Cost-Effectiveness of Sacubitril/Valsartan in Germany: An Application of the Efficiency Frontier.

BACKGROUND: To assess the cost-effectiveness of new treatments in Germany, the efficiency frontier (EF) method has been developed. We compared the cost-effectiveness analysis using international standards and the German methodology, using the heart failure drug sacubitril/valsartan as an example. METHODS: A previously developed Markov model was adapted to include 4 treatment options: no treatment, enalapril, candesartan, and sacubitril/valsartan. The internationally used incremental cost-effectiveness ratio (ICER) was calculated, as well as cost-effectiveness acceptability curves. Additionally, EFs, net monetary benefits (NMBs), and price-acceptability curves were created according to German guidelines. All analyses were performed from the perspective of the German Statutory Health Insurance. RESULTS: The base-case ICER for sacubitril/valsartan compared to enalapril is €19 300/quality-adjusted life-year. On the cost-effectiveness acceptability curve, sacubitril/valsartan is most likely to be cost-effective, out of all included comparators, from a hypothetical willingness-to-pay threshold of €18 250/quality-adjusted life-year onward. No EF could be constructed for the base case. Taking the uncertainty of the input parameters into account for the probabilistic sensitivity analysis, a NMB of around -€14 000 was calculated, depending on the outcome considered, with the NMB being zero at a daily price for sacubitril/valsartan ranging from €1.52 to €1.67. CONCLUSION: We calculated an ICER for Germany, comparable to previously published cost-effectiveness analyses for Europe, which widely concluded sacubitril/valsartan to be cost-effective. Using the German EF approach, a considerable discount needs to be applied before sacubitril/valsartan can be considered cost-effective.

An Economic Evaluation of Sacubitril/Valsartan for Heart Failure Patients in the Netherlands.

BACKGROUND: In September 2014, the PARADIGM-HF trial showed the heart failure drug combination sacubitril/valsartan to be superior to enalapril for patients with a reduced ejection fraction. OBJECTIVES: To determine the incremental cost-effectiveness of sacubitril/valsartan compared with enalapril in the Netherlands using the clinical data from the PARADIGM-HF trial. METHODS: To compare sacubitril/valsartan and enalapril in a cost-effectiveness study, a Markov model was developed using the effectiveness data from the PARADIGM-HF trial. A health care payer's perspective was applied in the economic evaluation. The developed model was used to evaluate the cost-effectiveness for sacubitril/valsartan at different per diem prices. RESULTS: The base-case analysis showed that sacubitril/valsartan can be cost-effective at maximum daily costs of €5.50 and €14.14 considering willingness-to-pay thresholds of €20,000 and €50,000 per quality-adjusted life-year (QALY), respectively. Sensitivity analysis demonstrated the robustness of the model, identifying only the price of sacubitril/valsartan and the mortality within the sacubitril/valsartan group as significant drivers of the cost-effectiveness ratio. Sacubitril/valsartan was cost-effective at a willingness-to-pay threshold of €20,000 per QALY (€50,000 per QALY) in more than 80% of the replications with certainty at the price point of €3 (€10). CONCLUSIONS: Sacubitril/valsartan can be considered a cost-effective treatment at a daily price of €5.25. Unless priced lower than enalapril (<€0.045 per day), sacubitril/valsartan is very unlikely to be cost-saving/dominant.

A New Statistical Method to Determine the Degree of Validity of Health Economic Model Outcomes against Empirical Data.

BACKGROUND: The validation of health economic (HE) model outcomes against empirical data is of key importance. Although statistical testing seems applicable, guidelines for the validation of HE models lack guidance on statistical validation, and actual validation efforts often present subjective judgment of graphs and point estimates. OBJECTIVES: To discuss the applicability of existing validation techniques and to present a new method for quantifying the degrees of validity statistically, which is useful for decision makers. METHODS: A new Bayesian method is proposed to determine how well HE model outcomes compare with empirical data. Validity is based on a pre-established accuracy interval in which the model outcomes should fall. The method uses the outcomes of a probabilistic sensitivity analysis and results in a posterior distribution around the probability that HE model outcomes can be regarded as valid. RESULTS: We use a published diabetes model (Modelling Integrated Care for Diabetes based on Observational data) to validate the outcome "number of patients who are on dialysis or with end-stage renal disease." Results indicate that a high probability of a valid outcome is associated with relatively wide accuracy intervals. In particular, 25% deviation from the observed outcome implied approximately 60% expected validity. CONCLUSIONS: Current practice in HE model validation can be improved by using an alternative method based on assessing whether the model outcomes fit to empirical data at a predefined level of accuracy. This method has the advantage of assessing both model bias and parameter uncertainty and resulting in a quantitative measure of the degree of validity that penalizes models predicting the mean of an outcome correctly but with overly wide credible intervals.

