Exploring challenges to TTO utilities: valuing states worse than dead

A Feasible Estimation of a "Corrected" EQ-5D Social Tariff

OBJECTIVES

To demonstrate the feasibility of estimating a social tariff free of utility curvature and probability weighting biases and to test transferability between riskless and risky contexts.

METHODS

Valuations for a selection of EQ-5D-3L health states were collected from a large and representative sample (N = 1676) of the Spanish general population through computer-assisted personal interviewing. Two elicitation methods were used: the traditional time trade-off (TTO) and a novel risky-TTO procedure. Both methods are equivalent for better than death states, which allowed us to test transferability of utilities across riskless and risky contexts. Corrective procedures applied are based on rank-dependent utility theory, identifying parameter estimates at the individual level. All corrections are health-state specific, which is a unique feature of our corrective approach.

RESULTS

Two corrected value sets for the EQ-5D-3L system are estimated, highlighting the feasibility of developing national tariffs under nonexpected utility theories, such as rank-dependent utility. Furthermore, transferability was not supported for at least half of the health states valued by our sample.

CONCLUSIONS

It is feasible to estimate a social tariff by using interviewing techniques, sample sizes, and sample representativeness equivalent to prior studies designed to generate national value sets for the EQ-5D. Utilities obtained in distinct contexts may not be interchangeable. Our findings caution against routinely taking transferability of utility for granted.

Valuing quality of life for economic evaluations in cancer: navigating multiple methods

INTRODUCTION

Utility values offer a quantitative means to evaluate the impact of novel cancer treatments on patients' quality of life (QoL). However, the multiple methods available for valuing QoL present challenges in selecting the most appropriate method across different contexts.

AREAS COVERED

This review provides cancer clinicians and researchers with an overview of methods to value QoL for economic evaluations, including standalone and derived preference-based measures (PBMs) and direct preference elicitation methods. Recent developments are described, including the comparative performance of cancer-specific PBMs versus generic PBMs, measurement of outcomes beyond health-related QoL, and increased use of discrete choice experiments to elicit preferences. Recommendations and considerations are provided to guide the choice of method for cancer research.

EXPERT OPINION

We foresee continued adoption of the QLU-C10D and FACT-8D in cancer clinical trials given the extensive use of the EORTC QLQ-C30 and FACT-G in cancer research. While these cancer-specific PBMs offer the convenience of eliciting utility values without needing a standalone PBM, researchers should consider potential limitations if they intend to substitute them for generic PBMs. As the field advances, there is a greater need for consensus on the approach to selection and integration of various methods in cancer clinical trials.

Beyond 10-year lead-times in EQ-5D-5L: leveraging alternative lead-times in willingness-to-accept questions to capture preferences for worse-than-dead states and their implication

BACKGROUND

A fixed 10-year lead-time in composite time-trade-off (C-TTO) tasks might compromise the precision of utility values below - 1. This study explored how alternative lead-times (ALTs) influence EQ-5D-5L value sets and their implications in economic evaluations.

METHODS

Leveraging data from Taiwan's EQ-5D-5L valuation and capitalizing on its exploratory willingness-to-accept question, we explored participants' quantification of "worse-than-dead (WTD)" health states with ALTs up to 50 years. We then derived alternative value sets incorporating these ALTs through interval regression and compared them against those from conventional models. To evaluate their impact on health change valuation, we simulated utility differences for all possible EQ-5D-5L health-state-pairs using each value set.

RESULTS

With a salient floor effect observed in the C-TTO values, the model with ALT led to a wider range of predicted utilities ( - 2.3897 ~ 1), compared with those of conventional models (generalized least squares (GLS):  - 0.7773 ~ 1; Tobit-GLS:  - 0.9583 ~ 1). Compared to the Tobit-GLS model, the model with ALT increased the numerical distance in 80% of health-state-pairs, with 11% decreasing and 9% altering direction (e.g., positive to negative) in utility differences.

CONCLUSIONS

While ALTs offer insights into patient preferences, their integration into economic evaluations might require rescaling. Future research should prioritize advanced rescaling methods or enhanced elicitation strategies for populations with substantial censoring. This is pivotal for improving the elicitation of extreme WTD states and accurately discerning the relative distances between health states. Countries developing EQ-5D-5L value sets should consider pilot studies and incorporating region-specific questions on social determinants, especially where pronounced floor effects are suspected.

Valuation of EQ-5D-5L health states from cancer patients' perspective: a feasibility study

OBJECTIVES

To assess the feasibility of estimating an EQ-5D-5L value set using a small study design in cancer patients and to compare the EQ-5D-5L values based on the preferences of cancer patients with those of the general public.

METHODS

Patients with clinically diagnosed cancers were recruited from two hospitals in Shanghai, China. In face-to-face interviews using the EQ-PVT survey, health states were valued by cancer patients using both cTTO and DCE methods. cTTO data was modelled alone or jointly with DCE data. Forty-eight models using different model specifications (cross-attribute level effect [CALE] and additive models), random/fixed effects model assumptions, data heteroscedasticity and censoring were estimated. The best performed model was identified in terms of monotonicity of estimated model coefficients and out-of-sample prediction accuracy.

RESULTS

Data collected from 221 cancer patients who participated in the study were included. The hybrid CALE model using both TTO and DCE data performed best in terms of prediction accuracy (Lin's concordance coefficient = 0.989; root mean squared error = 0.058) and suggested that pain/discomfort and anxiety/depression were the most undesirable health problems. Compared to values based on general Chinese public's health preferences, the values based on cancer patients' preferences were much higher and lower for health states characterized by extreme mobility problems and severe/extreme pain or discomfort, respectively.

