Alternative decision modelling techniques for the evaluation of health care technologies: Markov processes versus discrete event simulation

The Cost-Effectiveness of Pertuzumab for the Treatment of Metastatic HER2+ Breast Cancer in Czechia: A Semi-Markov Model Using Cost States

OBJECTIVES

This article estimates the cost-effectiveness of adding pertuzumab to the combination of trastuzumab and docetaxel within the first-line treatment for metastatic breast cancer with the amplification of HER2+.

METHODS

Data from Czech clinical practice recorded in the BREAST register are used. A semi-Markov model with states derived based on the treatment phases (first-line medication, no medication, next-line medication, death) is defined to estimate costs from the healthcare payers' perspective. The benefits are estimated as patient survival until death. The Kaplan-Meier estimates are supplemented by the Cox proportional hazard and the accelerated failure time models to control for patient characteristics. Health-related quality-of-life indicators are derived from relevant literature.

RESULTS

Based on the used data, adding pertuzumab does not result in statistically significantly longer survival while inducing higher treatment costs (€163 360 compared with €90 112 per patient in 2018 prices). Statistically longer survival was not supported by the log-rank test (P = .97), the Cox proportional hazard model, or the accelerated failure time model using the Gompertz distribution. The incremental cost-effectiveness ratio (€87 200) substantially exceeds the willingness to pay for 1 quality-adjusted life-year (€46 500).

CONCLUSIONS

This analysis indicates that adding pertuzumab cannot be considered cost-effective in Czechia. However, the observed phenomenon may be attributed to the limited duration of patient follow-up periods at the time of the study's execution (mean of 20-21 months). Importantly, we find that using states connected to specific treatment phases is appropriate for a retrospective analysis of patient-level clinical data.

Health Economic Research Assessing the Value of Early Detection of Cardiovascular Disease: A Systematic Review

INTRODUCTION

Cardiovascular disease (CVD) is the most prominent cause of death worldwide and has a major impact on healthcare budgets. While early detection strategies may reduce the overall CVD burden through earlier treatment, it is unclear which strategies are (most) efficient.

AIM

This systematic review reports on the cost effectiveness of recent early detection strategies for CVD in adult populations at risk.

METHODS

PubMed and Scopus were searched to identify scientific articles published between January 2016 and May 2022. The first reviewer screened all articles, a second reviewer independently assessed a random 10% sample of the articles for validation. Discrepancies were solved through discussion, involving a third reviewer if necessary. All costs were converted to 2021 euros. Reporting quality of all studies was assessed using the CHEERS 2022 checklist.

RESULTS

In total, 49 out of 5552 articles were included for data extraction and assessment of reporting quality, reporting on 48 unique early detection strategies. Early detection of atrial fibrillation in asymptomatic patients was most frequently studied (n = 15) followed by abdominal aortic aneurysm (n = 8), hypertension (n = 7) and predicted 10-year CVD risk (n = 5). Overall, 43 strategies (87.8%) were reported as cost effective and 11 (22.5%) CVD-related strategies reported cost reductions. Reporting quality ranged between 25 and 86%.

CONCLUSIONS

Current evidence suggests that early CVD detection strategies are predominantly cost effective and may reduce CVD-related costs compared with no early detection. However, the lack of standardisation complicates the comparison of cost-effectiveness outcomes between studies. Real-world cost effectiveness of early CVD detection strategies will depend on the target country and local context.

REGISTRATION OF SYSTEMATIC REVIEW

CRD42022321585 in International Prospective Registry of Ongoing Systematic Reviews (PROSPERO) submitted at 10 May 2022.

Process mining for healthcare decision analytics with micro-costing estimations

Managing constrained healthcare resources is an important and inescapable role of healthcare decision makers. Allocative decisions are based on downstream consequences of changes to care processes: judging whether the costs involved are offset by the magnitude of the consequences, and therefore whether the change represents value for money. Process mining techniques can inform such decisions by quantitatively discovering, comparing and detailing care processes using recorded data, however the scope of techniques typically excludes anything 'after-the-process' i.e., their accumulated costs and resulting consequences. Cost considerations are increasingly incorporated into process mining techniques, but the majority of healthcare costs for service and overhead components are commonly apportioned and recorded at the patient (trace) level, hiding event level detail. Within decision-analysis, event-driven and individual-level simulation models are sometimes used to forecast the expected downstream consequences of process changes, but are expensive to manually operationalise. In this paper, we address both of these gaps within and between process mining and decision analytics, by better linking them together. In particular, we introduce a new type of process model containing trace data that can be used in individual-level or cohort-level decision-analytical model building. Furthermore, we enhance these models with process-based micro-costing estimations. The approach was evaluated with health economics and decision modelling experts, with discussion centred on how the outputs could be used, and how similar information would otherwise be compiled.

Discrete-Event Simulation Modeling in Healthcare: A Comprehensive Review

Discrete-event simulation (DES) is a stochastic modeling approach widely used to address dynamic and complex systems, such as healthcare. In this review, academic databases were systematically searched to identify 231 papers focused on DES modeling in healthcare. These studies were sorted by year, approach, healthcare setting, outcome, provenance, and software use. Among the surveys, conceptual/theoretical studies, reviews, and case studies, it was found that almost two-thirds of the theoretical articles discuss models that include DES along with other analytical techniques, such as optimization and lean/six sigma, and one-third of the applications were carried out in more than one healthcare setting, with emergency departments being the most popular. Moreover, half of the applications seek to improve time- and efficiency-related metrics, and one-third of all papers use hybrid models. Finally, the most popular DES software is Arena and Simul8. Overall, there is an increasing trend towards using DES in healthcare to address issues at an operational level, yet less than 10% of DES applications present actual implementations following the modeling stage. Thus, future research should focus on the implementation of the models to assess their impact on healthcare processes, patients, and, possibly, their clinical value. Other areas are DES studies that emphasize their methodological formulation, as well as the development of frameworks for hybrid models.

Modeling Early Warning Systems: Construction and Validation of a Discrete Event Simulation Model for Heart Failure

OBJECTIVES

Developing and validating a discrete event simulation model that is able to model patients with heart failure managed with usual care or an early warning system (with or without a diagnostic algorithm) and to account for the impact of individual patient characteristics in their health outcomes.

METHODS

The model was developed using patient-level data from the Trans-European Network - Home-Care Management System study. It was coded using RStudio Version 1.3.1093 (version 3.6.2.) and validated along the lines of the Assessment of the Validation Status of Health-Economic decision models tool. The model includes 20 patient and disease characteristics and generates 8 different outcomes. Model outcomes were generated for the base-case analysis and used in the model validation.

RESULTS

Patients managed with the early warning system, compared with usual care, experienced an average increase of 2.99 outpatient visits and a decrease of 0.02 hospitalizations per year, with a gain of 0.81 life years (0.45 quality-adjusted life years) and increased average total costs of €11 249. Adding a diagnostic algorithm to the early warning system resulted in a 0.92 life year gain (0.57 quality-adjusted life years) and increased average costs of €9680. These patients experienced a decrease of 0.02 outpatient visits and 0.65 hospitalizations per year, while they avoided being hospitalized 0.93 times. The model showed robustness and validity of generated outcomes when comparing them with other models addressing the same problem and with external data.

CONCLUSIONS

This study developed and validated a unique patient-level simulation model that can be used for simulating a wide range of outcomes for different patient subgroups and treatment scenarios. It provides useful information for guiding research and for developing new treatment options by showing the hypothetical impact of these interventions on a large number of important heart failure outcomes.

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

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.

Comparison of Decision Modeling Approaches for Health Technology and Policy Evaluation

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.