Higher Adherence During Reimbursement of Pharmacological Smoking Cessation Treatments.

INTRODUCTION: In the Netherlands, pharmacologic Smoking Cessation Treatments (pSCTs) were reimbursed in 2011. In 2012 the reimbursement was discontinued. As of 2013, pSCTs were again reimbursed, provided they are accompanied by behavioral counseling. The aim of this article is to assess the impact of changes in reimbursement policy on use of-and adherence to-pSCTs. METHODS: A retrospective dispensing database analysis was performed on real-world observational data (2010-2013) from the Netherlands. Data on use and adherence was collected, in patients who were dispensed bupropion or varenicline in community pharmacies for the first time. Using the InterActionDataBase (iadb.nl), adherence per patient that initiated varenicline or bupropion was calculated by adding up all dispenses between initiation of the therapy and the 120 days thereafter. Good adherence was defined as using minimal 80% of the recommended duration and intensity of use. RESULTS: The prevalence of patients initiating pSCTs was stable at 0.4 per 1000 inhabitants per quarter during 2010. In 2011, the prevalence was on average 0.7, with peaks in the first (0.8 per 1000) and fourth (1.0 per 1000) quarters of 2011. In 2012, the prevalence was stable again at 0.3. In 2013, prevalence was on average 0.4, with a small peak in the first quarter. Adherence was 15.4% in 2010 versus 20.1% in 2011 (P = .002). In 2012, adherence was 13.9%, compared with 18.9% in 2013 (P = .008). CONCLUSIONS: Not only the likelihood of initiating pSCTs, but also the extent of adherence to these treatments, although generally low, seems higher during reimbursement.

Economics of chronic diseases protocol: cost-effectiveness modelling and the future burden of non-communicable disease in Europe.

BACKGROUND: The majority of chronic disease is caused by risk factors which are mostly preventable. Effective interventions to reduce these risks are known and proven to be applicable to a variety of settings. Chronic disease is generally developed long before the fatal outcome, meaning that a lot of people spend a number of years in poor health. Effective prevention measures can prolong lives of individuals and significantly improve their quality of life. However, the methods to measure cost-effectiveness are a subject to much debate. The Economics of Chronic Diseases project aims to establish the best possible methods of measuring cost-effectiveness as well as develop micro-simulation models apt at projecting future burden of chronic diseases, their costs and potential savings after implementation of cost-effective interventions. METHOD: This research project will involve eight European countries: Bulgaria, Finland, Greece, Lithuania, The Netherlands, Poland, Portugal and the United Kingdom (UK). A literature review will be conducted to identify scientific articles which critically review the methods of cost-effectiveness. Contact will be made health economists to inform and enrich this review. This evidence will be used as a springboard for discussion at a meeting with key European stakeholders and experts with the aim of reaching a consensus on recommendations for cost-effectiveness methodology. Epidemiological data for coronary heart disease, chronic kidney disease, type 2 diabetes and chronic obstructive pulmonary disease will be collected along with data on time trends in three major risk factors related to these diseases, specifically tobacco consumption, blood pressure and body mass index. Economic and epidemiological micro-simulation models will be developed to asses the future distributions of risks, disease outcomes, healthcare costs and the cost-effectiveness of interventions to reduce the burden of chronic diseases in Europe. DISCUSSION: This work will help to establish the best methods of measuring cost-effectiveness of health interventions as well as test a variety of scenarios to reduce the risk factors associated with selected chronic diseases. The modelling projections could be used to inform decisions and policies that will implement the best course of action to curb the rising incidence of chronic diseases.

Crossing borders: factors affecting differences in cost-effectiveness of smoking cessation interventions between European countries.