CONCLUSION

This study demonstrated the feasibility of using a small design to develop EQ-5D-5L value sets based on cancer patients' health preferences. Since there were signs of differences between preferences of patients and general population, it may be valuable to develop patient-specific value sets and use them in clinical decision making and economic evaluations.

Different Frameworks, Similar Results? Head-to-Head Comparison of the Generic Preference-Based Health-Outcome Measures CS-Base and EQ-5D-5L

OBJECTIVE

We compared two generic, preference-based health-outcome measures: the novel patient-centered Château-Santé Base (CS-Base), entailing a multi-attribute preference response framework, and the widely used EQ-5D-5L, regarding effects of different measurement frameworks and different descriptive systems.

METHODS

We conducted a cross-sectional study using a random sample of patients (3019 reached, 1988 included) in the USA with various health conditions. The CS-Base (12 attributes, each with four levels), EQ-5D-5L and the 5D-4L (an ad hoc, multi-attribute preference response-based measure that includes five attributes similar to the EQ-5D-5L, but with four levels) were used as health-outcome measures. We compared the proportions of problems reported on health attributes, statistical robustness and face validity of coefficients, attribute importance, differentiation between health states based on health-state values obtained with these measures, and user experience.

RESULTS

All the CS-Base and 5D-4L coefficients had logical orders and significant differences from the reference level (p < 0.001). Substantial differences were observed in the CS-Base and 5D-4L coefficients between all levels on all attributes, while subtle differences were seen in those of the EQ-5D-5L. Attribute importance of usual (daily) activities were lowest or second lowest in all the three health-outcome measures. Attributes with the highest importance in the CS-Base, 5D-4L, and EQ-5D-5L were respectively mobility, anxiety/depression, and pain/discomfort. Four attributes are similar between the CS-Base and EQ-5D-5L, eight are exclusive to CS-Base. Of the eight, vision and hearing had the highest importance. Health-state values showed a smoother distribution with minimal discontinuity in the CS-Base and EQ-5D-5L than in the 5D-4L. In user experience evaluation, both CS-Base and the 5D-4L showed mean scores above 50 (indicating positive evaluation) in terms of the description of health and ease of understanding.

CONCLUSIONS

This study demonstrated that CS-Base and 5D-4L, which are grounded in the multi-attribute preference response framework, produced statistically robust coefficients, with better face validity than those for the EQ-5D-5L. CS-Base and the EQ-5D-5L outperformed the 5D-4L in differentiating between health states. Notwithstanding differences in content, measurement frameworks, and estimated coefficients, the computed health-state values were similar between CS-Base and EQ-5D-5L.

Valuing selected WAItE health states using the Time Trade-Off methodology: findings from an online interviewer-assisted remote survey

PURPOSE

The Weight-Specific Adolescent Instrument for Economic Evaluation (WAItE) is a physical weight-specific patient reported outcome measure for use in adolescence. The purpose of this study was to use the Time Trade-Off (TTO) methodology, administered using an online interviewer-assisted remote survey, to obtain utility values for several health states from the WAItE descriptive system from a sample of the UK adult general population.

METHODS

The adult sample was gathered using a market research company and a sample of local residents. All participants completed the same interviewer-assisted remote survey, which included rating WAItE states of varying impairment using the TTO.

RESULTS

42 adults completed the survey. Utility values were gathered for four health states, ranging from low impairment to the most severe health from the WAItE descriptive system (the Pits state). Consistent orderings of the WAItE health states were observed; the health state with the lowest level of impairment was valued highest and the Pits state was valued lowest. Several respondents (n = 7, 17%) considered the Pits state to be worse than death; however, the mean value of this health state was 0.23.

CONCLUSIONS

The utility value of the Pits state relative to death generated from this study will be used to anchor latent values for WAItE health states generated from a Discrete Choice Experiment onto the 0 = death, 1 = full health Quality Adjusted Life Year (QALY) scale as part of a valuation study for the WAItE in the UK population. This study also provides further evidence that interviewer-assisted digital studies are feasible for collecting TTO data.

Testing Nonmonotonicity in Health Preferences

OBJECTIVE

The main aim of this article is to test monotonicity in life duration. Previous findings suggest that, for poor health states, longer durations are preferred to shorter durations up to some threshold or maximum endurable time (MET), and shorter durations are preferred to longer ones after that threshold.

METHODS

Monotonicity in duration is tested through 2 ordinal tasks: choices and rankings. A convenience sample (n = 90) was recruited in a series of experimental sessions in which participants had to rank-order health episodes and to choose between them, presented in pairs. Health episodes result from the combination of 7 EQ-5D-3L health states and 5 durations. Monotonicity is tested comparing the percentage rate of participants whose preferences were monotonic with the percentage of participants with nonmonotonic preferences for each health state. In addition, to test the existence of preference reversals, we analyze the fraction of people who switch their preference from rankings to choices.

RESULTS

Monotonicity is frequently violated across the 7 EQ-5D health states. Preference patterns for individuals describe violations ranging from almost 49% with choices to about 71% with rankings. Analysis performed by separate states shows that the mean rates of violations with choices and ranking are about 22% and 34%, respectively. We also find new evidence of preference reversals and some evidence-though scarce-of transitivity violations in choices.

CONCLUSIONS

Our results show that there is a medium range of health states for which preferences are nonmonotonic. These findings support previous evidence on MET preferences and introduce a new "choice-ranking" preference reversal. It seems that the use of 2 tasks with a similar response scale may make preference reversals less substantial, although it remains important and systematic.