Quantitative Evidence Synthesis Methods for the Assessment of the Effectiveness of Treatment Sequences for Clinical and Economic Decision Making: A Review and Taxonomy of Simplifying Assumptions

Sequential use of alternative treatments for chronic conditions represents a complex intervention pathway; previous treatment and patient characteristics affect both the choice and effectiveness of subsequent treatments. This paper critically explores the methods for quantitative evidence synthesis of the effectiveness of sequential treatment options within a health technology assessment (HTA) or similar process. It covers methods for developing summary estimates of clinical effectiveness or the clinical inputs for the cost-effectiveness assessment and can encompass any disease condition. A comprehensive review of current approaches is presented, which considers meta-analytic methods for assessing the clinical effectiveness of treatment sequences and decision-analytic modelling approaches used to evaluate the effectiveness of treatment sequences. Estimating the effectiveness of a sequence of treatments is not straightforward or trivial and is severely hampered by the limitations of the evidence base. Randomised controlled trials (RCTs) of sequences were often absent or very limited. In the absence of sufficient RCTs of whole sequences, there is no single best way to evaluate treatment sequences; however, some approaches could be re-used or adapted, sharing ideas across different disease conditions. Each has advantages and disadvantages, and is influenced by the evidence available, extent of treatment sequences (number of treatment lines or permutations), and complexity of the decision problem. Due to the scarcity of data, modelling studies applied simplifying assumptions to data on discrete treatments. A taxonomy for all possible assumptions was developed, providing a unique resource to aid the critique of existing decision-analytic models.

Decision Modeling for Economic Evaluation in Otolaryngology-Head and Neck Surgery: Review of Techniques

Decision making in health care is complex, and substantial uncertainty can be involved. Structured, systematic approaches to the integration of available evidence, assessment of uncertainty, and determination of choice are of significant benefit in an era of "value-based care." This is especially true for otolaryngology-head and neck surgery, where technological advancements are frequent and applicable to an array of subspecialties. Decision analysis aims to achieve these goals through various modeling techniques, including (1) decision trees, (2) Markov process, (3) microsimulation, and (4) discrete event simulation. While decision models have been used for decades, many clinicians and researchers continue to have difficulty deciphering them. In this review, we present an overview of various decision analysis modeling techniques, their purposes, how they can be interpreted, and commonly used syntax to promote understanding and use of these approaches. Throughout, we provide a sample research question to facilitate discussion of the advantages and disadvantages of each technique.

Patient and Public Involvement (PPI) in Health Economics Methodology Research: Reflections and Recommendations

Patient and public involvement (PPI) can be used in methods research, as well as applied research, in health economics. However, methods research goals may seem quite abstract when compared to the lived experiences of lay participants. This article draws on 4 years of PPI in a research project to develop methods for including family carer outcomes in economic evaluation. Key challenges in using PPI for health economics methods research relate to (1) training and preparation, (2) maintaining involvement, and (3) selecting suitable tasks. We suggest three criteria for selecting a research task for PPI input based on task importance, professional researcher skills gap, and potential PPI contribution.

How Useful Are Early Economic Models? Comment on "Problems and Promises of Health Technologies: The Role of Early Health Economic Modelling"

Early economic modelling has long been recommended to aid research and development (R&D) decisions in medical innovation, although they are less frequently published and critically appraised. A review of 30 innovations by Grutters et al provides an opportunity to evaluate how early models are used in practice. The evidence of early models can be used to inform two types of decision: to continue development ("stop or go") or to alter future R&D activities. I argue that early models have limited use in stop or go decisions, as less resource and data undermine the reliability of the models’ indicative estimates of cost-effectiveness. Whilst they are far more useful for informing future R&D directions, the best techniques available from statistical decision science, such as value of information analysis, are not regularly used. It is highly recommended that early models adopt these methods to best deal with uncertainty, quantify the potential value of further research, identify areas of study with the greatest potential benefit and generate recommendations on study design and sample size.

Healthcare Costs of Treating Privately Insured Patients with Acute Myeloid Leukemia in the United States from 2004 to 2014: A Generalized Additive Modeling Approach

OBJECTIVES

The primary objective of this study was to predict healthcare cost trajectories for patients with newly diagnosed acute myeloid leukemia (AML) receiving allogeneic hematopoietic cell transplantation (alloHCT), as a function of days since chemotherapy initiation, days relative to alloHCT, and days before death or last date of insurance eligibility (LDE). An exploratory objective examined patients with AML receiving chemotherapy only.

METHODS

We used Optum's de-identified Clinformatics^® Data Mart Database to construct cumulative cost trajectories from chemotherapy initiation to death or LDE (through 31 December 2014) for US patients aged 20-74 years diagnosed between 1 March 2004 and 31 December 2013 (n = 187 alloHCT; n = 253 chemotherapy only). We used generalized additive modeling (GAM) to predict expected trajectories and bootstrapped confidence intervals (CIs) at user-specified intervals conditional on dates of alloHCT and death or LDE relative to chemotherapy initiation.

RESULTS

Expected costs (in 2017 values) for a hypothetical patient receiving alloHCT 60 days after chemotherapy initiation and followed for 5 years were $US572,000 (95% CI 517,000-633,000); $US119,000 (95% CI 51,000-192,000); $US102,000 (95% CI 0-285,000); $US79,000 (95% CI 0-233,000), for years 1-4, respectively, and either $US494,000 (95% CI 212,000-799,000) or $US108,000 (95% CI 0-230,000) in year 5, whether the patient died or was lost to follow-up on day 1825, respectively.

CONCLUSIONS

Rates of cost accrual varied over time since chemotherapy initiation, with accelerations around the time of alloHCT and death. GAM is a potentially useful approach for imputing longitudinal costs relative to treatment initiation and one or more intercurrent, clinical, or terminal events in randomized controlled trials or registries with unrecorded costs or for dynamic decision-analytic models.

Systematic Review of Decision Analytic Modelling in Economic Evaluations of Low Back Pain and Sciatica

BACKGROUND

Low back pain (LBP) and sciatica place significant burden on individuals and healthcare systems, with societal costs alone likely to be in excess of £15 billion. Two recent systematic reviews for LBP and sciatica identified a shortage of modelling studies in both conditions.

OBJECTIVES

The aim of this systematic review was to document existing model-based economic evaluations for the treatment and management of both conditions; critically appraise current modelling techniques, analytical methods, data inputs, and structure, using narrative synthesis; and identify unresolved methodological problems and gaps in the literature.

METHODS

A systematic literature review was conducted whereby 6512 records were extracted from 11 databases, with no date limits imposed. Studies were abstracted according to a predesigned protocol, whereby they must be economic evaluations that employed an economic decision model and considered any management approach for LBP and sciatica. Study abstraction was initially performed by one reviewer who removed duplicates and screened titles to remove irrelevant studies. Overall, 133 potential studies for inclusion were then screened independently by other reviewers. Consensus was reached between reviewers regarding final inclusion.

RESULTS

Twenty-one publications of 20 unique models were included in the review, five of which were modelling studies in LBP and 16 in sciatica. Results revealed a poor standard of modelling in both conditions, particularly regarding modelling techniques, analytical methods, and data quality. Specific issues relate to inappropriate representation of both conditions in terms of health states, insufficient time horizons, and use of inappropriate utility values.

CONCLUSION

High-quality modelling studies, which reflect modelling best practice, as well as contemporary clinical understandings of both conditions, are required to enhance the economic evidence for treatments for both conditions.

An introduction to the methods of decision-analytic modelling used in economic evaluations for Dermatologists

Economic evaluations are used to identify which health treatments or preventions offer the most effective use of resources, or value for money. This is achieved by identifying, measuring and valuing the inputs and outcomes of alternative interventions. These evaluations are often conducted alongside clinical trials; however, these trials may end before the outcomes of economic interest have been observed and measured. An alternative to within trial economic evaluation is to use decision modelling, which can model the cost-effectiveness of interventions over an extended time period. This paper aims to provide an overview for clinicians of the different modelling techniques used within health economic evaluations and to introduce methods for critical appraisal. The most common modelling approaches, and their associated strengths and weaknesses, were discussed. Alongside this, practical examples specific to dermatology were given. These examples include studies where the model chosen or the methods used may not have been the most appropriate. Methods for critical appraisal were also highlighted. Common modelling approaches include Decision Trees, Markov Cohort, extensions to the Markov model (Monte Carlo Simulation) and Discrete Event Simulation models. Items of the Philips Checklist were discussed in the context of performing critical appraisal. Health economic decision models are multi-faceted and can often be complex. Full critical appraisal requires clinicians' unique knowledge, which is complementary to the knowledge of health economists.

