽  
Virtual Simulation ◽  
Air Force Research Laboratory

Simulation approaches generally fall into two categories: discrete time or discrete event. For military modeling and simulation needs, the two approaches typically align with virtual simulation, which implies human interaction with the simulation program, and constructive simulation, which implies no human interaction. The Air Force Research Laboratory develops and distributes AFSIM (Advanced Framework for Simulation, Integration, and Modeling) to a user community that uses both virtual and constructive simulation. This paper documents the software design and primary algorithms that provide AFSIM’s support for both modes, which is termed a hybrid simulation.

scholarly journals Time-to-event analysis in economic evaluations: a comparison of modelling methods to assess the cost-effectiveness of transplanting a marginal quality kidney

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Sameera Senanayake ◽  
Nicholas Graves ◽  
Helen Healy ◽  
Keshwar Baboolal ◽  
Adrian Barnett ◽  
...  
Keyword(s):  
Cost Effectiveness ◽  
Time Horizon ◽  
Transition Probabilities ◽  
Discrete Event ◽  
Economic Evaluations ◽  
Event Analysis ◽  
Time To Event ◽  
Modelling Method ◽  
Cycle Lengths ◽  
Modelling Methods

Abstract Background Economic-evaluations using decision analytic models such as Markov-models (MM), and discrete-event-simulations (DES) are high value adds in allocating resources. The choice of modelling method is critical because an inappropriate model yields results that could lead to flawed decision making. The aim of this study was to compare cost-effectiveness when MM and DES were used to model results of transplanting a lower-quality kidney versus remaining waitlisted for a kidney. Methods Cost-effectiveness was assessed using MM and DES. We used parametric survival models to estimate the time-dependent transition probabilities of MM and distribution of time-to-event in DES. MMs were simulated in 12 and 6 monthly cycles, out to five and 20-year time horizon. Results DES model output had a close fit to the actual data. Irrespective of the modelling method, the cycle length of MM or the time horizon, transplanting a low-quality kidney as compared to remaining waitlisted was the dominant strategy. However, there were discrepancies in costs, effectiveness and net monetary benefit (NMB) among different modelling methods. The incremental NMB of the MM in the 6-months cycle lengths was a closer fit to the incremental NMB of the DES. The gap in the fit of the two cycle lengths to DES output reduced as the time horizon increased. Conclusion Different modelling methods were unlikely to influence the decision to accept a lower quality kidney transplant or remain waitlisted on dialysis. Both models produced similar results when time-dependant transition probabilities are used, most notable with shorter cycle lengths and longer time-horizons.

Sign in / Sign up

Export Citation Format

Share Document