HIGHLIGHTS

Two procedures based on ordinal comparisons are used to elicit preferences: direct choices and rankings. Our study reports significant rates of nonmonotonic preferences (or maximum endurable time [MET]-type preferences) for different combinations of durations and EQ-5D health states.Analysis for separate health states shows that the mean rates of nonmonotonicity range from 22% (choices) to 34% (rankings), but within-subject analysis shows that nonmonotonicity is even higher, ranging from 49% (choices) to 71% (rankings). These violations challenge the validity of multiplicative QALY models.We find that the MET phenomenon may affect particularly those EQ-5D health states that are in the middle of the severity scale and not so much the extreme health states (i.e., very mild and very severe states).We find new evidence of preference reversals even using 2 procedures of a similar (ordinal) nature. Percentage rates of preference reversals range from 1.5% to 33%. We also find some (although scarce) evidence on violations of transitivity.

Health state utilities associated with hyperphagia: Data for use in cost-utility models

Objective

Rare genetic diseases of obesity typically present with hyperphagia, a pathologic desire to consume food. Cost-utility models assessing the value of treatments for these rare diseases will require health state utilities representing hyperphagia. This study estimated utilities associated with various hyperphagia severity levels.

Methods

Four health state vignettes were developed using published literature and clinician input to represent various severity levels of hyperphagia. Utilities were estimated for these health states in a time trade-off elicitation study in a UK general population sample.

Results

In total, 215 participants completed interviews (39.5% male; mean age 39.1 years). Mean (SD) utilities were 0.98 (0.02) for no hyperphagia, 0.91 (0.10) for mild hyperphagia, 0.70 (0.30) for moderate hyperphagia, and 0.22 (0.59) for severe hyperphagia. Mean (SD) disutilities were -0.08 (0.10) for mild, -0.28 (0.30) for moderate, and -0.77 (0.58) for severe hyperphagia.

Conclusions

These data show increasing severity of hyperphagia is associated with decreased utility. Utilities associated with severe hyperphagia are similar to those of other health conditions severely impacting quality of life (QoL). These findings highlight that treatments addressing substantial QoL impacts of severe hyperphagia are needed. Utilities estimated here may be useful in cost-utility models of treatments for rare genetic diseases of obesity.

Health state utilities for beta-thalassemia: a time trade-off study

BACKGROUND

Beta-thalassemia (BT) is an inherited blood disorder characterized by reduced levels of functional hemoglobin resulting in phenotypes ranging from clinically asymptomatic to severely anemic. Patients with BT may require lifelong regular blood transfusions supported by appropriate iron chelation therapy (ICT). This study aimed to determine how the UK general population values BT health states associated with differing transfusion burden and ICT.

METHODS

Composite time trade-off (cTTO) methodology was employed to elicit health state utilities in BT. Relevant BT literature related to symptom and quality-of-life impact, including physical, functional, and emotional well-being, and safety profiles of BT treatments were considered when drafting health state descriptions. Eleven health state descriptions were developed and validated by hematologists and patient advocates for clinical accuracy and completeness. 200 individuals from the UK general population participated in the cTTO interviews.

RESULTS

The mean age of participants was 41.50 years (SD 16.01, range 18-81); 88 (46.8%) were female. Utility values ranged from 0.78 (SD 0.34) for non-transfusion dependent BT with oral ICT to 0.37 (SD 0.50) for high transfusion burden with subcutaneous ICT in transfusion-dependent BT.

CONCLUSIONS

This study provides health utilities for a range of BT health states from the UK general population perspective. Importantly, lower transfusion burden and lower burden of anemia were associated with higher utilities. To a lesser extent, differential modes of ICT were found to impact utility valuations in patients with BT. The utilities obtained in this study can be employed as inputs in cost-effectiveness analyses of BT therapies.

Estimating health state utilities in primary hyperoxaluria type 1: a valuation study

AIMS

Quantitative data on health state utilities in primary hyperoxaluria type 1 (PH1) are lacking. This study was conducted to estimate utility values in PH1 using 3 standard valuation techniques.

MATERIALS AND METHODS

Health state vignettes were developed with input from expert clinicians to describe different stages of chronic kidney disease (CKD) within the setting of PH1, along with a post-combined liver and kidney transplant (CLKT) health state ≥12 months following transplantation. The utility associated with living in each PH1 health state, as described by the vignettes, was evaluated by members of the UK general public using standard utility assessment techniques, including EQ-5D-5L, Visual Analog Scale, and Time Trade-Off.

RESULTS

A similar pattern across the three valuation methods was observed. Utility values were roughly constant from CKD stage 1-3b and then dropped sharply from these states to CKD stage 4. Decreases in utility in the later stages of CKD (stages 4-5) were followed by a recovery in quality of life in the post-CLKT health state.

LIMITATIONS

Vignettes may not fully capture the burden of living with PH1.

CONCLUSIONS

This study serves as an informative source of data on how the burden of PH1 varies across disease stages of CKD and post-CLKT in patients with PH1. The study findings highlight the unique clinical features of PH1 compared with non-PH1-related CKD, such as the need for earlier and more intensive hemodialysis, the risk of systemic oxalosis, and the potential need for CLKT (as opposed to kidney-only transplant). The impact of PH1 on health-related quality of life, which worsens once hemodialysis is required and systemic disease manifestations occur, is captured in this study using quantitative health state utilities. These data provide an understanding of the impact of PH1 on health-related quality of life and will facilitate health economic evaluation of future treatments.

A threshold explanation for the lack of variation in negative composite time trade-off values

Background

Recent studies concluded that for health states considered worse than dead (WTD), as measured with the time trade-off (cTTO) method, negative mean values were insensitive to health state severity, which represents a validity problem for the cTTO. However, the aforementioned studies analysed negative values in isolation, which causes selection bias as the value distribution is truncated.