Decision modelling for economic evaluation of liver transplantation

As the gap between a shortage of organs and the immense demand for liver grafts persists, every available donor liver needs to be optimized for utility, urgency and equity. To overcome this challenge, decision modelling might allow us to gather evidence from previous studies as well as compare the costs and consequences of alternative options. For public health policy and clinical intervention assessment, it is a potentially powerful tool. The most commonly used types of decision analytical models include decision trees, the Markov model, microsimulation, discrete event simulation and the system dynamic model. Analytic models could support decision makers in the field of liver transplantation when facing specific problems by synthesizing evidence, comprising all relevant options, generalizing results to other contexts, extending the time horizon and exploring the uncertainty. For modeling studies of economic evaluation for transplantation, understanding the current nature of the disease is crucial, as well as the selection of appropriate modelling techniques. The quality and availability of data is another key element for the selection and development of decision analytical models. In addition, good practice guidelines should be complied, which is important for standardization and comparability between economic outputs.

Matching the model with the evidence: comparing discrete event simulation and state-transition modeling for time-to-event predictions in a cost-effectiveness analysis of treatment in metastatic colorectal cancer patients

BACKGROUND

Individual patient data, e.g. from clinical trials, often need to be extrapolated or combined with additional evidence when assessing long-term impact in cost-effectiveness modeling studies. Different modeling methods can be used to represent the complex dynamics of clinical practice; the choice of which may impact cost-effectiveness outcomes. We compare the use of a previously designed cohort discrete-time state-transition model (DT-STM) with a discrete event simulation (DES) model.

METHODS

The original DT-STM was replicated and a DES model developed using AnyLogic software. Models were populated using individual patient data of a phase III study in metastatic colorectal cancer patients, and compared based on their evidence structure, internal validity, and cost-effectiveness outcomes. The DT-STM used time-dependent transition probabilities, whereas the DES model was populated using parametric distributions.

RESULTS

The estimated time-dependent transition probabilities for the DT-STM were irregular and more sensitive to single events due to the required small cycle length and limited number of event observations, whereas parametric distributions resulted in smooth time-to-event curves for the DES model. Although the DT-STM and DES model both yielded similar time-to-event curves, the DES model represented the trial data more accurately in terms of mean health-state durations. The incremental cost-effectiveness ratio (ICER) was €172,443 and €168,383 per Quality Adjusted Life Year gained for the DT-STM and DES model, respectively.

CONCLUSION

DES represents time-to-event data from clinical trials more naturally and accurately than DT-STM when few events are observed per time cycle. As a consequence, DES is expected to yield a more accurate ICER.

Discrete Event Simulation for Decision Modeling in Health Care: Lessons from Abdominal Aortic Aneurysm Screening

Markov models are often used to evaluate the cost-effectiveness of new healthcare interventions but they are sometimes not flexible enough to allow accurate modeling or investigation of alternative scenarios and policies. A Markov model previously demonstrated that a one-off invitation to screening for abdominal aortic aneurysm (AAA) for men aged 65 y in the UK and subsequent follow-up of identified AAAs was likely to be highly cost-effective at thresholds commonly adopted in the UK (£20,000 to £30,000 per quality adjusted life-year). However, new evidence has emerged and the decision problem has evolved to include exploration of the circumstances under which AAA screening may be cost-effective, which the Markov model is not easily able to address. A new model to handle this more complex decision problem was needed, and the case of AAA screening thus provides an illustration of the relative merits of Markov models and discrete event simulation (DES) models. An individual-level DES model was built using the R programming language to reflect possible events and pathways of individuals invited to screening v. those not invited. The model was validated against key events and cost-effectiveness, as observed in a large, randomized trial. Different screening protocol scenarios were investigated to demonstrate the flexibility of the DES. The case of AAA screening highlights the benefits of DES, particularly in the context of screening studies.

Quality assurance target for community-based breast cancer screening in China: a model simulation

Background

We aimed to clarify the feasibility of a community-based screening strategy for breast cancer in Tianjin, China; to identify the factors that most significantly influenced its feasibility; and to identify the reference range for quality control.

Methods

A state-transition Markov model simulated a hypothetical cohort of 100,000 healthy women, the start aged was set at 35 years and the time horizon was set to 50 years. The primary outcome for the model was the incremental cost-utility ratio (ICUR), defined as the program’s cost per quality-adjusted life year (QALY) gained. Three screening strategies providing by community health service for women aged 35 to 69 years was compared regarding to different intervals.

Result

The probability of the ICUR being below 20 272USD (i.e., triple the annual gross domestic product [3 GDPs]) per QALY saved was 100% for annual screening strategy and screening every three years. Only when the attendance rate was > 50%, the probability for annual screening would be cost effective > 95%. The probability for the annual screening strategy being cost effective could reach to 95% for a willingness-to-pay (WTP) of 2 GDPs when the compliance rate for transfer was > 80%. When 10% stage I tumors were detected by screening, the probability of the annual screening strategy being cost effective would be up to 95% for a WTP > 3 GDPs.

Conclusion

Annual community-based breast cancer screening was cost effective for a WTP of 3 GDP based on the incidence of breast cancer in Tianjin, China. Measures are needed to ensure performance indicators to a desirable level for the cost-effectiveness of breast cancer screening.

Electronic supplementary material

The online version of this article (10.1186/s12885-018-4168-1) contains supplementary material, which is available to authorized users.

The Importance of Model Structure in the Cost-Effectiveness Analysis of Primary Care Interventions for the Management of Hypertension

BACKGROUND

Management of hypertension can lead to significant reductions in blood pressure, thereby reducing the risk of cardiovascular disease. Modeling the course of cardiovascular disease is not without complications, and uncertainty surrounding the structure of a model will almost always arise once a choice of a model structure is defined.

OBJECTIVES

To provide a practical illustration of the impact on the results of cost-effectiveness of changing or adapting model structures in a previously published cost-utility analysis of a primary care intervention for the management of hypertension Targets and Self-Management for the Control of Blood Pressure in Stroke and at Risk Groups (TASMIN-SR).

METHODS

The case study assessed the structural uncertainty arising from model structure and from the exclusion of secondary events. Four alternative model structures were implemented. Long-term cost-effectiveness was estimated and the results compared with those from the TASMIN-SR model.

RESULTS

The main cost-effectiveness results obtained in the TASMIN-SR study did not change with the implementation of alternative model structures. Choice of model type was limited to a cohort Markov model, and because of the lack of epidemiological data, only model 4 captured structural uncertainty arising from the exclusion of secondary events in the case study model.

CONCLUSIONS

The results of this study indicate that the main conclusions drawn from the TASMIN-SR model of cost-effectiveness were robust to changes in model structure and the inclusion of secondary events. Even though one of the models produced results that were different to those of TASMIN-SR, the fact that the main conclusions were identical suggests that a more parsimonious model may have sufficed.

Comparison of Timed Automata with Discrete Event Simulation for Modeling of Biomarker-Based Treatment Decisions: An Illustration for Metastatic Castration-Resistant Prostate Cancer

BACKGROUND

With the advent of personalized medicine, the field of health economic modeling is being challenged and the use of patient-level dynamic modeling techniques might be required.

OBJECTIVES

To illustrate the usability of two such techniques, timed automata (TA) and discrete event simulation (DES), for modeling personalized treatment decisions.

METHODS

An early health technology assessment on the use of circulating tumor cells, compared with prostate-specific antigen and bone scintigraphy, to inform treatment decisions in metastatic castration-resistant prostate cancer was performed. Both modeling techniques were assessed quantitatively, in terms of intermediate outcomes (e.g., overtreatment) and health economic outcomes (e.g., net monetary benefit). Qualitatively, among others, model structure, agent interactions, data management (i.e., importing and exporting data), and model transparency were assessed.