OBJECTIVES: Many different factors affect the transferability of cost-effectiveness results between countries. The objective is to quantify the impact of nine potential causes of variation in cost-effectiveness of pharmacological smoking cessation therapies (SCTs) between The Netherlands (reference case), Germany, Sweden, UK, Belgium, and France. METHODS: The life-time benefits of smoking cessation were calculated using the Benefits of Smoking Cessation on Outcomes model, following a cohort of smokers making an unaided quit attempt, or using nicotine replacement therapy (NRT), bupropion, or varenicline. We investigated the impact of between-country differences in nine factors-demography, smoking prevalence, mortality, epidemiology and costs of smoking-related diseases, resource use and unit costs of SCTs, utility weights and discount rates-on the incremental net monetary benefit (INMB), using a willingness-to-pay (WTP) of euro20,000 per quality adjusted life year (QALY). RESULTS: The INMB of 1000 quit attempts with NRT versus unaided, varies from euro0.39 million (Germany) to euro1.47 million (France). The differences between the countries were primarily due to differences in discount rates, causing the INMB to change between -65% to +62%, incidence and mortality rates (epidemiology) of smoking-related diseases (-43% to +35%) and utility weights. Impact also depended on the WTP for a QALY and time horizon: at a low WTP or a short time horizon, the resource use and unit costs of SCTs had the highest impact on INMB. CONCLUSIONS: Although all INMBs were positive, there were significant differences across countries. These were primarily related to choice of discount rate and epidemiology of diseases.

The risk of overestimating cost savings from hospital-at-home schemes: A literature review.

BACKGROUND: The concept of hospital-at-home means that home treatment is provided to patients who would otherwise have been treated in the hospital. This may lead to lower costs, but estimates of savings may be overstated if inpatient hospital costs are priced incorrectly. OBJECTIVE: The objective of this study was to evaluate the quality of cost analyses of hospital-at-home studies for acute conditions published from 1996 through 2019 and to present an overview of evidence. DESIGN: Literature review DATA SOURCES: The PubMed and NHS EED databases were searched. REVIEW METHODS: The overall quality of studies was evaluated based on Quality of Health Economic Studies (QHES) score, design, sample size, alignment of cost calculation with study perspective, time horizon, use of tariffs or real resource use and clarity of calculations. Furthermore, we systematically assessed whether cost savings were likely to be overestimated, based on criteria about the costing of inpatient hospital days, informal care costs and bias. RESULTS: We identified 48 studies. The average QHES score was 60 out of a maximum of 100 points. Almost all studies violated one or more criteria for the risk of overestimation of cost savings. The most frequent problems were the use of average unit prices per inpatient day (not taking into account the decreasing intensity of care) and biased designs. Most studies found cost differences in favour of hospital-at-home; the range varied from savings of €8773 to a cost increase of €2316 per patient. CONCLUSION: Overall quality of studies was not good, with some exceptions. Many cost savings were probably overestimated.

Costs of clinical events in type 2 diabetes mellitus patients in the Netherlands: A systematic review.

BACKGROUND: Type 2 diabetes mellitus (T2DM) is an established risk factor for cardiovascular and nephropathic events. In the Netherlands, prevalence of T2DM is expected to be as high as 8% by 2025. This will result in significant clinical and economic impact, highlighting the need for well-informed reimbursement decisions for new treatments. However, availability and consistent use of costing methodologies is limited. OBJECTIVE: We aimed to systematically review recent costing data for T2DM-related cardiovascular and nephropathic events in the Netherlands. METHODS: A systematic literature review in PubMed and Embase was conducted to identify available Dutch cost data for T2DM-related events, published in the last decade. Information extracted included costs, source, study population, and costing perspective. Finally, papers were evaluated using the Consolidated Health Economic Evaluation Reporting Standards (CHEERS). RESULTS: Out of initially 570 papers, 36 agreed with the inclusion criteria. From these studies, 150 cost estimates for T2DM-related clinical events were identified. In total, 29 cost estimates were reported for myocardial infarction (range: €196-€27,038), 61 for stroke (€495-€54,678), fifteen for heart failure (€325-€16,561), 24 for renal failure (€2,438-€91,503), and seventeen for revascularisation (€3,000-€37,071). Only four estimates for transient ischaemic attack were available, ranging from €587 to €2,470. Adherence to CHEERS was generally high. CONCLUSIONS: The most expensive clinical events were related to renal failure, while TIA was the least expensive event. Generally, there was substantial variation in reported cost estimates for T2DM-related events. Costing of clinical events should be improved and preferably standardised, as accurate and consistent results in economic models are desired.

Research Costs Investigated: A Study Into the Budgets of Dutch Publicly Funded Drug-Related Research.