Aim

To investigate the validity of aforementioned studies and of negative values in general.

Methods

The ‘threshold explanation’ was formulated: beyond a certain severity threshold, preferences change from better than dead (BTD) to WTD. This threshold differs between respondents. Thus, negative values across severity are obtained from different respondents, and responses added for higher severity contribute negative values close to zero, explaining the aforementioned insensitivity. This explanation was tested using data from the Dutch EQ-5D-5L valuation study. Respondents valued 10 health states. Based on respondents’ number of WTD preferences, segments were constructed, containing respondents with similar severity thresholds. Using regression models for each individual respondent, we examined the relation between values and severity and compared respondents between segments.

Results

Negative values, when analysed in isolation, were insensitive to severity. However, for individual respondents and within most segments, cTTO values and severity were negatively related. For individual respondents, negative slopes were steeper for segments with more WTD preferences, as predicted by the threshold explanation.

Discussion

Analysing negative values in isolation leads to biased estimates. Analyses of cTTO values for individual respondents refute the insensitivity of negative cTTO values.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11136-022-03155-6.

An initiative to develop capability-adjusted life years in Sweden (CALY-SWE): Selecting capabilities with a Delphi panel and developing the questionnaire

Introduction

Capability-adjusted life years Sweden (CALY-SWE) are a new Swedish questionnaire-based measure for quality of life based on the capability approach. CALY-SWE are targeted towards use in cost-effectiveness evaluations of social welfare consequences. Here, we first motivate the measure both from a theoretical and from a Swedish policy-making perspective. Then, we outline the core principles of the measure, namely the relation to the capability approach, embedded equity considerations inspired by the fair-innings approach, and the bases for which capabilities should be considered. The aims were to 1) the most vital capabilities for individuals in Sweden, 2) to define a sufficient level of each identified capability to lead a flourishing life, and to 3) develop a complete questionnaire for the measurement of the identified capabilities.

Material and methods

For the selection of capabilities, we used a Delphi process with Swedish civil society representants. To inform the questionnaire development, we conducted a web survey in three versions, with each Swedish 500 participants, to assess the distribution of capabilities that resulted from the Delphi process in the Swedish population. Each version was formulated with different strictness so that less strict wordings of a capability level would apply to a larger share of participants. All versions also included questions on inequality aversion regarding financial, educational, and health capabilities.

Results

The Delphi process resulted in the following six capabilities: Financial situation & housing, health, social relations, occupations, security, and political & civil rights. We formulated the final phrasing for the questionnaire based on normative reasons and the distribution of capabilities in the population while taking into account inequality aversion.

Conclusion

We developed a capability-based model for cost effectiveness economic evaluations of broader social consequences, specific to the Swedish context.

Japanese preference weights of the Adult Social Care Outcomes Toolkit for Carers (ASCOT-Carer)

Purpose

We developed preference weights of the Adult Social Care Outcomes Toolkit for Carers (ASCOT-Carer) in Japan.

Methods

We used best–worst scaling (BWS) and composite time trade-off (cTTO) to determine the preference weights for ASCOT-Carer states in the general population. TTO values were applied to convert the BWS scores to utilities. The sample number was approximately 1000 for the BWS survey and 200 for the TTO survey. Whereas face-to-face surveys by computer-assisted interviewing were adopted for the TTO tasks, a web-based survey was used for the BWS tasks. In the BWS tasks, the ASCOT-Carer states were presented, and the “best,” “worst,” “second best,” and “second worst” domains in a profile were selected. A mixed logit model was applied to the BWS data.

Results

The respondents’ background was similar to that of the general population, although the number of people in the age and sex categories was equal. The preference weights for calculating the utilities of the ASCOT-Carer states were estimated. The estimated utilities of the ASCOT-Carer states were distributed between 1 and 0.02. All preference weights were consistent. The item with the highest preference weight was level 1 in the “space and time to be yourself.” The least preferred item was level 4 in the “space and time to be yourself” and “control over daily life” domains.

Conclusion

We established Japanese preference weights for ASCOT-Carer states, the first weights of an Asian country. The estimated utilities can contribute to the measurement of caregivers’ social care-related QoL and perform of cost-effectiveness analyses.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11136-021-03076-w.

Behavioral Theories That Have Influenced the Way Health State Preferences Are Elicited and Interpreted: A Bibliometric Mapping Analysis of the Time Trade-Off Method With VOSviewer Visualization

Aim

The aim of this paper is to develop an understanding of how behavioral theories have influenced the way preferences for health-related quality of life are elicited and interpreted. We focus on the Time Trade-off (TTO) method given it represents the quality-adjusted life-year (QALY) concept-that survival in less-than-full health can be deemed equivalent to a shorter survival in full health. To our knowledge this is the first review using a combination of systematic scoping review, bibliometrics and VOSviewer visualization to map the development of ideas in health economics.

Methods

A priori, we selected three behavioral theories to explore within our review, referred to here as Expected Utility Theory, Non-Expected Utility Theory and Probabilistic Choice Theory. A fourth topic, Order Effects, is defined broadly to encompass behavioral theories around timing/sequence of events. For the main search, Scopus was used to identify literature that had (a) elicited TTO values and/or (b) contributed to the way TTO values were elicited and interpreted, from inception to July 2021. Papers that focused on the latter category were given the label "behavioral" and underwent additional analyses. A two stage-screening was applied to assess eligibility. Co-citation, co-authorship and co-occurrence of keywords was used to chart the development of TTO over time.