RESULTS

Both models yielded realistic and similar intermediate and health economic outcomes. Overtreatment was reduced by 6.99 and 7.02 weeks by applying circulating tumor cell as a response marker at a net monetary benefit of -€1033 and -€1104 for the TA model and the DES model, respectively. Software-specific differences were observed regarding data management features and the support for statistical distributions, which were considered better for the DES software. Regarding method-specific differences, interactions were modeled more straightforward using TA, benefiting from its compositional model structure.

CONCLUSIONS

Both techniques prove suitable for modeling personalized treatment decisions, although DES would be preferred given the current software-specific limitations of TA. When these limitations are resolved, TA would be an interesting modeling alternative if interactions are key or its compositional structure is useful to manage multi-agent complex problems.

Model-Based Economic Evaluation of Treatments for Depression: A Systematic Literature Review

BACKGROUND

An increasing number of model-based studies that evaluate the cost effectiveness of treatments for depression are being published. These studies have different characteristics and use different simulation methods.

OBJECTIVE

We aimed to systematically review model-based studies evaluating the cost effectiveness of treatments for depression and examine which modelling technique is most appropriate for simulating the natural course of depression.

METHODS

The literature search was conducted in the databases PubMed, EMBASE and PsycInfo between 1 January 2002 and 1 October 2016. Studies were eligible if they used a health economic model with quality-adjusted life-years or disability-adjusted life-years as an outcome measure. Data related to various methodological characteristics were extracted from the included studies. The available modelling techniques were evaluated based on 11 predefined criteria.

RESULTS

This methodological review included 41 model-based studies, of which 21 used decision trees (DTs), 15 used cohort-based state-transition Markov models (CMMs), two used individual-based state-transition models (ISMs), and three used discrete-event simulation (DES) models. Just over half of the studies (54%) evaluated antidepressants compared with a control condition. The data sources, time horizons, cycle lengths, perspectives adopted and number of health states/events all varied widely between the included studies. DTs scored positively in four of the 11 criteria, CMMs in five, ISMs in six, and DES models in seven.

CONCLUSION

There were substantial methodological differences between the studies. Since the individual history of each patient is important for the prognosis of depression, DES and ISM simulation methods may be more appropriate than the others for a pragmatic representation of the course of depression. However, direct comparisons between the available modelling techniques are necessary to yield firm conclusions.

Markov Influence Diagrams

Markov influence diagrams (MIDs) are a new type of probabilistic graphical model that extends influence diagrams in the same way that Markov decision trees extend decision trees. They have been designed to build state-transition models, mainly in medicine, and perform cost-effectiveness analyses. Using a causal graph that may contain several variables per cycle, MIDs can model various patient characteristics without multiplying the number of states; in particular, they can represent the history of the patient without using tunnel states. OpenMarkov, an open-source tool, allows the decision analyst to build and evaluate MIDs—including cost-effectiveness analysis and several types of deterministic and probabilistic sensitivity analysis—with a graphical user interface, without writing any code. This way, MIDs can be used to easily build and evaluate complex models whose implementation as spreadsheets or decision trees would be cumbersome or unfeasible in practice. Furthermore, many problems that previously required discrete event simulation can be solved with MIDs; i.e., within the paradigm of state-transition models, in which many health economists feel more comfortable.

An empirical comparison of Markov cohort modeling and discrete event simulation in a capacity-constrained health care setting

OBJECTIVES

To empirically compare Markov cohort modeling (MM) and discrete event simulation (DES) with and without dynamic queuing (DQ) for cost-effectiveness (CE) analysis of a novel method of health services delivery where capacity constraints predominate.

METHODS

A common data-set comparing usual orthopedic care (UC) to an orthopedic physiotherapy screening clinic and multidisciplinary treatment service (OPSC) was used to develop a MM and a DES without (DES-no-DQ) and with DQ (DES-DQ). Model results were then compared in detail.

RESULTS

The MM predicted an incremental CE ratio (ICER) of $495 per additional quality-adjusted life-year (QALY) for OPSC over UC. The DES-no-DQ showed OPSC dominating UC; the DES-DQ generated an ICER of $2342 per QALY.

CONCLUSIONS

The MM and DES-no-DQ ICER estimates differed due to the MM having implicit delays built into its structure as a result of having fixed cycle lengths, which are not a feature of DES. The non-DQ models assume that queues are at a steady state. Conversely, queues in the DES-DQ develop flexibly with supply and demand for resources, in this case, leading to different estimates of resource use and CE. The choice of MM or DES (with or without DQ) would not alter the reimbursement of OPSC as it was highly cost-effective compared to UC in all analyses. However, the modeling method may influence decisions where ICERs are closer to the CE acceptability threshold, or where capacity constraints and DQ are important features of the system. In these cases, DES-DQ would be the preferred modeling technique to avoid incorrect resource allocation decisions.

Understanding the Effects of Competition for Constrained Colonoscopy Services with the Introduction of Population-level Colorectal Cancer Screening

BACKGROUND

Median wait times for gastroenterology services in Canada exceed consensus-recommended targets and have worsened substantially over the past decade. Meanwhile, efforts to control colorectal cancer have shifted their focus to screening asymptomatic, average-risk individuals. Along with increasing prevalence of colorectal cancer due to an aging population, screening programs are expected to add substantially to the existing burden on colonoscopy services, and create competition for limited services among individuals of varying risk. Failure to understand the effects of operational programmatic screening decisions may cause unintended harm to both screening participants and higher-risk patients, make inefficient use of limited health care resources, and ultimately hinder a program's success.

METHODS

We present a new simulation model (Simulation of Cancer Outcomes for Planning Exercises, or SCOPE) for colorectal cancer screening which, unlike many other colorectal cancer screening models, reflects the effects of competition for limited colonoscopy services between patient groups and can be used to guide planning to ensure adequate resource allocation. We include verification and validation results for the SCOPE model.

RESULTS

A discrete event simulation model was developed based on an epidemiological representation of colorectal cancer in a sample population. Colonoscopy service and screening modules were added to allow observation of screening scenarios and resource considerations. The model reproduces population-based data on prevalence of colorectal cancer by stage, and mortality by cause of death, age, and sex, and attendant demand and wait times for colonoscopy services.

CONCLUSIONS

The study model differs from existing screening models in that it explicitly considers the colonoscopy resource implications of screening activities and the impact of constrained resources on screening effectiveness.

Health Impact Assessment for Second-Hand Smoke Exposure in Germany--Quantifying Estimates for Ischaemic Heart Diseases, COPD, and Stroke

Evidence of the adverse health effects attributable to second-hand smoke (SHS) exposure is available. This study aims to quantify the impact of SHS exposure on ischaemic heart diseases (IHD), chronic obstructive pulmonary diseases (COPD), and stroke in Germany. Therefore, this study estimated and forecasted the morbidity for the three outcomes in the German population. Furthermore, a health impact assessment was performed using DYNAMO-HIA, which is a generic software tool applying a Markov model. Overall 687,254 IHD cases, 231,973 COPD cases, and 288,015 stroke cases were estimated to be attributable to SHS exposure in Germany for 2014. Under the assumption that the population prevalence of these diseases and the prevalence of SHS exposure remain constant, the total number of cases will increase due to demographic aging. Assuming a total eradication of SHS exposure beginning in 2014 leads to an estimated reduction of 50% in cases, compared to the reference scenario in 2040 for all three diseases. The results highlight the relevance of SHS exposure because it affects several chronic disease conditions and has a major impact on the population’s health. Therefore, public health campaigns to protect non-smokers are urgently needed.