BACKGROUND: The costs of performing research are an important input in value of information (VOI) analyses but are difficult to assess. OBJECTIVE: The aim of this study was to investigate the costs of research, serving two purposes: (1) estimating research costs for use in VOI analyses; and (2) developing a costing tool to support reviewers of grant proposals in assessing whether the proposed budget is realistic. METHODS: For granted study proposals from the Netherlands Organization for Health Research and Development (ZonMw), type of study, potential cost drivers, proposed budget, and general characteristics were extracted. Regression analysis was conducted in an attempt to generate a 'predicted budget' for certain combinations of cost drivers, for implementation in the costing tool. RESULTS: Of 133 drug-related research grant proposals, 74 were included for complete data extraction. Because an association between cost drivers and budgets was not confirmed, we could not generate a predicted budget based on regression analysis, but only historic reference budgets given certain study characteristics. The costing tool was designed accordingly, i.e. with given selection criteria the tool returns the range of budgets in comparable studies. This range can be used in VOI analysis to estimate whether the expected net benefit of sampling will be positive to decide upon the net value of future research. CONCLUSION: The absence of association between study characteristics and budgets may indicate inconsistencies in the budgeting or granting process. Nonetheless, the tool generates useful information on historical budgets, and the option to formally relate VOI to budgets. To our knowledge, this is the first attempt at creating such a tool, which can be complemented with new studies being granted, enlarging the underlying database and keeping estimates up to date.

Mix and match. A simulation study on the impact of mixed-treatment comparison methods on health-economic outcomes.

BACKGROUND: Decision-analytic cost-effectiveness (CE) models combine many parameters, often obtained after meta-analysis. AIM: We compared different methods of mixed-treatment comparison (MTC) to combine transition and event probabilities derived from several trials, especially with respect to health-economic (HE) outcomes like (quality adjusted) life years and costs. METHODS: Trials were drawn from a simulated reference population, comparing two of four fictitious interventions. The goal was to estimate the CE between two of these. The amount of heterogeneity between trials was varied in scenarios. Parameter estimates were combined using direct comparison, MTC methods proposed by Song and Puhan, and Bayesian generalized linear fixed effects (GLMFE) and random effects models (GLMRE). Parameters were entered into a Markov model. Parameters and HE outcomes were compared with the reference population using coverage, statistical power, bias and mean absolute deviation (MAD) as performance indicators. Each analytical step was repeated 1,000 times. RESULTS: The direct comparison was outperformed by the MTC methods on all indicators, Song's method yielded low bias and MAD, but uncertainty was overestimated. Puhan's method had low bias and MAD and did not overestimate uncertainty. GLMFE generally had the lowest bias and MAD, regardless of the amount of heterogeneity, but uncertainty was overestimated. GLMRE showed large bias and MAD and overestimated uncertainty. Song's and Puhan's methods lead to the least amount of uncertainty, reflected in the shape of the CE acceptability curve. GLMFE showed slightly more uncertainty. CONCLUSIONS: Combining direct and indirect evidence is superior to using only direct evidence. Puhan's method and GLMFE are preferred.

Economic Burden in Chinese Patients with Diabetes Mellitus Using Electronic Insurance Claims Data.

BACKGROUND: There is a paucity of studies that focus on the economic burden in daily care in China using electronic health data. The aim of this study is to describe the development of the economic burden of diabetic patients in a sample city in China from 2009 to 2011 using electronic data of patients' claims records. METHODS: This study is a retrospective, longitudinal study in an open cohort of Chinese patients with diabetes. The patient population consisted of people living in a provincial capital city in east China, covered by the provincial urban employee basic medical insurance (UEBMI). We included any patient who had at least one explicit diabetes diagnosis or received blood glucose lowering medication in at least one registered outpatient visit or hospitalization during a calendar year in the years 2009-2011. Cross-sectional descriptions of different types of costs, prevalence of diabetic complications and related diseases, medication use were performed for each year separately and differences between three years were compared using a chi-square test or the non-parametric Kruskal-Wallis H test. RESULTS: Our results showed an increasing trend in total medical cost (from 2,383 to 2,780 USD, p = 0.032) and diabetes related cost (from 1,655 to 1,857 USD) for those diabetic patients during the study period. The diabetes related economic burden was significantly related to the prevalence of complications and related diseases (p<0.001). The overall medication cost during diabetes related visits also increased (from 1,335 to 1,383 USD, p = 0.021). But the use pattern and cost of diabetes-related medication did not show significant changes during the study period. CONCLUSION: The economic burden of diabetes increased significantly in urban China. It is important to improve the prevention and treatment of diabetes to contribute to the sustainability of the Chinese health-care system.

Unremarked or Unperformed? Systematic Review on Reporting of Validation Efforts of Health Economic Decision Models in Seasonal Influenza and Early Breast Cancer.