Results

A total of 1,727 records were retrieved from Scopus and were supplemented by an additional 188 papers. There were 856 applied and 280 behavioral papers included in the final corpus, with the behavioral set split equally into four sets of 70 papers to chart the development of keywords over time: (1) 1972-1999; (2) 2000-2010, (3) 2010-2015 and (4) 2015-2021.

Discussion

The keyword analysis suggested that whilst some ideas transition quickly from economic theory to the TTO literature, such as the impact of Order Effects, others take longer to be assimilated, for example Non-Expected Utility models or failure of constant discounting. It is therefore important that researchers within health economics work more closely with those in mainstream economics and keep abreast of the wider economics and behavioral sciences to expedite the uptake of new and relevant ideas.

Developing a preference-based measure for weight-specific health-related quality of life in adolescence: the WAItE UK valuation study protocol

INTRODUCTION

Adolescent obesity is a public health problem in the UK. The Weight-Specific Adolescent Instrument for Economic Evaluation (WAItE) has been developed as the first weight-specific health-related quality of life measure appropriate for economic evaluation, but currently cannot be used to generate quality-adjusted life years (QALYs), which are the basis of cost-utility analysis. Generic measures (such as the EQ-5D-Y or CHU-9D) may be insensitive to small but important health changes in overweight or obese adolescents. This study aims to generate a preference-based scoring algorithm for the WAItE.

METHODS AND ANALYSIS

A discrete choice experiment (DCE) will be administered to value health states described by the WAItE classification system. These health states will be presented to members of the adult general population of the UK via an online survey. A range of regression models will be used to produce the utility algorithm for the WAItE. The DCE-visual analogue scale and time trade-off (TTO) anchoring methods will be used anchor the value set on to the 0-1 QALY scale.

ETHICS AND DISSEMINATION

The Newcastle University Medical School Ethics Committee approved the study (references 4772/2020 (DCE) and 9978/2020 (TTO)). The developed algorithm can be applied to future economic evaluations of weight management interventions and treatments for adolescents.

What Justifies the Allocation of Health Care Resources to Patients with Disorders of Consciousness?

This paper critically engages ethical issues in the allocation of novel, and potentially costly, health care resources to patients with disorders of consciousness. First, we review potential benefits of novel health care resources for patients and their families and outline preliminary considerations to address concerns about cost. We then address two problems regarding the allocation of health care resources to patients with disorders of consciousness: (1) the problem of uncertain moral status; and (2) the problem of accurately measuring the welfare burdens these resources would relieve. We conclude by suggesting that opportunity-based frameworks might complement standard approaches for justifying resources allocation to patients with disorders of consciousness.

An overview of the time trade-off method: concept, foundation, and the evaluation of distorting factors in putting a value on health

INTRODUCTION

Preference-based instruments measuring health status express the value of specific health states in a single number. One method used is time trade-off (TTO). Health-status values are key elements in calculating quality-adjusted life years (QALYs) and are pertinent for resource allocation. Since they are used in economic evaluations of healthcare, searching for a theoretical foundation of TTO in economics is justified.

AREA COVERED

This paper provides an overview of TTO, including its relation to economic theory, and discusses biases and distortions, compiled from recent and older research. Inconsistencies between TTO and random utility theory were detected; The TTO is confounded by time preferences and by respondents' life expectancies. TTO is cognitively challenging, therefore guidance during the interviews is needed, producing interview effects. TTO does not measure one thing at a time, nor are the values independent of other states that are being valued in the same task. That is, TTO does not exhibit theoretical measurement properties such as unidimensionality and the invariance principle.

EXPERT OPINION

We conclude that the TTO may be a pragmatic method of eliciting health state values, but the limitations in regard to measurement theory and practical elicitation problems makes it prone to inconsistencies and arbitrariness.

Hybrid Methodology to Improve Health Status Utility Values Derivation Using EQ-5D-5L and Advanced Multi-Criteria Techniques

This paper presented a new approach to the calculation of quality-adjusted life years (QALY) based on multi-criteria decision-making (MCDM) methods and using the EQ-5D-5L questionnaire. The health status utility values were calculated through a hybrid methodology. We combined the analytic hierarchy process (AHP), the AHP with a D-number extended fuzzy preference relation (D-AHP), the fuzzy analytic hierarchy process (F-AHP), and the technique for order preference by similarity to the ideal solution (TOPSIS) to obtain individual and aggregated utility values. The preference data were elicited using a sample of individuals from a Colombian university. In all tested methods, the ordinal preferences were consistent, and the weights were compared using the Euclidean distance criterion (EDC). We identified F-AHP-TOPSIS as the optimal method; its benefits were associated with modeling the response options of the EQ-5D in linguistic terms, it gave the best approximation to the initial preferences according to EDC, and it could be used as an alternative to the known prioritization method. This hybrid methodology was particularly useful in certain medical decisions concerned with understanding how a specific person values his or her current health or possible health outcomes from different interventions in small population samples and studies carried out in low- and middle-low-income countries.

A Head-On Ordinal Comparison of the Composite Time Trade-Off and the Better-Than-Dead Method

OBJECTIVES

The valuation of health states worse than dead is challenging. Currently used time trade-off methods face problems in (1) detecting time-dependent preferences and (2) insensitivity toward severity for states worse than dead. The better-than-dead (BTD) method has the potential to detect time-dependent preferences. This study compares the BTD and composite time trade-off (cTTO) methods at the ordinal level.

METHODS

An experiment was conducted in a convenience sample in which respondents (N = 200) valued the same set of 7 health states in the BTD method and cTTO method. Binary BTD responses were used, with response categories of better than dead and worse than dead. Ternary cTTO responses were used, with the additional equal-to-dead response category. Polychoric correlations were used to determine the agreement between these methods. Consistency and test-retest reliability were assessed within methods.

