Jörgen Möller

Modeling Using Discrete Event Simulation A Report of the ISPOR-SMDM Modeling Good Research Practices Task Force–4

Discrete event simulation (DES) is a form of computer-based modeling that provides an intuitive and flexible approach to representing complex systems. It has been used in a wide range of health care applications. Most early applications involved analyses of systems with constrained resources, where the general aim was to improve the organization of delivered services. More recently, DES has increasingly been applied to evaluate specific technologies in the context of health technology assessment. The aim of this article was to provide consensus-based guidelines on the application of DES in a health care setting, covering the range of issues to which DES can be applied. The article works through the different stages of the modeling process: structural development, parameter estimation, model implementation, model analysis, and representation and reporting. For each stage, a brief description is provided, followed by consideration of issues that are of particular relevance to the application of DES in a health care setting. Each section contains a number of best practice recommendations that were iterated among the authors, as well as among the wider modeling task force.

Discrete Event Simulation: The Preferred Technique for Health Economic Evaluations?

OBJECTIVES To argue that discrete event simulation should be preferred to cohort Markov models for economic evaluations in health care. METHODS The basis for the modeling techniques is reviewed. For many health-care decisions, existing data are insufficient to fully inform them, necessitating the use of modeling to estimate the consequences that are relevant to decision-makers. These models must reflect what is known about the problem at a level of detail sufficient to inform the questions. Oversimplification will result in estimates that are not only inaccurate, but potentially misleading. RESULTS Markov cohort models, though currently popular, have so many limitations and inherent assumptions that they are inadequate to inform most health-care decisions. An event-based individual simulation offers an alternative much better suited to the problem. A properly designed discrete event simulation provides more accurate, relevant estimates without being computationally prohibitive. It does require more data and may be a challenge to convey transparently, but these are necessary trade-offs to provide meaningful and valid results. CONCLUSION In our opinion, discrete event simulation should be the preferred technique for health economic evaluations today.

Modeling Using Discrete Event Simulation

Discrete event simulation (DES) is a form of computer-based modeling that provides an intuitive and flexible approach to representing complex systems. It has been used in a wide range of health care applications. Most early applications involved analyses of systems with constrained resources, where the general aim was to improve the organization of delivered services. More recently, DES has increasingly been applied to evaluate specific technologies in the context of health technology assessment. The aim of this article is to provide consensus-based guidelines on the application of DES in a health care setting, covering the range of issues to which DES can be applied. The article works through the different stages of the modeling process: structural development, parameter estimation, model implementation, model analysis, and representation and reporting. For each stage, a brief description is provided, followed by consideration of issues that are of particular relevance to the application of DES in a health care setting. Each section contains a number of best practice recommendations that were iterated among the authors, as well as the wider modeling task force.

Cost-effectiveness of novel relapsed-refractory multiple myeloma therapies in Norway: lenalidomide plus dexamethasone vs bortezomib

Abstract Objective: To estimate the cost-effectiveness (cost per additional life-year [LY] and quality-adjusted life-year [QALY] gained) of lenalidomide plus dexamethasone (LEN/DEX) compared with bortezomib for the treatment of relapsed-refractory multiple myeloma (rrMM) in Norway. Methods: A discrete-event simulation model was developed to predict patients’ disease course using patient data, best response, and efficacy levels obtained from LEN/DEX MM-009/-010 trials and the bortezomib (APEX) published clinical trial. Predictive equations for time-to-progression (TTP) and post-progression survival (PPS) were developed by identifying the best fitting parametric survival distributions and selecting the most significant predictors. Disease and adverse event management was obtained via survey from Norwegian experts. Costs, derived from official Norwegian pricing data bases, included drug, administration, monitoring, and adverse event management costs. Results: Complete or partial responders were 65% for LEN/DEX compared to 43% for bortezomib. Derived median TTP was 11.45 months for LEN/DEX compared to 5.15 months for bortezomib. LYs and QALYs were higher for LEN/DEX (4.06 and 2.95, respectively) than for bortezomib (3.11 and 2.19, respectively). The incremental costs per QALY and LY gained from LEN/DEX were NOK 247,978 and NOK 198,714, respectively, compared to bortezomib. Multiple sensitivity analyses indicated the findings were stable. The parameters with the greatest impact were 4-year time horizon (NOK 441,457/QALY) and higher bound confidence intervals for PPS (NOK 118,392). Limitations: The model analyzed two therapies not compared in head-to-head trials, and predicted results using an equation incorporating patient-level characteristics. It is a limited estimation of the costs and outcomes in a Norwegian setting. Conclusions: The simulation model showed that treatment with LEN/DEX leads to greater LYs and QALYs when compared to bortezomib in the treatment of rrMM patients. The incremental cost-effectiveness ratio indicated treatment with LEN/DEX to be cost-effective and was the basis of the reimbursement approval of LEN/DEX in Norway.

