Preference Elicitation Techniques Research Articles

A Methodological Study to Compare Alternative Modes of Administration with Value EQ-5D Using Preference-Elicitation Techniques

ObjectivesTime trade-off (TTO) and discrete choice experiment (DCE) preference-elicitation techniques can be administered using face-to-face interviews (F2F), unassisted online (UO) surveys, or remote-assisted (RA) interviews. The objective of this study was to explore how the mode of administration affects the quality and reliability of preference-elicitation data. MethodsEQ-5D-5L health states were valued using composite TTO (cTTO) and DCE approaches by the UK general population. Participants were allocated to 1 of 2 study groups. Group A completed both F2F and UO surveys (n = 271), and group B completed both RA and UO surveys (n = 223). The feasibility of survey completion and the reliability and face-validity of data collected were compared across all modes of administration. ResultsFewer participants reported receiving sufficient guidance on the cTTO tasks during the UO survey compared with the 2 assisted modes. Participants across all modes typically reported receiving sufficient guidance on the DCE tasks.cTTO data were less reliable from the UO survey compared with both assisted modes, but there were no differences in DCE data reliability. cTTO data from all modes demonstrated face-validity; however, the UO survey produced higher utilities for moderate and severe health states than both assisted modes. Both F2F and RA modes provided comparably reliable data. ConclusionsThe reliability of DCE data is not affected by the mode of administration. Interviewer-assisted modes of administration (F2F or RA) yield more reliable cTTO data than unassisted surveys. Both F2F and RA surveys produced similar-quality data.

Using the Online Elicitation of Personal Utility Functions Approach to Derive a Patient-Based 5-Level Version of EQ-5D Value Set: A Study in 122 Patients With Rheumatic Diseases From Germany

ObjectivesTraditional preference elicitation methods, such as discrete choice experiments or time trade-off, usually require large sample sizes. This can limit their applicability in patient populations, where recruiting enough participants can be challenging.The objective of this study was to test a new method, called the Online elicitation of Personal Utility Functions (OPUF) approach, to derive an EQ-5D-5L value set from a relatively small sample of patients with rheumatic diseases. MethodsOPUF is a new type of online survey that implements compositional preference elicitation techniques. Central to the method are 3 valuation steps: (1) dimension weighting, (2) level rating, and (3) anchoring. An English demo version of the OPUF survey can be accessed at https://valorem.health/eq5d5l. From the responses, a personal EQ-5D-5L utility function can be constructed for each participant, and a group-level value set can be derived by aggregating model coefficients across participants. ResultsA total of 122 patients with rheumatic disease from Germany completed the OPUF survey. The survey was generally well received; most participants completed the survey in less than 20 minutes and were able to derive a full EQ-5D-5L value set. The precision of mean coefficients was high, despite the small sample size. ConclusionsOur findings demonstrate that OPUF can be used to derive an EQ-5D-5L value set from a relatively small sample of patients. Although the method is still under development, we think that it has the potential to be a valuable preference elicitation tool and to complement traditional methods in several areas.

A ranking approach for robust portfolio decision analysis based on multilinear portfolio utility functions and incomplete preference information

Uncertainties are often present in typical portfolio decision analysis (PDA), and it is prudent to conduct a robust PDA to identify project portfolios that perform reasonably well across the entire range of uncertain parameter values. Previous research on robust PDA and applications has almost entirely relied on additive portfolio preference functions, with little research on robustness under multilinear portfolio utility functions (PUFs). A ranking approach for robust PDA is proposed based on multilinear PUFs, incomplete preference information, and the algorithm for accelerating stochastic multicriteria acceptability analysis (SMAA). First, the initial set of portfolios is screened through various constraints to produce a smaller set of desirable portfolios. Second, several types of incomplete preference information are thought to be elicited with regard to multilinear PUFs, including preference information on coefficients, portfolios, and uncertainties on preference elicitation techniques. Third, the SMAA decomposition algorithm is implemented as a ranking approach to perform robust PDA on the basis of multilinear PUFs, incomplete preference information, and potentially numerous portfolios in the set of desirable portfolios. Finally, the approach is applied to a case to validate its effectiveness, and comparisons among additive, multiplicative, and multilinear PUFs in terms of decision recommendations are performed.