How Do Diabetes Models Measure Up? A Review of Diabetes Economic Models and ADA Guidelines

Introduction: Economic models and computer simulation models have been used for assessing short-term cost-effectiveness of interventions and modelling long-term outcomes and costs. Several guidelines and checklists have been published to improve the methods and reporting. This article presents an overview of published diabetes models with a focus on how well the models are described in relation to the considerations described by the American Diabetes Association (ADA) guidelines. Methods: Relevant electronic databases and National Institute for Health and Care Excellence (NICE) guidelines were searched in December 2012. Studies were included in the review if they estimated lifetime outcomes for patients with type 1 or type 2 diabetes. Only unique models, and only the original papers were included in the review. If additional information was reported in subsequent or paired articles, then additional citations were included. References and forward citations of relevant articles, including the previous systematic reviews were searched using a similar method to pearl growing. Four principal areas were included in the ADA guidance reporting for models: transparency, validation, uncertainty, and diabetes specific criteria. Results: A total of 19 models were included. Twelve models investigated type 2 diabetes, two developed type 1 models, two created separate models for type 1 and type 2, and three developed joint type 1 and type 2 models. Most models were developed in the United States, United Kingdom, Europe or Canada. Later models use data or methods from earlier models for development or validation. There are four main types of models: Markov-based cohort, Markov-based microsimulations, discrete-time microsimulations, and continuous time differential equations. All models were long-term diabetes models incorporating a wide range of compilations from various organ systems. In early diabetes modelling, before the ADA guidelines were published, most models did not include descriptions of all the diabetes specific components of the ADA guidelines but this improved significantly by 2004. Conclusion: A clear, descriptive short summary of the model was often lacking. Descriptions of model validation and uncertainty were the most poorly reported of the four main areas, but there exist conferences focussing specifically on the issue of validation. Interdependence between the complications was the least well incorporated or reported of the diabetes-specific criterion.

Dealing with Time in Health Economic Evaluation: Methodological Issues and Recommendations for Practice

Time is an important aspect of health economic evaluation, as the timing and duration of clinical events, healthcare interventions and their consequences all affect estimated costs and effects. These issues should be reflected in the design of health economic models. This article considers three important aspects of time in modelling: (1) which cohorts to simulate and how far into the future to extend the analysis; (2) the simulation of time, including the difference between discrete-time and continuous-time models, cycle lengths, and converting rates and probabilities; and (3) discounting future costs and effects to their present values. We provide a methodological overview of these issues and make recommendations to help inform both the conduct of cost-effectiveness analyses and the interpretation of their results. For choosing which cohorts to simulate and how many, we suggest analysts carefully assess potential reasons for variation in cost effectiveness between cohorts and the feasibility of subgroup-specific recommendations. For the simulation of time, we recommend using short cycles or continuous-time models to avoid biases and the need for half-cycle corrections, and provide advice on the correct conversion of transition probabilities in state transition models. Finally, for discounting, analysts should not only follow current guidance and report how discounting was conducted, especially in the case of differential discounting, but also seek to develop an understanding of its rationale. Our overall recommendations are that analysts explicitly state and justify their modelling choices regarding time and consider how alternative choices may impact on results.

A systematic review and economic evaluation of intraoperative tests [RD-100i one-step nucleic acid amplification (OSNA) system and Metasin test] for detecting sentinel lymph node metastases in breast cancer

BACKGROUND

In breast cancer patients, sentinel lymph node biopsy is carried out at the same time as the removal of the primary tumour to postoperatively test with histopathology for regional metastases in the sentinel lymph node. Those patients with positive test results are then operated on 2-4 weeks after primary surgery to remove the lymph nodes from the axilla (axillary lymph node dissection, ALND). New molecular tests RD-100i [one-step nucleic acid amplification (OSNA); based on messenger RNA amplification to identify the cytokeratin-19 (CK19) gene marker] (Sysmex, Norderstedt, Germany) and Metasin (using the CK19 and mammaglobin gene markers) (Cellular Pathology, Princess Alexandra Hospital NHS Trust, Harlow, UK) are intended to provide an intraoperative diagnosis, thereby avoiding the need for postoperative histopathology and, in positive cases, a second operation for ALND.

OBJECTIVE

To evaluate the clinical effectiveness and cost-effectiveness of using OSNA and Metasin in the NHS in England for the intraoperative diagnosis of sentinel lymph nodes metastases, compared with postoperative histopathology, the current standard.

DATA SOURCES

Electronic databases including MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, EMBASE, The Cochrane Library and the Health Economic Evaluations Database as well as clinical trial registries, grey literature and conference proceedings were searched up to July 2012.

REVIEW METHODS

A systematic review of the evidence was carried out using standard methods. Single-gate studies were used to estimate the accuracy of OSNA with histopathology as the reference standard. The cost-effectiveness analysis adapted an existing simulation model of the long-term costs and health implications of early breast cancer diagnostic outcomes. The model accounted for the costs of an extended first operation with intraoperative testing, the loss of health-related quality of life (disutility) from waiting for postoperative test results, disutility and costs of a second operation, and long-term costs and disutility from lymphoedema related to ALND, adjuvant therapy, locoregional recurrence and metastatic recurrence.

RESULTS

A total of 724 references were identified in the searches, of which 17 studies assessing test accuracy were included in the review, 15 on OSNA and two on Metasin. Both Metasin studies were unpublished. OSNA sensitivity of 84.5% [95% confidence interval (CI) 74.7% to 91.0%] and specificity of 91.8% (95% CI 87.8% to 94.6%) for patient nodal status were estimated in a meta-analysis of five studies [unadjusted for tissue allocation bias (TAB)]. At these values and a 20% node-positive rate, OSNA resulted in lifetime discounted cost-savings of £498 and a quality-adjusted life-year (QALY) loss of 0.048 relative to histopathology, that is, £4324 saved per QALY lost. The most favourable plausible scenario for OSNA in terms of the node-positive rate (range 10-40%), diagnostic accuracy values (91.3% sensitivity and 94.2% specificity, from three reports that adjusted for TAB), the costs of histopathology, OSNA and second surgery, and long-term costs and utilities resulted in a maximum saving per QALY lost of £10,500; OSNA sensitivity and specificity would need to be ≥ 95% for this figure to be ≥ £20,000.

LIMITATIONS

There is limited evidence on the diagnostic test accuracy of intraoperative tests. The quality of information on costs of resource utilisation during the diagnostic pathway is low and no evidence exists on the disutility of waiting for a second surgery. No comparative studies exist that report clinical outcomes of intraoperative diagnostic tests. These knowledge gaps have more influence on the decision than current uncertainty in the performance of postoperative histopathology in standard practice.

CONCLUSIONS

One-step nucleic acid amplification is not cost-effective for the intraoperative diagnosis of sentinel lymph node metastases. OSNA is less accurate than histopathology and the consequent loss of health benefits in this patient group is not compensated for by health gains elsewhere in the health system that may be obtained with the cost-savings made. The evidence on Metasin is insufficient to evaluate its cost-effectiveness.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42012002889.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.

Declining liver graft quality threatens the future of liver transplantation in the United States

National liver transplantation (LT) volume has declined since 2006, in part because of worsening donor organ quality. Trends that degrade organ quality are expected to continue over the next 2 decades. We used the United Network for Organ Sharing (UNOS) database to inform a 20-year discrete event simulation estimating LT volume from 2010 to 2030. Data to inform the model were obtained from deceased organ donors between 2000 and 2009. If donor liver utilization practices remain constant, utilization will fall from 78% to 44% by 2030, resulting in 2230 fewer LTs. If transplant centers increase their risk tolerance for marginal grafts, utilization would decrease to 48%. The institution of "opt-out" organ donation policies to increase the donor pool would still result in 1380 to 1866 fewer transplants. Ex vivo perfusion techniques that increase the use of marginal donor livers may stabilize LT volume. Otherwise, the number of LTs in the United States will decrease substantially over the next 15 years. In conclusion, the transplant community will need to accept inferior grafts and potentially worse posttransplant outcomes and/or develop new strategies for increasing organ donation and utilization in order to maintain the number of LTs at the current level.