Invasive meningococcal disease epidemiology and control measures: a framework for evaluation

BackgroundMeningococcal disease can have devastating consequences. As new vaccines emerge, it is necessary to assess their impact on public health. In the absence of long-term real world data, modeling the effects of different vaccination strategies is required. Discrete event simulation provides a flexible platform with which to conduct such evaluations.MethodsA discrete event simulation of the epidemiology of invasive meningococcal disease was developed to quantify the potential impact of implementing routine vaccination of adolescents in the United States with a quadrivalent conjugate vaccine protecting against serogroups A, C, Y, and W-135. The impact of vaccination is assessed including both the direct effects on individuals vaccinated and the indirect effects resulting from herd immunity. The simulation integrates a variety of epidemiologic and demographic data, with core information on the incidence of invasive meningococcal disease and outbreak frequency derived from data available through the Centers for Disease Control and Prevention. Simulation of the potential indirect benefits of vaccination resulting from herd immunity draw on data from the United Kingdom, where routine vaccination with a conjugate vaccine has been in place for a number of years. Cases of disease are modeled along with their health consequences, as are the occurrence of disease outbreaks.ResultsWhen run without a strategy of routine immunization, the simulation accurately predicts the age-specific incidence of invasive meningococcal disease and the site-specific frequency of outbreaks in the Unite States. 2,807 cases are predicted annually, resulting in over 14,000 potential life years lost due to invasive disease. In base case analyses of routine vaccination, life years lost due to infection are reduced by over 45% (to 7,600) when routinely vaccinating adolescents 12 years of age at 70% coverage. Sensitivity analyses indicate that herd immunity plays an important role when this population is targeted for vaccination. While 1,100 cases are avoided annually when herd immunity effects are included, in the absence of any herd immunity, the number of cases avoided with routine vaccination falls to 380 annually. The duration of vaccine protection also strongly influences results.ConclusionIn the absence of appropriate real world data on outcomes associated with large-scale vaccination programs, decisions on optimal immunization strategies can be aided by discrete events simulations such as the one described here. Given the importance of herd immunity on outcomes associated with routine vaccination, published estimates of the economic efficiency of routine vaccination with a quadrivalent conjugate vaccine in the United States may have considerably underestimated the benefits associated with a policy of routine immunization of adolescents.