OP94 Online Elicitation Of Personal Utility Functions (OPUF): A New Tool For Eliciting EQ-5D-5L Value Sets On The Societal-, Group-, Subgroup-, And Individual-Level

IntroductionThe ‘Online elicitation of Personal Utility Functions’ (OPUF) is a new method for valuing health states. It is based on compositional preference elicitation techniques. In contrast to established, decompositional techniques, such as time trade-off or discrete choice experiment (DCE), the OPUF approach does not require hundreds or thousands of respondents, but allows estimating utility functions for small (patient) groups and even on the individual level. The objective of this study was to generate and compare EQ-5D-5L value sets on the societal-, group-, subgroup-, and individual-level.MethodsA demonstration version of the EQ-5D-5L OPUF Tool is available at: https://eq5d5l.me. It broadly consists of three valuation steps: dimension weighting, level rating, and anchoring. Responses were combined on the individual level to construct personal utility functions. Every respondent also completed three conventional DCEs. Preferences were aggregated across individuals to estimate a societal and various group-level preference functions. We then assessed the heterogeneity of preferences between groups using descriptive statistics and k-means cluster analysis.ResultsA representative sample (n = 1,000) of the United Kingdom (UK) population was recruited through the prolific online platform. On average, it took participants about 7 minutes to complete the survey. Data of 874 respondents were included in the analysis. For each respondent, we constructed a personal EQ-5D-5L value set. The derived utility functions predicted respondents’ choices in DCE with an accuracy of 78 percent. On the societal level, the predicted values for the EQ-5D-5L health states ranged from -0.376 to 1. Health state preference varied greatly between groups. This was largely due to differences in the anchoring, while there was near consensus on the relative importance of the five EQ-5D dimensions between groups. Demographic characteristics explained only a small proportion of the variability.ConclusionsUsing the OPUF approach, we were not only able to estimate a new EQ-5D-5L value set for the UK, but also to examine the underlying individual preferences in an unprecedented level of detail.

Developing an Australian utility value set for the Early Childhood Oral Health Impact Scale-4D (ECOHIS-4D) using a discrete choice experiment

PurposePreference-based quality of life measures (PBMs) are used to generate quality-adjusted life years (QALYs) in economic evaluations. A PBM consists of (1) a health state classification system and (2) a utility value set that allows the instrument responses to be converted to QALYs. A new, oral health-specific classification system, the Early Childhood Oral Health Impact Scale-4D (ECOHIS-4D) has recently been developed. The aim of this study was to generate an Australian utility value set for the ECOHIS-4D.MethodsA discrete choice experiment with duration (DCETTO) was used as the preference elicitation technique. An online survey was administered to a representative sample of Australian adults over 18 years. Respondents were given 14 choice tasks (10 tasks from the DCE design of 50 choice sets blocked into five blocks, 2 practice tasks, a repeated and a dominant task). Data were analyzed using the conditional logit model.ResultsA total of 1201 respondents from the Australian general population completed the survey. Of them, 69% (n = 829) perceived their oral health status to be good, very good, or excellent. The estimated coefficients from the conditional logit models were in the expected directions and were statistically significant (p < 0.001). The utility values for health states defined by the ECOHIS-4D ranged from 0.0376 to 1.0000.ConclusionsThis newly developed utility value set will enable the calculation of utility values for economic evaluations of interventions related to oral diseases such as dental caries among young children. This will facilitate more effective resource allocation for oral health services.

EE14 A Systematic Review of Preference Elicitation Techniques Used in Valuing Children's Health-Related Quality-of-Life

Valuing children’s health states for use in economic evaluations is important globally, and particularly so for jurisdictions where cost-utility analysis is the preferred analysis for decision making. However, there remains debate about the approaches to elicitation of values. The main objectives of this paper were to identify and describe methods used to value children’s health states and determine the specific considerations required when using the methods.