Economics of stratified medicine in rheumatoid arthritis

Clinically relevant examples of stratified medicine are available for patients with rheumatoid arthritis (RA). The aim of this study was to understand the current economic evidence for stratified medicine in RA. Two systematic reviews were conducted to identify: (1) all economic evaluations of stratified treatments for rheumatoid arthritis, or those which have used a subgroup analysis, and (2) all stated preference studies of treatments for rheumatoid arthritis. Ten economic evaluations of stratified treatments for RA, 38 economic evaluations including with a subgroup analysis and eight stated preference studies were identified. There was some evidence to support that stratified approaches to treating a patient with RA may be cost-effective. However, there remain key gaps in the economic evidence base needed to support the introduction of stratified medicine in RA into healthcare systems and considerable uncertainty about how proposed stratified approaches will impact future patient preferences, outcomes and costs when used in routine practice.

A Systematic Review of Research Guidelines in Decision-Analytic Modeling

BACKGROUND

Decision-analytic modeling (DAM) has been increasingly used to aid decision making in health care. The growing use of modeling in economic evaluations has led to increased scrutiny of the methods used.

OBJECTIVE

The objective of this study was to perform a systematic review to identify and critically assess good practice guidelines, with particular emphasis on contemporary developments.

METHODS

A systematic review of English language articles was undertaken to identify articles presenting guidance for good practice in DAM in the evaluation of health care. The inclusion criteria were articles providing guidance or criteria against which to assess good practice in DAM and studies providing criteria or elements for good practice in some areas of DAM. The review covered the period January 1990 to March 2014 and included the following electronic bibliographic databases: Cochrane Library, Cochrane Methodology Register and Health Technology Assessment, NHS Economic Evaluation Database, MEDLINE, and PubMed (Embase). Additional studies were identified by searching references.

RESULTS

Thirty-three articles were included in this review. A practical five-dimension framework was developed that describe the key elements of good research practice that should be considered and reported to increase the credibility of results obtained from DAM in the evaluation of health care.

CONCLUSIONS

This study is the first to critically review all available guidelines and statements of good practice in DAM since 2006. The development of good practice guidelines is an ongoing process, and important efforts have been made to identify what is good practice and to keep these guidelines up to date.

Do different decision-analytic modeling approaches produce different results? A systematic review of cross-validation studies

When choosing a modeling approach for health economic evaluation, certain criteria are often considered (e.g., population resolution, interactivity, time advancement mechanism, resource constraints). However, whether these criteria and their associated modeling approach impacts results remain poorly understood. A systematic review was conducted to identify cross-validation studies (i.e., modeling a problem using different approaches with the same body of evidence) to offer insight on this topic. With respect to population resolution, reviewed studies suggested that both aggregate- and individual-level models will generate comparable results, although a practical trade-off exists between validity and feasibility. In terms of interactivity, infectious-disease models consistently showed that, depending on the assumptions regarding probability of disease exposure, dynamic and static models may produce dissimilar results with opposing policy recommendations. Empirical evidence on the remaining criteria is limited. Greater discussion will therefore be necessary to promote a deeper understanding of the benefits and limits to each modeling approach.

Protocol for the economic evaluation of the diarrhea alleviation through zinc and oral rehydration salt therapy at scale through private and public providers in rural Gujarat and Uttar Pradesh, India

BACKGROUND

Child diarrhea persists as a leading public health problem in India despite evidence supporting zinc and low osmolarity oral rehydration salts as effective treatments. Across 2 years in 2010-2013, the Diarrhea Alleviation using Zinc and Oral Rehydration Salts Therapy (DAZT) program was implemented to operationalize delivery of these interventions at scale through private and public sector providers in rural Gujarat and Uttar Pradesh, India.

METHODS/DESIGN

This study evaluates the cost-effectiveness of DAZT program activities relative to status quo conditions existing before the study, comparing a Monte Carlo simulation method with net-benefit regression, discussing the strengths and weaknesses of each approach. A control group was not included in the 'before and after' study design as zinc has proven effectiveness for diarrhea treatment. Costs will be calculated using a societal perspective including program implementation and household out-of-pocket payments for care seeking, as well as estimates of wages lost. Outcomes will be measured in terms of episodes averted in net-benefit regression and in terms of the years of life lost component of disability-adjusted life years in the method based on Monte Carlo simulation. The Lives Saved Tool will be used to model anticipated changes in mortality over time and deaths averted based on incremental changes in coverage of oral rehydration salts and zinc. Data will derive from cross-sectional surveys at the start, midpoint, and endpoint of the program. In addition, Lives Saved Tool (LiST) projections will be used to define the reference case value for the ceiling ratio in terms of natural units.

DISCUSSION

This study will be useful both in its application to an economic evaluation of a public health program in its implementation phase but also in its comparison of two methodological approaches to cost-effectiveness analysis. Both policy recommendations and methodological lessons learned will be discussed, recognizing the limitations in drawing strong policy conclusions due to the uncontrolled study design. It is expected that this protocol will be useful to researchers planning what method to use for the evaluation of similar before and after studies.

Modelling outcomes of complex treatment strategies following a clinical guideline for treatment decisions in patients with rheumatoid arthritis

BACKGROUND

Management of rheumatoid arthritis (RA) is characterised by a sequence of disease-modifying antirheumatic drugs (DMARDs) and biological response modifiers (BRMs). In most of the Western countries, the drug sequences are determined based on disease activity and treatment history of the patients. A model for realistic patient outcomes should reflect the treatment pathways relevant for patients with specific characteristics.

OBJECTIVE

This study aimed at developing a model that could simulate long-term patient outcomes and cost effectiveness of treatment strategies with and without inclusion of BRMs following a clinical guideline for treatment decisions.

METHODS

Discrete event simulation taking into account patient characteristics and treatment history was used for model development. Treatment effect on disease activity, costs, health utilities and times to events were estimated using Dutch observational studies. Long-term progression of physical functioning was quantified using a linear mixed-effects model. Costs and health utilities were estimated using two-part models. The treatment strategy recommended by the Dutch Society for Rheumatology where both DMARDs and BRMs were available (Strategy 2) was compared with the treatment strategy without BRMs (Strategy 1). Ten thousand theoretical patients were tracked individually until death. In the probabilistic sensitivity analysis, Monte Carlo simulations were performed with 1,000 sets of parameters sampled from appropriate probability distributions.

RESULTS

The simulated changes over time in disease activity and physical functioning were plausible. The incremental cost per quality-adjusted life-year gained of Strategy 2 compared with Strategy 1 was <euro>124,011. At a willingness-to-pay threshold higher than <euro>119,167, Strategy 2 dominated Strategy 1 in terms of cost effectiveness but the probability that the Strategy 2 is cost effective never exceeded 0.87.

CONCLUSIONS

It is possible to model the outcomes of complex treatment strategies based on a clinical guideline for the management of RA. Following the Dutch guideline and using real-life data, inclusion of BRMs in the treatment strategy for RA appeared to be less favourable in our model than in most of the existing models that compared drug sequences independent of patient characteristics and used data from randomised controlled clinical trials. Despite complexity and demand for extensive data, our modelling approach can help to identify the knowledge gaps in clinical guidelines for RA management and priorities for future research.

A systematic and critical review of model-based economic evaluations of pharmacotherapeutics in patients with bipolar disorder

BACKGROUND

Bipolar disorder (BD) is a chronic and relapsing mental illness with a considerable health-related and economic burden. The primary goal of pharmacotherapeutics for BD is to improve patients' well-being. The use of decision-analytic models is key in assessing the added value of the pharmacotherapeutics aimed at treating the illness, but concerns have been expressed about the appropriateness of different modelling techniques and about the transparency in the reporting of economic evaluations.

OBJECTIVES

This paper aimed to identify and critically appraise published model-based economic evaluations of pharmacotherapeutics in BD patients.

METHODS

A systematic review combining common terms for BD and economic evaluation was conducted in MEDLINE, EMBASE, PSYCINFO and ECONLIT. Studies identified were summarised and critically appraised in terms of the use of modelling technique, model structure and data sources. Considering the prognosis and management of BD, the possible benefits and limitations of each modelling technique are discussed.