RESULTS

Overall agreement between the cTTO and BTD method equaled 77.1% and differed between health states and respondents. For both methods, there were few inconsistencies, and the test-retest reliability was comparable (88%). Health states were more often considered worse than dead in the BTD method (BTD: 54.7%, cTTO: 37.2%).

CONCLUSIONS

The high agreement between both methods and the comparable amount of inconsistencies and test-retest reliability suggest that the methods have similar measurement properties. The BTD method yielded higher frequencies of worse-than-dead responses while essentially asking respondents to make similar choices. This accounts for part of the disagreement between the methods. Several explanations are offered for this difference, yet more research is needed to explain this phenomenon.

A new tool for creating personal and social EQ-5D-5L value sets, including valuing 'dead'

The EuroQol Group's health descriptive systems, the EQ-5D-3L and its successor introduced in 2009, the EQ-5D-5L, are widely used worldwide for valuing health-related quality of life for cost-utility analysis and patient-reported health outcome measures. A new online tool for creating personal and social EQ-5D-5L value sets was recently developed and trialled in New Zealand (NZ). The tool, which includes extensive checks of the quality of participants' data, implements the PAPRIKA method - a novel type of adaptive discrete choice experiment in the present context - and a binary search algorithm to identify any health states worse than dead. After development and testing, the tool was distributed in an online survey in February and March 2018 to a representative sample of NZ adults (N = 5112), whose personal value sets were created. The tool's extensive data quality checks resulted in a 'high-quality' sub-sample of 2468 participants whose personal value sets were, in effect, averaged to create a social value set for NZ. These results overall as well as feedback from participants indicates that the new valuation tool is feasible and acceptable to participants, enabling valuation data to be relatively easily and cheaply collected. The tool could also be used in other countries, tested against other methods for creating EQ-5D-5L value sets, applied in personalised medicine and adapted to create value sets for other health descriptive systems.

Women's Preferences for Maternal and Neonatal Morbidity and Mortality in Childbirth

Purpose. To measure utility values that describe women's willingness to tradeoff maternal morbidity for fetal benefit among pregnant and nonpregnant women of reproductive age. Methods. We recruited English-speaking women aged 18 to 45 years in clinical and community-based settings. Eight health states were studied: 4 maternal (healthy, stroke, hysterectomy, death) and 4 neonatal (healthy, severe cerebral palsy [CP], severe mental retardation [MR], death). Utilities were assessed on a subset of 9 pairs of mom/baby delivery outcomes. Participants ranked the 9 pairs of outcomes in order of preference, then standard gamble methods were used to calculate utilities. Numeracy skills were assessed. Results. Utilities were obtained from 477 participants (recruitment rate = 94%). Twenty-one percent were pregnant, 63% were parents, and 54% were African American. Utilities did not differ significantly between pregnant and nonpregnant women or based on numeracy score. The highest (nonhealthy) values were assigned to baby healthy/mom hysterectomy (0.999), baby healthy/mom stroke (0.946), and baby CP/mom healthy (0.940). The lowest values were assigned to baby death/mom hysterectomy (0.203), baby MR/mom death (0.150), and baby death/mom stroke (0.087). Nonwhite participants assigned a significantly higher value to baby MR/mom death (P = 0.01), baby MR/mom stroke (P = 0.02), baby MR/mom healthy (P < 0.01), and baby MR/mom hysterectomy (P = 0.02) than white participants. Conclusion. When asked to value pairs of maternal/fetal outcomes that required a tradeoff of morbidity and mortality, women tended to assign the highest utility to combinations that resulted in a "healthy baby." They assigned the lowest values to combinations that resulted in a baby's death or MR. Our findings highlight the importance of 1) assessing individual preferences and goals of care for pregnancy outcomes and 2) measuring utilities among reproductive-aged women when modeling obstetric decision making.

Principles of Need and the Aggregation Thesis

Principles of need are constantly referred to in health care priority setting. The common denominator for any principle of need is that it will ascribe some kind of special normative weight to people being worse off. However, this common ground does not answer the question how a plausible principle of need should relate to the aggregation of benefits across individuals. Principles of need are sometimes stated as being incompatible with aggregation and sometimes characterized as accepting aggregation in much the same way as utilitarians do. In this paper we argue that if one wants to take principles of need seriously both of these positions have unreasonable implications. We then characterize and defend a principle of need consisting of sufficientarian elements as well as prioritarian which avoids these unreasonable implications.

Latent Class Models Reveal Poor Agreement between Discrete-Choice and Time Tradeoff Preferences

Background. In health economics, there has been interest in using discrete-choice experiments (DCEs) to derive preferences for health states in lieu of previously established approaches like time tradeoff (TTO). We examined whether preferences elicited through DCEs are associated and agree with preferences elicited through TTO tasks. Methods. We used data from 1073 respondents to the Canadian EQ-5D-5L valuation study. Multivariate mixed-effects models specified a common likelihood for the TTO and discrete-choice data, with separate but correlated random effects for the TTO and DCE data, for each of the 5 EQ-5D-5L dimensions. Multivariate latent class models allowed separate but associated latent classes for the DCE and TTO data. Results. Correlation between the random effects for the 2 tasks ranged from -0.12 to 0.75, with only pain/discomfort and anxiety/depression having at least a 50% posterior probability of strong (>0.6) correlation. Latent classes for the TTO and DCE data both featured 1 latent class capturing participants attaching large disutilities to pain/discomfort, another capturing participants attaching large disutility to anxiety/depression, and the third class capturing the remainder. Agreement in class membership was poor (κ coefficient: 0.081; 95% credible interval, 0.033-0.13). Fewer respondents expressed strong disutilities for problems with anxiety/depression or pain/discomfort in the TTO than the DCE data (17% v. 55%, respectively). Conclusions. Stated preferences using TTO and DCEs show association across dimensions but poor agreement at the level of individual health states within respondents. Joint models that assume agreement between DCE and TTO have been used to develop national value sets for the EQ-5D-5L. This work indicates that when combining data from both techniques, methods requiring association but not agreement are needed.