What drives differences in preferences for health states between patients and the public? A qualitative investigation of respondents’ thought processes

Cost-effectiveness analyses using quality-adjusted life-years (QALYs) are used in decision-making regarding which interventions are available via many national healthcare systems. QALYs are calculated based on health state values provided by preference elicitation techniques. Several national decision-making bodies recommend that health state values should be based on preferences elicited from general populations, rather than from patients. Previous studies have shown systematic differences between health state values elicited from members of the general population and from patients. Various explanations for this phenomenon have been proposed, however empirical evidence for these is scarce. We aimed to explore possible reasons for discrepancies between public and patient valuations by undertaking qualitative cognitive interviews, asking 14 members of the general population and 12 people with multiple sclerosis (MS) to think aloud while completing a preference elicitation task (time trade-off) for MS-related health states. The interviews were undertaken between December 2016 and October 2017 in the South West region of England, and were analysed using the Framework Method. As anticipated, we found that participants with MS had more experience of health problems and used this experience to consider how they might adapt to the health states over time, and which dimensions of health-related quality of life were most important to them. We found no evidence that participants with MS were less affected by framing effects and focusing illusions, more likely to prioritise non-physical dimensions of health, or more prone to loss aversion, endowment effects and non-compensatory decision-making. These findings contribute to our understanding of how patients and members of the general population respond to preference elicitation exercises, and why their preferences may differ, and may help to inform developing areas of research, such as the joint presentation of cost-effectiveness results from multiple perspectives, and the use of preferences elicited from patients for experienced health states.

PNS99 Using an Online Survey to Elicit Preferences in China: Comparison of Different Methods Based on SF-6DV2

To compare the performance of 3 different elicitation methods in an online survey environment in China. An online panel of general population in China participated in a web survey. Each participant was randomized to complete two out of three preference elicitation technique tasks, which were 8 random Discrete Choice Experiment (DCE) tasks, 6 random Time Trade Off (TTO) tasks, and 10 random Best Worst Scaling (BWS) tasks. Ordinal method results were self-anchored on the death values to estimate the utilities. Models assuming homogeneous preference (Conditional Logit and linear regressions (OLS)) and preference heterogeneity (Hierarchical Bayes and Mixed Effects Linear Hierarchical Bayes) were used. The number of significant levels, percent certainty and Root Likelihood (RLH) were employed for the comparison. A total of 961 participants (40.6% <40 years, 49 % female, 76.4% healthy) who passed the quality checks were utilized in the analysis. Based on the number of significant levels, percent certainty and RLH, the BWS homogenous preference model was the best (0, 30.5%, 0.39) compared to DCE (6, 20%, 0.38) and TTO (17, 26%, 0.27). The worst state (PITS) was valued at 0.17, -0.69 and -0.033 respectively. The TTO heterogenous preference model had a whole dimension insignificant. Between BWS and DCE, the preferred model specifications changed for the DCE and both had none insignificant levels. The Percent certainty was 70 & 55.2, RLH at 0.58 & 0.61 and PITS valued at 0.197 and 0.359 for BWS and DCE models respectively. Using online DCE or BWS to elicit preference in china is feasible while TTO requires further optimization. Data from ordinal methods modeled by heterogenous preference were more consistent than the one from the cardinal method (TTO) or modeled assuming homogenous preference. Cardinal methods still perform worse than ordinal methods in online surveys and generates less reliable results.