BACKGROUND: Transparent reporting of validation efforts of health economic models give stakeholders better insight into the credibility of model outcomes. In this study we reviewed recently published studies on seasonal influenza and early breast cancer in order to gain insight into the reporting of model validation efforts in the overall health economic literature. METHODS: A literature search was performed in Pubmed and Embase to retrieve health economic modelling studies published between 2008 and 2014. Reporting on model validation was evaluated by checking for the word validation, and by using AdViSHE (Assessment of the Validation Status of Health Economic decision models), a tool containing a structured list of relevant items for validation. Additionally, we contacted corresponding authors to ask whether more validation efforts were performed other than those reported in the manuscripts. RESULTS: A total of 53 studies on seasonal influenza and 41 studies on early breast cancer were included in our review. The word validation was used in 16 studies (30 %) on seasonal influenza and 23 studies (56 %) on early breast cancer; however, in a minority of studies, this referred to a model validation technique. Fifty-seven percent of seasonal influenza studies and 71 % of early breast cancer studies reported one or more validation techniques. Cross-validation of study outcomes was found most often. A limited number of studies reported on model validation efforts, although good examples were identified. Author comments indicated that more validation techniques were performed than those reported in the manuscripts. CONCLUSIONS: Although validation is deemed important by many researchers, this is not reflected in the reporting habits of health economic modelling studies. Systematic reporting of validation efforts would be desirable to further enhance decision makers' confidence in health economic models and their outcomes.

The Missing Stakeholder Group: Why Patients Should be Involved in Health Economic Modelling.

Evaluations of healthcare interventions, e.g. new drugs or other new treatment strategies, commonly include a cost-effectiveness analysis (CEA) that is based on the application of health economic (HE) models. As end users, patients are important stakeholders regarding the outcomes of CEAs, yet their knowledge of HE model development and application, or their involvement therein, is absent. This paper considers possible benefits and risks of patient involvement in HE model development and application for modellers and patients. An exploratory review of the literature has been performed on stakeholder-involved modelling in various disciplines. In addition, Dutch patient experts have been interviewed about their experience in, and opinion about, the application of HE models. Patients have little to no knowledge of HE models and are seldom involved in HE model development and application. Benefits of becoming involved would include a greater understanding and possible acceptance by patients of HE model application, improved model validation, and a more direct infusion of patient expertise. Risks would include patient bias and increased costs of modelling. Patient involvement in HE modelling seems to carry several benefits as well as risks. We claim that the benefits may outweigh the risks and that patients should become involved.

Cost-Analysis of Seven Nosocomial Outbreaks in an Academic Hospital.

OBJECTIVES: Nosocomial outbreaks, especially with (multi-)resistant microorganisms, are a major problem for health care institutions. They can cause morbidity and mortality for patients and controlling these costs substantial amounts of funds and resources. However, how much is unclear. This study sets out to provide a comparable overview of the costs of multiple outbreaks in a single academic hospital in the Netherlands. METHODS: Based on interviews with the involved staff, multiple databases and stored records from the Infection Prevention Division all actions undertaken, extra staff employment, use of resources, bed-occupancy rates, and other miscellaneous cost drivers during different outbreaks were scored and quantified into Euros. This led to total costs per outbreak and an estimated average cost per positive patient per outbreak day. RESULTS: Seven outbreaks that occurred between 2012 and 2014 in the hospital were evaluated. Total costs for the hospital ranged between €10,778 and €356,754. Costs per positive patient per outbreak day, ranged between €10 and €1,369 (95% CI: €49-€1,042), with a mean of €546 and a median of €519. Majority of the costs (50%) were made because of closed beds. CONCLUSIONS: This analysis is the first to give a comparable overview of various outbreaks, caused by different microorganisms, in the same hospital and all analyzed with the same method. It shows a large variation within the average costs due to different factors (e.g. closure of wards, type of ward). All outbreaks however cost considerable amounts of efforts and money (up to €356,754), including missed revenue and control measures.

Cost-effectiveness of obinutuzumab for chronic lymphocytic leukaemia in The Netherlands.