Valuation of Health States Considered to Be Worse Than Death-An Analysis of Composite Time Trade-Off Data From 5 EQ-5D-5L Valuation Studies

OBJECTIVES

To evaluate the discriminative ability of negative values measured in 5-level EuroQol 5-dimensional questionnaire (EQ-5D-5L) value set studies.

METHODS

This is a secondary analysis of EQ-5D-5L value set studies from Singapore, the Netherlands, China, Thailand, and Canada in which health state values were elicited from a general population sample using a composite time trade-off (TTO) method. Mean values were calculated for health states with same severity. The association between the mean values and severity was evaluated using Pearson correlation (r). A linear mixed model using severity as the fixed effect was fitted for values. The analyses were performed separately for positive values (from a conventional TTO for health states considered "better than death") and negative values (from a lead time TTO for health states considered "worse than death").

RESULTS

In Singapore (N = 1000; negative values 32.6%), the mean decreased with severity from 0.89 to 0.21 for positive values and increased with severity from -0.98 to -0.89 for negative values. The correlation between values and severity was much lower for negative values (r = -0.016) than for positive values (r = -0.614). The coefficient of severity in the linear mixed model for negative values was much smaller (coefficient = -0.009; pseudo-R² < 0.001) compared with the model for positive values (coefficient = -0.041; pseudo-R² = 0.337). Results using data sets from the other countries were similar.

CONCLUSIONS

Negative values are not associated with severity of health states in EQ-5D-5L valuation studies, suggesting poor discriminative ability of the lead time TTO method in valuing health states considered worse than death.

Estimating utilities/disutilities for high-risk metastatic hormone-sensitive prostate cancer (mHSPC) and treatment-related adverse events

PURPOSE

To capture UK societal health utility values for high-risk metastatic hormone-sensitive prostate cancer (mHSPC) and the disutility associated with treatment-related adverse events (AEs) to inform future cost-utility analyses.

METHODS

A literature review, and patient and clinical expert interviews informed the development of health states characterising mHSPC symptoms and the impact of treatment-related AEs on health-related quality of life (HRQL). Three base health states were developed describing a typical patient with high-risk mHSPC: receiving androgen deprivation therapy (ADT) [Base State 1]; receiving docetaxel plus ADT [Base State 2]; completed docetaxel and still receiving ADT whose disease has not yet progressed [Base State 3]. Six additional health states described treatment-related AEs. The health states were validated with experts and piloted with general public participants. Health state utilities were obtained using the time trade-off (TTO) method with 200 members of the UK general population. A generalised estimating equation (GEE) model was used to estimate disutility weights.

RESULTS

Mean TTO scores for Base State 1 to 3 were 0.71 (SD = 0.26), 0.64 (SD = 0.27), and 0.68 (SD = 0.26), respectively, indicating that receiving docetaxel plus ADT was most impactful on HRQL. The GEE model indicated when compared to Base State 2 that the nausea and vomiting AE had the most impact on HRQL (- 0.21), while alopecia was least burdensome (- 0.04).

CONCLUSIONS

The study highlights the differences in utility between base health states and the significant impact of treatment-related AEs on the HRQL of patients with mHSPC. These findings underline the importance of accounting for impaired HRQL when assessing treatments for mHSPC.

Handling Data Quality Issues to Estimate the Spanish EQ-5D-5L Value Set Using a Hybrid Interval Regression Approach

BACKGROUND

The Spanish five-level EuroQol five-dimensional questionnaire (EQ-5D-5L) valuation study was the first to use the EuroQol Valuation Technology protocol, including composite time trade-off (C-TTO) and discrete choice experiments (DCE). In this study, its investigators noticed that some interviewers did not fully explain the C-TTO task to respondents. Evidence from a follow-up study in 2014 confirmed that when interviewers followed the protocol, the distribution of C-TTO responses widened.

OBJECTIVES

To handle the data quality issues in the C-TTO responses by estimating a hybrid interval regression model to produce a Spanish EQ-5D-5L value set.

METHODS

Four different models were tested. Model 0 integrated C-TTO and DCE responses in a hybrid model and models 1 to 3 altered the interpretation of the C-TTO responses: model 1 allowed for censoring of the C-TTO responses, whereas model 2 incorporated interval responses and model 3 included the interviewer-specific protocol violations. For external validation, the predictions of the four models were compared with those of the follow-up study using the Lin's concordance correlation coefficient.

RESULTS

This stepwise approach to modeling C-TTO and DCE responses improved the concordance between the valuation and follow-up studies (concordance correlation coefficient: 0.948 [model 0], 0.958 [model 1], 0.952 [model 2], and 0.989 [model 3]). We recommend the estimates from model 3, because its hybrid interval regression model addresses the data quality issues found in the valuation study.

CONCLUSIONS

Protocol violations may occur in any valuation study; handling them in the analysis can improve external validity. The resulting EQ-5D-5L value set (model 3) can be applied to inform Spanish health technology assessments.

People in states worse than dead according to the EQ-5D UK value set: would they rather be dead?