Learning Modulo Theories for constructive preference elicitation

This paper introduces CLEO, a novel preference elicitation algorithm capable of recommending complex configurable objects characterized by both discrete and continuous attributes and constraints defined over them. While existing preference elicitation techniques focus on searching for the best instance in a database of candidates, CLEO takes a constructive approach to recommendation through interactive optimization in a space of feasible configurations. The algorithm assumes minimal initial information, i.e., a set of catalog attributes, and defines decisional features as logic formulae combining Boolean and algebraic constraints over the attributes. The (unknown) utility of the decision maker is modeled as a weighted combination of features. CLEO iteratively alternates a preference elicitation step, where pairs of candidate configurations are selected based on the current utility model, and a refinement step where the utility is refined by incorporating the feedback received. The elicitation step leverages a Max-SMT solver to return optimal configurations according to the current utility model. The refinement step is implemented as learning to rank, and a sparsifying norm is used to favor the selection of few informative features in the combinatorial space of candidate decisional features.A major feature of CLEO is that it can recommend optimal configurations in hybrid domains (i.e., including both Boolean and numeric attributes), thanks to the use of Max-SMT technology, while retaining uncertainty in the decision-maker's utility and noisy feedback. In so doing, it adapts the recently introduced learning modulo theory framework to the preference elicitation setting. The combinatorial formulation of the utility function coupled with the feature selection capabilities of 1-norm regularization allow to effectively deal with the uncertainty in the DM utility while retaining high expressiveness. Experimental results on complex recommendation tasks show the ability of CLEO to quickly identify optimal configurations, as well as its capacity to recover from suboptimal initial choices. Our empirical evaluation highlights how CLEO outperforms a state-of-the-art Bayesian preference elicitation algorithm when applied to a purely discrete task

Pairwise comparisons or constrained optimization? A usability evaluation of techniques for eliciting decision priorities

AbstractDecision support methodologies provide notations for expressing and communicating the priorities that inform a decision. Although a substantial literature has explored the theoretical merits of such notations and methodologies, much less work has investigated their usability in practice, which is of vital importance for their widespread adoption by users. In this paper, we explore the usability of two well‐known preference elicitation techniques, pairwise comparisons and constrained optimization. The techniques were explored through two contrasting crowd worker experiments, a preliminary one evaluating recognition, that is, the ability to identify the most suitable formulation for a given task, and the other synthesis, that is, the ability to construct formulations for a given task. The tasks are based on a case study involving source selection, a well‐known problem in the data integration domain. The results of the empirical evaluation show that, overall, pairwise comparisons resulted in significantly higher performance than constrained optimization, yet there is negligible difference between the usability appraisals for each technique. Furthermore, we observed that the technique that participants perform better with is not necessarily the one that they consider more usable.

3125 steps to perfect health: a nonparametric approach to developing the EQ-5D-5L value set.

The EQ-5D-5L is a commonly used instrument for assessing the utility of different health states. Health state utility values are a key component of health technology evaluations. Such evaluations are used to support evidence-based decisions surrounding health resource allocations and therefore rely on the accuracy of the valuation set used. This paper takes an alternative approach to developing an EQ-5D-5L value set for Canada. The aim is to introduce a robust method that is likely to generate a value set with improved accuracy and that can be used to generate value sets for other populations without the need for modification. The common approach to developing a valuation set for preference-based instruments is to ask a population sample to value a subset of the health states using an established preference elicitation technique. The relationship between the elicited health states and the preferences is used to inform a model to predict the utility values for the unsampled health states described by the instrument. The true relationship is unknown and the functional forms chosen in the modelling process vary across valuation studies. We use nonparametric local constant regression to estimate an EQ-5D-5L value set for Canada and propose this method as an alternative for value set development because it does not require the specification of a functional form at the outset. Compared to the existing valuation model for Canada, the nonparametric method improves in-sample fit, reducing the average squared prediction error by 94.46% and the mean absolute error by 79.37%. In four of five sets of out-of-sample studies, this new approach performs significantly better than 9 comparison models. Despite lacking any restriction on the functional form of the resulting valuations, the valuation set generated by this new approach is logically consistent. 100% of the pairs of health states in which one state is dominant have health state values which respect this ordering. The value set also appears to differ substantially from the comparators. Overall, the results suggest that nonparametric regression is a promising tool for the estimation of EQ-5D-5L valuation sets and may be a good option in a standardised methodology for value set development.