BACKGROUND: Obinutuzumab combined with chlorambucil (GClb) has shown to be superior to rituximab combined with chlorambucil (RClb) and chlorambucil (Clb) in newly diagnosed patients with chronic lymphocytic leukaemia (CLL). This study evaluates the cost-effectiveness per life-year and quality-adjusted life-year (QALY) of GClb compared to RClb, Clb, and ofatumumab plus chlorambucil (OClb) in The Netherlands. METHODS: A Markov model was developed to assess the cost-effectiveness of GClb, RClb, Clb and other treatments in the United Kingdom. A country adaptation was made to estimate the cost-effectiveness of these therapies in The Netherlands using Dutch unit costs and Dutch data sources for background mortality and post-progression survival. RESULTS: An incremental gain of 1.06 and 0.64 QALYs was estimated for GClb compared to Clb and RClb respectively, at additional costs of €23,208 and €7254 per patient. Corresponding incremental cost-effectiveness ratios (ICERs) were €21,823 and €11,344 per QALY. Indirect treatment comparisons showed an incremental gain varying from 0.44 to 0.77 QALYs for GClb compared to OClb and additional costs varying from €7041 to €5028 per patient. The ICER varied from €6556 to €16,180 per QALY. Sensitivity analyses showed the robustness of the results. CONCLUSION: GClb appeared to be a cost-effective treatment strategy compared to RClb, OClb and Clb.

Financial evaluations of antibiotic stewardship programs-a systematic review.

INTRODUCTION: There is an increasing awareness to counteract problems due to incorrect antimicrobial use. Interventions that are implemented are often part of an Antimicrobial Stewardship Program (ASPs). Studies publishing results from these interventions are increasing, including reports on the economical effects of ASPs. This review will look at the economical sections of these studies and the methods that were used. METHODS: A systematic review was performed of articles found in the PubMed and EMBASE databases published from 2000 until November 2014. Included studies found were scored for various aspects and the quality of the papers was assessed following an appropriate check list (CHEC criteria list). RESULTS: 1233 studies were found, of which 149 were read completely. Ninety-nine were included in the final review. Of these studies, 57 only mentioned the costs associated with the antimicrobial medication. Others also included operational costs (n = 23), costs for hospital stay (n = 18), and/or other costs (n = 19). Nine studies were further assessed for their quality. These studies scored between 2 and 14 out of a potential total score of 19. CONCLUSIONS: This review gives an extensive overview of the current financial evaluation of ASPs and the quality of these economical studies. We show that there is still major potential to improve financial evaluations of ASPs. Studies do not use similar nor consistent methods or outcome measures, making it impossible draw sound conclusions and compare different studies. Finally, we make some recommendations for the future.

A few years later. Update of the cost-effectiveness of infant pneumococcal vaccination in Dutch children.

This study aimed to calculate the cost-effectiveness of infant pneumococcal vaccination in the Netherlands, using the 13-valent PCV13 vs. the currently used 10-valent PCV10. We adapted a previously published model, using recent estimates of epidemiological and efficacy data. In 12 scenarios, we explored the impact of different assumptions on the incremental cost-effectiveness ratio (ICER) of PCV13 over PCV10.Taking only direct effects on invasive pneumococcal disease into account, PCV13 was not found to be cost-effective at a price difference of €11 per dose. If herd protection, replacement and non-invasive disease were also taken into account, the ICER of PCV13 compared with PCV10 was below €30 000/QALY gained in 11 of 12 scenarios. PCV13 was considered dominant in the primary scenario with a price difference below €2.63 per dose.

Not simply more of the same: distinguishing between patient heterogeneity and parameter uncertainty.

In cost-effectiveness (CE) Markov models, heterogeneity in the patient population is not automatically taken into account. We aimed to compare methods of dealing with heterogeneity on estimates of CE, using a case study in chronic obstructive pulmonary disease (COPD). We first present a probabilistic sensitivity analysis (PSA) in which we sampled only from distributions representing parameter uncertainty. This ignores any heterogeneity. Next, we explored heterogeneity by presenting results for subgroups, using a method that samples parameter uncertainty simultaneously with heterogeneity in a single-loop PSA. Finally, we distinguished parameter uncertainty from heterogeneity in a double-loop PSA by performing a nested simulation within each PSA iteration. Point estimates and uncertainty differed substantially between methods. The incremental CE ratio (ICER) ranged from € 4900 to € 13,800. The single-loop PSA led to a substantially different shape of the CE plane and an overestimation of the uncertainty compared with the other 3 methods. The CE plane for the double-loop PSA showed substantially less uncertainty and a stronger negative correlation between the difference in costs and the difference in effects compared with the other methods. This came at the cost of higher calculation times. Not accounting for heterogeneity, subgroup analysis and the double-loop PSA can be viable options, depending on the decision makers' information needs. The single-loop PSA should not be used in CE research. It disregards the fundamental differences between heterogeneity and sampling uncertainty and overestimates uncertainty as a result.