Decision-Analytic Models: Current Methodological Challenges

Modelers seeking to help inform decisions about insurance (public or private) coverage of the cost of pharmaceuticals or other health care interventions face various methodological challenges. In this review, which is not meant to be comprehensive, we cover those that in our experience are most vexing. The biggest challenge is getting decision makers to trust the model. This is a major problem because most models undergo only cursory validation; our field has lacked the motivation, time, and data to properly validate models intended to inform health care decisions. Without documented, adequate validation, there is little basis for decision makers to have confidence that the model’s results are credible and should be used in a health technology appraisal. A fundamental problem for validation is that the models are very artificial and lack sufficient depth to adequately represent the reality they are simulating. Typically, modelers assume that all resources have infinite capacity so any patient needing care receives it immediately; there are no waiting times or queues, contrary to the common experience in actual practice. Moreover, all the patients enter the model simultaneously at time zero rather than over time as happens in actuality; differences between patients are ignored or minimized and structural modeling choices that make little sense (e.g., using states to represent events) are forced by commitment to a technique (and even to specific spreadsheet software!). The resulting structural uncertainty is rarely addressed, because methods are lacking and even probabilistic analysis of parameter uncertainty suffers from weak consideration of correlation and arbitrary distribution choices. Stakeholders must see to it that models are fit for the stated purpose and provide the best possible estimates given available data—the decisions at stake deserve nothing less.

Economic and health consequences of managing bradycardia with dual-chamber compared to single-chamber ventricular pacemakers in Italy.

Objective This study sought to estimate the economic implications of managing bradycardia due to sinoatrial node disease or atrioventricular block with dual compared to single-chamber ventricular pacemakers from an Italian government perspective. Dual-chamber pacemakers lower the risk of developing atrial fibrillation and pacemaker syndrome. Methods A discrete event simulation of a patients course for 5 years following pacemaker implantation. Each patient may experience the following: complications, pacemaker syndrome, atrial fibrillation, stroke, or death. Risk functions were based on published data from the Canadian Trial of Physiologic Pacing and Mode Selection Trial in Sinus-Node Dysfunction. Identical patients were simulated after receiving a single or dual-chamber pacemaker. Quality-adjusted life-years (QALYs) and direct medical costs were estimated (2004 Euros). Benefits and costs were discounted at 3%. Results The model predicts that implanting the dual-chamber device in 1000 patients will prevent 36 patients from developing atrial fibrillation, 168 from developing severe pacemaker syndrome, but will lead to 13 additional hospitalizations with complications over 5 years. Health benefits are achieved at an incremental cost of &U20AC; 23 per patient, and 0.09 QALY, yielding an incremental cost–effectiveness ratio of &U20AC; 260/QALY. Sensitivity analysis shows that device replacement rates due to pacemaker syndrome have the biggest impact on the final results. Conclusions In the long term, higher initial costs of the dual-chamber device may be offset by a reduction in costs associated with reoperations and atrial fibrillation.

Advantages and disadvantages of discrete-event simulation for health economic analyses.

Health technology assessments (HTA) are carried out to inform decision-makers of the possible consequences of agreeing to pay for a particular medication or other intervention. To carry out these a...

Economic implications of growth hormone use in patients with short bowel syndrome

ABSTRACT Objective: Short bowel syndrome is a rare, lifethreatening condition that can result in nutritional malabsorption. Parenteral nutrition provides life-saving support but can lead to complications and affect quality of life. Recombinant human growth hormone, somatropin (rDNA origin), has been shown to significantly reduce dependence on nutritional support (p < 0.05). This study evaluates the economic impact of somatropin use in the management of short bowel syndrome. Methods: A discrete event simulation (DES) model was developed to estimate the benefits and costs associated with somatropin use. Risks of treatment complications and of disease-related events were modeled in identical patient pairs – one receiving parenteral nutrition alone, the other receiving 4 weeks of somatropin – for 2 years following initiation of treatment. Life expectancy was assumed equivalent. Risk functions were estimated from the literature and one randomized clinical trial. Total and component costs associated with each strategy were determined. The distribution of patients reducing parenteral nutrition need and the final parenteral nutrition frequency were also estimated. Sensitivity analyses were completed for key inputs. Direct medical costs are reported in USD

Validation of a DICE Simulation Against a Discrete Event Simulation Implemented Entirely in Code

2004. Results: The model predicted that 96.0% of patients receiving somatropin reduce or eliminate parenteral nutrition within 6 weeks: average use was reduced by 2.8 days and one-third weaned completely. Based on 1.9 L of parenteral nutrition per day, estimated costs were