Field comparisons of incentive-compatible preference elicitation techniques

Knowledge of consumer demand is important for firms, policy-makers, and economists. One common tool for incentive-compatible demand elicitation, the Becker-DeGroot-Marschak (BDM) mechanism, has been widely used in laboratory settings but rarely evaluated for reliability at large scale in the field. In two field experiments (for a new agricultural information service and rainfall index insurance) we compare demand curves estimated from BDM implementations with demand curves estimated from choices at individually randomized fixed prices. In the test (for an agricultural information service) we obtain demand curves from BDM and fixed prices that align closely and are unable to reject equivalence of the demand curves. For the second test (of rainfall index insurance) the results are mixed, with the distributions lining up well at certain points of the demand curve and deviating at others. Overall, we find no evidence for systematic bias. Our evidence suggests that a ‘reframed’ version of BDM that uses discrete prices may minimize non-standard bidding behavior and is well suited to future experimental work in low-literacy environments.

The Graph Model for Conflict Resolution: Reflections on Three Decades of Development

The fundamental design and inherent capabilities of the Graph Model for Conflict Resolution (GMCR) to address a rich range of complex real world conflict situations are put into perspective by tracing its historical development over a period spanning more than 30 years, and highlighting great opportunities for meaningful future expansions within an era of artificial intelligence (AI) and intensifying conflict in an over-crowded world. By constructing a sound theoretical foundation for GMCR based upon assumptions reflecting what actually occurs in reality, a fascinating story is narrated on how GMCR was able to expand in bold new directions as well as take advantage of many important legacy decision technologies built within the earlier Metagame Analysis and later Conflict Analysis paradigms. From its predecessors, for instance, GMCR could benefit by the employment of option form put forward within Metagame Analysis for effectively recording a conflict, as well as preference elicitation techniques and solution concepts for defining chess-like behavior when calculating stability of states from the realm of Conflict Analysis. The key ideas outlined in the paper underlying the current and projected capabilities of GMCR include the development of four different ways to handle preference uncertainty in the presence of either transitive or intransitive preferences; a wide range of solution concepts for describing many kinds of human behavior under conflict; unique coalition analysis algorithms for determining if a given decision maker can fare better in a dispute via cooperation; tracing the evolution of a conflict over time; and the matrix formulation of GMCR for computational efficiency when calculating stability and also theoretically expanding GMCR in bold new directions. Inverse engineering is mentioned as an AI extension of GMCR for computationally determining the preferences required by decision makers in order to reach a desirable state, such as a climate change agreement in which all nations significantly cut back on their greenhouse gas emissions. The basic design of a decision support system for permitting researchers and practitioners to readily apply the foregoing and other advancements in GMCR to tough real world controversies is discussed. Although GMCR has been successfully applied to challenging disputes arising in many different fields, a simple climate change negotiation conflict between the US and China is utilized to explain clearly key concepts mentioned throughout the fascinating historical journey surrounding GMCR.

PNS203 USE OF DISCRETE CHOICE EXPERIMENTS TO INFORM HTA DECISION MAKING.

Discrete choice experiments (DCEs) are a regularly used preference elicitation technique in health economics to elicit preferences for both healthcare interventions and programmes. There is growing recognition that DCEs have the potential to contribute more directly to outcome research for use in economic evaluation, such as health technology assessments (HTA). DCE methodology has recently developed to facilitate the use of the technique to elicit preferences of health states on a scale suitable for informing estimates of quality adjusted life years (QALYs). The aim of this pragmatic literature review was to understand the extent to which DCEs are used to inform HTA decision-making. We conducted targeted searches of selected HTA agency websites using Google’s site search function for terms such as “discrete choice experiment". The four agencies (NICE/England, CADTH/Canada, HAS/France and ICER/USA), were chosen for their prominence in the field of health technology appraisal and the transparency of the processes by which they evaluate healthcare technologies. A bespoke template was developed and used in the data extraction. All four agencies have used DCEs to inform HTA decision-making, however, the degree to which they are used varies between agencies. NICE has predominantly used DCEs within its Clinical Guidelines/Public Health programmes. DCE has also been used within its Technology Appraisal programme, although this has been limited to date. CADTH has used methodology involving DCEs to inform its HTA decision-making processes. HAS have also used DCE to inform its decision-making processes; however, much less frequently. Lastly, ICER has referenced DCEs within both its evidence reviews and scoping documents. Although uptake of this preference elicitation technique has been limited, it is clear that all four agencies recognise DCEs as an effective method for preference elicitation in the context of HTA.