RESULTS

Fourteen studies were identified using model-based economic evaluations of pharmacotherapeutics in BD patients. Of these 14 studies, nine used Markov, three used discrete-event simulation (DES) and two used decision-tree models. Most of the studies (n = 11) did not include the rationale for the choice of modelling technique undertaken. Half of the studies did not include the risk of mortality. Surprisingly, no study considered the risk of having a mixed bipolar episode.

CONCLUSIONS

This review identified various modelling issues that could potentially reduce the comparability of one pharmacotherapeutic intervention with another. Better use and reporting of the modelling techniques in the future studies are essential. DES modelling appears to be a flexible and comprehensive technique for evaluating the comparability of BD treatment options because of its greater flexibility of depicting the disease progression over time. However, depending on the research question, modelling techniques other than DES might also be appropriate in some cases.

When to use discrete event simulation (DES) for the economic evaluation of health technologies? A review and critique of the costs and benefits of DES

Modelling in economic evaluation is an unavoidable fact of life. Cohort-based state transition models are most common, though discrete event simulation (DES) is increasingly being used to implement more complex model structures. The benefits of DES relate to the greater flexibility around the implementation and population of complex models, which may provide more accurate or valid estimates of the incremental costs and benefits of alternative health technologies. The costs of DES relate to the time and expertise required to implement and review complex models, when perhaps a simpler model would suffice. The costs are not borne solely by the analyst, but also by reviewers. In particular, modelled economic evaluations are often submitted to support reimbursement decisions for new technologies, for which detailed model reviews are generally undertaken on behalf of the funding body. This paper reports the results from a review of published DES-based economic evaluations. Factors underlying the use of DES were defined, and the characteristics of applied models were considered, to inform options for assessing the potential benefits of DES in relation to each factor. Four broad factors underlying the use of DES were identified: baseline heterogeneity, continuous disease markers, time varying event rates, and the influence of prior events on subsequent event rates. If relevant, individual-level data are available, representation of the four factors is likely to improve model validity, and it is possible to assess the importance of their representation in individual cases. A thorough model performance evaluation is required to overcome the costs of DES from the users' perspective, but few of the reviewed DES models reported such a process. More generally, further direct, empirical comparisons of complex models with simpler models would better inform the benefits of DES to implement more complex models, and the circumstances in which such benefits are most likely.

Markov modeling and discrete event simulation in health care: a systematic comparison

OBJECTIVES

The aim of this study was to assess if the use of Markov modeling (MM) or discrete event simulation (DES) for cost-effectiveness analysis (CEA) may alter healthcare resource allocation decisions.

METHODS

A systematic literature search and review of empirical and non-empirical studies comparing MM and DES techniques used in the CEA of healthcare technologies was conducted.

RESULTS

Twenty-two pertinent publications were identified. Two publications compared MM and DES models empirically, one presented a conceptual DES and MM, two described a DES consensus guideline, and seventeen drew comparisons between MM and DES through the authors' experience. The primary advantages described for DES over MM were the ability to model queuing for limited resources, capture individual patient histories, accommodate complexity and uncertainty, represent time flexibly, model competing risks, and accommodate multiple events simultaneously. The disadvantages of DES over MM were the potential for model overspecification, increased data requirements, specialized expensive software, and increased model development, validation, and computational time.

CONCLUSIONS

Where individual patient history is an important driver of future events an individual patient simulation technique like DES may be preferred over MM. Where supply shortages, subsequent queuing, and diversion of patients through other pathways in the healthcare system are likely to be drivers of cost-effectiveness, DES modeling methods may provide decision makers with more accurate information on which to base resource allocation decisions. Where these are not major features of the cost-effectiveness question, MM remains an efficient, easily validated, parsimonious, and accurate method of determining the cost-effectiveness of new healthcare interventions.

The development and validation of a decision-analytic model representing the full disease course of acute myeloid leukemia

BACKGROUND

The treatment of acute myeloid leukemia (AML) is moving towards personalized medicine. However, due to the low incidence of AML, it is not always feasible to evaluate the cost-effectiveness of personalized medicine using clinical trials. Decision analytic models provide an alternative data source.

OBJECTIVE

The aim of this study was to develop and validate a decision analytic model that represents the full disease course of AML.

METHODS

We used a micro simulation with discrete event components to incorporate both patient and disease heterogeneity. Input parameters were calculated from patient-level data. Two hematologists critically evaluated the model to ensure face validity. Internal and external validity was tested by comparing complete remission (CR) rates and survival outcomes of the model with original data, other clinical trials and a population-based study.

RESULTS

No significant differences in patient and treatment characteristics, CR rate, 5-year overall and disease-free survival were found between the simulated and original data. External validation showed no significant differences in survival between simulated data and other clinical trials. However, differences existed between the simulated data and a population-based study.

CONCLUSIONS

The model developed in this study is proved to be valid for analysis of an AML population participating in a clinical trial. The generalizability of the model to a broader patient population has not been proven yet. Further research is needed to identify differences between the clinical trial population and other AML patients and to incorporate these differences in the model.

A mathematical approach for evaluating Markov models in continuous time without discrete-event simulation

Markov models are a simple and powerful tool for analyzing the health and economic effects of health care interventions. These models are usually evaluated in discrete time using cohort analysis. The use of discrete time assumes that changes in health states occur only at the end of a cycle period. Discrete-time Markov models only approximate the process of disease progression, as clinical events typically occur in continuous time. The approximation can yield biased cost-effectiveness estimates for Markov models with long cycle periods and if no half-cycle correction is made. The purpose of this article is to present an overview of methods for evaluating Markov models in continuous time. These methods use mathematical results from stochastic process theory and control theory. The methods are illustrated using an applied example on the cost-effectiveness of antiviral therapy for chronic hepatitis B. The main result is a mathematical solution for the expected time spent in each state in a continuous-time Markov model. It is shown how this solution can account for age-dependent transition rates and discounting of costs and health effects, and how the concept of tunnel states can be used to account for transition rates that depend on the time spent in a state. The applied example shows that the continuous-time model yields more accurate results than the discrete-time model but does not require much computation time and is easily implemented. In conclusion, continuous-time Markov models are a feasible alternative to cohort analysis and can offer several theoretical and practical advantages.

Continuous time simulation and discretized models for cost-effectiveness analysis

The design of decision-analytic models for cost-effectiveness analysis has been the subject of discussion. The current work addresses this issue by noting that, when time is to be explicitly modelled, we need to represent phenomena occurring in continuous time. Models evaluated in continuous time may not have closed-form solutions, and in this case, two approximations can be used: simulation models in continuous time and discretized models at the aggregate level. Stylized examples were set up where both approximations could be implemented. These aimed to illustrate determinants of the use of the two approximations: cycle length and precision, the use of continuity corrections in discretized models and the discretization of rates into probabilities. The examples were also used to explore the impact of the approximations not only in terms of absolute survival but also cost effectiveness and incremental comparisons. Discretized models better approximate continuous time results if lower cycle lengths are used. Continuous time simulation models are inherently stochastic, and the precision of the results is determined by the simulation sample size. The use of continuity corrections in discretized models allows the use of greater cycle lengths, producing no significant bias from the discretization. How the process is discretized (the conversion of rates into probabilities) is key. Results show that appropriate discretization coupled with the use of a continuity correction produces results unbiased for higher cycle lengths. Alternative methods of discretization are less efficient, i.e. lower cycle lengths are needed to obtain unbiased results. The developed work showed the importance of acknowledging bias in estimating cost effectiveness. When the alternative approximations can be applied, we argue that it is preferable to implement a cohort discretized model rather than a simulation model in continuous time. In practice, however, it may not be possible to represent the decision problem by any conventionally defined discretized model, in which case other model designs need to be applied, e.g. a simulation model.