A Novel Preference Elicitation Technique Based on a Graph Model and Its Application to a Brownfield Redevelopment Conflict in China.

Disputes are very common and pervasive in brownfield redevelopment projects, in which multiple stakeholders or decision-makers (DMs) strategically interact with each other with a conflict of interest. The preference information of DMs involved plays a vital role in identifying possible outcomes or resolutions for resolving a tough brownfield conflict. In this research, a novel preference ranking technique is purposefully proposed within the graph model for conflict resolution (GMCR) paradigm to effectively and accurately garner DMs’ actual preferences, in which states are ranked according to their similarities and closeness to the most and least preferred states instead of subjective option statements or weights in traditional preference ranking methods. Finally, a real-world brownfield conflict which occurred in China is utilized to show how the proposed preference ranking method can be applied for conveniently obtaining the true preference information of DMs and strategically determining the equilibria of a given dispute. The case study indicates that the novel preference elicitation approach is more objective and reasonable than the traditional option prioritization method. Moreover, there exists an equilibrium which can provide strategic advice and meaningful insights for addressing the brownfield conflict.

Small-scale forestry and carbon offset markets: An empirical study of Vermont Current Use forest landowner willingness to accept carbon credit programs.

This study investigates the preferences of small forest landowners regarding forest carbon credit programs while documenting characteristics of potentially successful frameworks. We designed hypothetical carbon credit programs with aggregated carbon offset projects and requirements of existing voluntary and compliance protocols in mind. We administered a mail survey to 992 forest landowners in Vermont’s Current Use Program utilizing best-worst choice, a novel preference elicitation technique, to elicit their preferences about these programs. We found that small forest landowners see revenue as the most important factor in a carbon credit program and the duration of the program as the least important factor. Landowners reported that shorter program duration, higher revenue, and lower withdrawal penalties positively impact their willingness to accept forest carbon credit programs. Notably, our study includes carbon credit program implementer as a key program attribute, allowing us to quantify landowners’ tradeoffs between non-profit, for-profit, and government organizations. Overall, we found that landowners significantly prefer working with a non-profit organization. Based on monetary estimates of willingness-to-accept compensation, our results suggest that aggregated forest carbon offset projects incorporating small forest landowners could be piloted successfully in Vermont by non-profit organizations while maintaining relatively strict guidelines of existing carbon offset protocols.

Stated preferences for long-term care: a literature review

ABSTRACTPerson-centred provision of long-term care (LTC) requires information on how individuals value respective LTC services. The literature on LTC preferences has not been comprehensively reviewed, existing summaries are contradictory. An explorative, scoping review was conducted to provide a thorough methodological description and results synthesis of studies that empirically investigated LTC preference outcomes based on respondents’ statements. A wide search strategy, with 18 key terms relating to ‘LTC’ and 31 to ‘preferences’, was developed. Database searches in PubMed, Ovid and ScienceDirect were conducted in February 2016. The 59 studies meeting the inclusion criteria were grouped and methodically described based on preference elicitation techniques and methods. Despite substantial methodological heterogeneity between studies, certain findings consistently emerged for the investigated LTC preference outcomes. The large majority of respondents preferred to receive LTC in their known physical and social environment when care needs were moderate, but residential care when care needs were extensive. Preferences were found to depend on a variety of personal, environmental, social and cultural aspects. Dependent individuals aspired to preserve their personal and social identity, self-image, independence, autonomy, control and dignity, which suggests that LTC preferences are a function of the perceived ability of a specific LTC arrangement to satisfy peoples’ basic physiological and mental/social needs. Research on LTC preferences would greatly profit from a standardisation of respective concepts and methods.

Using neuroeconomics to understand environmental valuation

Contingent valuation, choice experiments, and other stated preference methods are frequently used to capture the nonmarket valuation of natural resources and ecosystem services. The emerging field of neuroeconomics, which assesses the neuroscience underlying decision-making, plots a promising course to explore the mechanisms underlying complex environmental valuation decisions. Neuroeconomic methods offer a unique capacity to isolate value components that contribute to willingness-to-pay (WTP), separating an individual's response to natural resource attributes that are of interest to economists from other attributes or influences on the decision process. Neuroimaging data can also aid in understanding differences in response between preference elicitation techniques and identifying the use of different decision processes and heuristics during valuation. This article surveys the benefits and limitations of using neuroeconomics methods to assess the value of environmental goods, and focuses on three examples where neuroeconomics may inform environmental valuation: protest responses, comparison of hypothetical and consequential choice contexts, and the evaluation of environmental attributes and optimization of study design in stated choice experiments. Neuroeconomics methods offer a foundation for positive collaboration between environmental economists and cognitive neuroscientists, yielding metrics that complement and augment current stated preference methods of determining environmental valuation.

Preferences for public involvement in health service decisions: a comparison between best-worst scaling and trio-wise stated preference elicitation techniques

Stated preference elicitation techniques, such as discrete choice experiments and best-worst scaling, are now widely used in health research to explore the public’s choices and preferences. In this paper, we propose an alternative stated preference elicitation technique, which we refer to as ‘trio-wise’. We explain this new technique, its relative advantages, modeling framework, and how it compares to the best-worst scaling method. To better illustrate the differences and similarities, we utilize best-worst scaling Case 2, where individuals make best and worst (most and least) choices for the attribute levels that describe a single profile. We demonstrate this new preference elicitation technique using an empirical case study that explores preferences among the general public for ways to involve them in decisions concerning the health care system. Our findings show that the best-worst scaling and trio-wise preference elicitation techniques both retrieve similar preferences. However, the capability of our trio-wise method to provide additional information on the strength of rank preferences and its ability to accommodate indifferent preferences lead us to prefer it over the standard best-worst scaling technique.

A comparison of methods for converting DCE values onto the full health-dead QALY scale.

Preference elicitation techniques such as time trade-off (TTO) and standard gamble (SG) receive criticism for their complexity and difficulties of use. Ordinal techniques such as discrete choice experiment (DCE) are arguably easier to understand but generate values that are not anchored onto the full health-dead 1-0 quality-adjusted life-year (QALY) scale required for use in economic evaluation. This article compares existing methods for converting modeled DCE latent values onto the full health-dead QALY scale: 1) anchoring DCE values using dead as valued in the DCE and 2) anchoring DCE values using TTO value for worst state to 2 new methods: 3) mapping DCE values onto TTO and 4) combining DCE and TTO data in a hybrid model. Models are compared using their ability to predict mean TTO health state values. We use postal DCE data (n = 263) and TTO data (n = 307) collected by interview in a general population valuation study of an asthma condition-specific measure (AQL-5D). New methods 3 and 4 using mapping and hybrid models are better able to predict mean TTO health state values (mean absolute difference [MAD], 0.052-0.084) than the anchor-based methods (MAD, 0.075-0.093) and were better able to predict mean TTO health state values even when using in their estimation a subsample of the available TTO data. These new mapping and hybrid methods have a potentially useful role for producing values on the QALY scale from data elicited using ordinal techniques such as DCE for use in economic evaluation that makes best use of the desirable properties of each elicitation technique and elicited data. Further research is encouraged.