A proposed model for economic evaluations of major depressive disorder

In countries like UK and Australia, the comparability of model-based analyses is an essential aspect of reimbursement decisions for new pharmaceuticals, medical services and technologies. Within disease areas, the use of models with alternative structures, type of modelling techniques and/or data sources for common parameters reduces the comparability of evaluations of alternative technologies for the same condition. The aim of this paper is to propose a decision analytic model to evaluate long-term costs and benefits of alternative management options in patients with depression. The structure of the proposed model is based on the natural history of depression and includes clinical events that are important from both clinical and economic perspectives. Considering its greater flexibility with respect to handling time, discrete event simulation (DES) is an appropriate simulation platform for modelling studies of depression. We argue that the proposed model can be used as a reference model in model-based studies of depression improving the quality and comparability of studies.

A critical review of model-based economic studies of depression: modelling techniques, model structure and data sources

Depression is the most common mental health disorder and is recognized as a chronic disease characterized by multiple acute episodes/relapses. Although modelling techniques play an increasingly important role in the economic evaluation of depression interventions, comparatively little attention has been paid to issues around modelling studies with a focus on potential biases. This, however, is important as different modelling approaches, variations in model structure and input parameters may produce different results, and hence different policy decisions. This paper presents a critical review of literature on recently published model-based cost-utility studies of depression. Taking depression as an illustrative example, through this review, we discuss a number of specific issues in relation to the use of decision-analytic models including the type of modelling techniques, structure of models and data sources. The potential benefits and limitations of each modelling technique are discussed and factors influencing the choice of modelling techniques are addressed. This review found that model-based studies of depression used various simulation techniques. We note that a discrete-event simulation may be the preferred technique for the economic evaluation of depression due to the greater flexibility with respect to handling time compared with other individual-based modelling techniques. Considering prognosis and management of depression, the structure of the reviewed models are discussed. We argue that a few reviewed models did not include some important structural aspects such as the possibility of relapse or the increased risk of suicide in patients with depression. Finally, the appropriateness of data sources used to estimate input parameters with a focus on transition probabilities is addressed. We argue that the above issues can potentially bias results and reduce the comparability of economic evaluations.

Cost effectiveness of quetiapine in patients with acute bipolar depression and in maintenance treatment after an acute depressive episode

BACKGROUND

Bipolar disorder has a significant impact upon a patient's quality of life, imposing a considerable economic burden on the individual, family members and society as a whole. Several medications are indicated for the acute treatment of mania and depression associated with bipolar disorder as well as for maintenance therapy; however, these have varying efficacy, tolerability and costs.

OBJECTIVE

The objective of this study was to develop a new discrete-event simulation model to analyse the long-term consequences of pharmacological therapy for the management of bipolar I and II disorders (acute treatment of episodes of mania and depression as well as maintenance therapy).

METHODS

Probabilities of remission and relapse were obtained from clinical trial data and meta-analyses. Costs (year 2011 values) were assessed from a UK healthcare payer's perspective, and included pharmacological therapy and resource use associated with the treatment of mood events and selected adverse events. The health effects were measured in terms of QALYs.

RESULTS

For a patient starting with acute depression or in remission at 40 years of age (which was the average age in the clinical trials), quetiapine 300 mg/day was a cost-effective strategy compared with olanzapine 15 mg/day over a 5-year time frame. With acute bipolar depression as a starting episode, the 5-year medical costs were £323 higher and QALYs were 0.038 higher for quetiapine compared with olanzapine, corresponding to a cost-effectiveness ratio of £8600 per QALY gained.

CONCLUSION

Compared with olanzapine, the results suggest that quetiapine is cost effective as a maintenance treatment for bipolar depression.

DYNAMO-HIA--a Dynamic Modeling tool for generic Health Impact Assessments

BACKGROUND

Currently, no standard tool is publicly available that allows researchers or policy-makers to quantify the impact of policies using epidemiological evidence within the causal framework of Health Impact Assessment (HIA). A standard tool should comply with three technical criteria (real-life population, dynamic projection, explicit risk-factor states) and three usability criteria (modest data requirements, rich model output, generally accessible) to be useful in the applied setting of HIA. With DYNAMO-HIA (Dynamic Modeling for Health Impact Assessment), we introduce such a generic software tool specifically designed to facilitate quantification in the assessment of the health impacts of policies.

METHODS AND RESULTS

DYNAMO-HIA quantifies the impact of user-specified risk-factor changes on multiple diseases and in turn on overall population health, comparing one reference scenario with one or more intervention scenarios. The Markov-based modeling approach allows for explicit risk-factor states and simulation of a real-life population. A built-in parameter estimation module ensures that only standard population-level epidemiological evidence is required, i.e. data on incidence, prevalence, relative risks, and mortality. DYNAMO-HIA provides a rich output of summary measures--e.g. life expectancy and disease-free life expectancy--and detailed data--e.g. prevalences and mortality/survival rates--by age, sex, and risk-factor status over time. DYNAMO-HIA is controlled via a graphical user interface and is publicly available from the internet, ensuring general accessibility. We illustrate the use of DYNAMO-HIA with two example applications: a policy causing an overall increase in alcohol consumption and quantifying the disease-burden of smoking.

CONCLUSION

By combining modest data needs with general accessibility and user friendliness within the causal framework of HIA, DYNAMO-HIA is a potential standard tool for health impact assessment based on epidemiologic evidence.

Population- versus cohort-based modelling approaches

While no single type of model can provide adequate answers under all circumstances, any modelling endeavour should incorporate three fundamental considerations in any decision-making question: the target population, the disease and the intervention characteristics. A target population is likely to be characterized by various types of heterogeneity and a dynamic evolution over time. It is therefore important to adequately capture these population effects on the results of a model. There are essentially two different approaches in modelling a population over time: a cohort-based approach and a population-based approach. In a cohort-based model, a closed group of individuals who have at least one specific characteristic or experience in common over a defined period of time is run through a state transition process. The cohort is generally composed of a hypothetical number of representative or 'average' individuals (i.e. the target population is considered to be a homogeneous group). The population-based approach projects the evolution of the estimated prevalent target population and intends to reflect as much as possible the demographic, epidemiological and clinical characteristics of the prevalent target population relevant for the decision problem. A cohort-based approach is generally used in most published healthcare decision models. However, this choice is rarely discussed by modellers. In this article, we challenge this assumption. To address the underlying decision problem, we affirm it is crucial that modellers consider the characteristics of the target population. Then, they could opt for using the most appropriate approach. Decision makers should also understand the impact on the results of both types of models in order to make informed healthcare decisions.

Breast cancer-related lymphedema: comparing direct costs of a prospective surveillance model and a traditional model of care

Secondary prevention involves monitoring and screening to prevent negative sequelae from chronic diseases such as cancer. Breast cancer treatment sequelae, such as lymphedema, may occur early or late and often negatively affect function. Secondary prevention through prospective physical therapy surveillance aids in early identification and treatment of breast cancer-related lymphedema (BCRL). Early intervention may reduce the need for intensive rehabilitation and may be cost saving. This perspective article compares a prospective surveillance model with a traditional model of impairment-based care and examines direct treatment costs associated with each program. Intervention and supply costs were estimated based on the Medicare 2009 physician fee schedule for 2 groups: (1) a prospective surveillance model group (PSM group) and (2) a traditional model group (TM group). The PSM group comprised all women with breast cancer who were receiving interval prospective surveillance, assuming that one third would develop early-stage BCRL. The prospective surveillance model includes the cost of screening all women plus the cost of intervention for early-stage BCRL. The TM group comprised women referred for BCRL treatment using a traditional model of referral based on late-stage lymphedema. The traditional model cost includes the direct cost of treating patients with advanced-stage lymphedema. The cost to manage early-stage BCRL per patient per year using a prospective surveillance model is $636.19. The cost to manage late-stage BCRL per patient per year using a traditional model is $3,124.92. The prospective surveillance model is emerging as the standard of care in breast cancer treatment and is a potential cost-saving mechanism for BCRL treatment. Further analysis of indirect costs and utility is necessary to assess cost-effectiveness. A shift in the paradigm of physical therapy toward a prospective surveillance model is warranted.