Purpose

Quality-adjusted life years (QALYs) measure health by combining length and quality of life. QALYs constitute the effect side of incremental cost-effectiveness ratios, describing the results of health economic evaluations. The objectives of this study were to (1) investigate the prevalence of states worse than dead (SWD) when using the EuroQol-5D UK value set, and (2) to study to what extent SWDs are reasonable with a starting point in experience-based valuations of health states.

Methods

Data from a Swedish cross-sectional population survey were used. The survey was directed to 10,000 persons 65 years and older and its primary aim was to investigate the prevalence and consequences of chronic pain. The survey included questions reflecting life situation and well-being. Some of these were used in order to characterise people in SWD.

Results

SWD were found in 1.8% of the 6611 respondents. The prevalence of SWD increased with advancing age and was more common among women than men. The control questions used indicated that most of the persons being in SWD according to the EQ-5D UK value set most probably would not judge themselves to be in a SWD.

Conclusions

Though negative QALY-weights are not very common, they constitute a non-negligible part of health states in a Swedish population 65 years and older. Prevalence of SWD is higher among women than men and increases with age. From responses to other questions on well-being and life situation, there is reason to doubt the reasonableness of experience-based negative QALY-weights in many cases.

Assessing the Use of a Feedback Module to Model EQ-5D-5L Health States Values in Hong Kong

BACKGROUND

An international valuation protocol exists for obtaining societal values for each of the 3125 health states of the five-level EuroQol-five dimensions (EQ-5D-5L) questionnaire. A feedback module (FM) that can be related to theoretical models used in behavioral economics was recently included in this protocol.

OBJECTIVES

Our objective was to assess the impact of using an FM to estimate an EQ-5D-5L value set in Hong Kong.

METHODS

EQ-5D-5L health states were elicited using a composite time trade-off (C-TTO) and a discrete-choice (DC) experiment. Use of the FM according to participant characteristics and the impact of the FM on the number of inconsistent C-TTO responses were assessed. We employed a main-effects hybrid model that combined data from both elicitation techniques.

RESULTS

In total, 1014 individuals completed the survey. The sample was representative of the general Chinese Hong Kong population in terms of sex, educational attainment, marital status, and most age groups but not for employment status. The use of the FM reduced the number of C-TTO inconsistencies. Participant characteristics differed significantly between those who used and did not use the FM. The model without a constant resulted in logical consistent coefficients and was therefore selected as the model to produce the value set. The predicted EQ-5D-5L values ranged from -0.864 to 1.

CONCLUSIONS

The use of an FM to allow participants to exclude C-TTO responses reduced the number of inconsistent responses and improved the quality of the data when estimating an EQ-5D-5L value set in Hong Kong.

A Computer-Assisted Personal Interview App in Research Electronic Data Capture for Administering Time Trade-off Surveys (REDCap): Development and Pretest

Background

The time trade-off (TTO) task is a method of eliciting health utility scores, which range from 0 (equivalent to death) to 1 (equivalent to perfect health). These scores numerically represent a person’s health-related quality of life. Software apps exist to administer the TTO task; however, most of these apps are poorly documented and unavailable to researchers.

Objective

To fill the void, we developed an online app to administer the TTO task for a research study that is examining general public proxy health-related quality of life estimates for persons with Alzheimer’s disease. This manuscript describes the development and pretest of the app.

Methods

We used Research Electronic Data Capture (REDCap) to build the TTO app. The app’s modular structure and REDCap’s object-oriented environment facilitated development. After the TTO app was built, we recruited a purposive sample of 11 members of the general public to pretest its functionality and ease of use.

Results

Feedback from the pretest group was positive. Minor modifications included clarity enhancements, such as rearranging some paragraph text into bullet points, labeling the app to delineate different question sections, and revising or deleting text. We also added a research question to enable the identification of respondents who know someone with Alzheimer’s disease.

Conclusions

We developed an online app to administer the TTO task. Other researchers may access and customize the app for their own research purposes.

Valuing health-related quality of life: An EQ-5D-5L value set for England

A new version of the EQ-5D, the EQ-5D-5L, is available. The aim of this study is to produce a value set to support use of EQ-5D-5L data in decision-making. The study design followed an international research protocol. Randomly selected members of the English general public completed 10 time trade-off and 7 discrete choice experiment tasks in face-to-face interviews. A 20-parameter hybrid model was used to combine time trade-off and discrete choice experiment data to generate values for the 3,125 EQ-5D-5L health states. Valuation data are available for 996 respondents. Face validity of the data has been demonstrated, with more severe health states generally given lower values. Problems with pain/discomfort and anxiety/depression received the greatest weight. Compared to the existing EQ-5D-3L value set, there are considerably fewer "worse than dead" states (5.1%, compared with over one third), and the minimum value is higher. Values range from -0.285 (extreme problems on all dimensions) to 0.950 (for health states 11211 and 21111). Results have important implications for users of the EQ-5D-5L both in England and internationally. Quality-adjusted life year gains from interventions seeking to improve very poor health may be smaller using this value set and may previously have been overestimated.

Valuation and Modeling of EQ-5D-5L Health States Using a Hybrid Approach

Supplemental Digital Content is available in the text.

Background:

The EQ-5D instrument is the most widely used preference-based health-related quality of life questionnaire in cost-effectiveness analysis of health care technologies. Recently, a version called EQ-5D-5L with 5 levels on each dimension was developed. This manuscript explores the performance of a hybrid approach for the modeling of EQ-5D-5L valuation data.

Methods:

Two elicitation techniques, the composite time trade-off, and discrete choice experiments, were applied to a sample of the Spanish population (n=1000) using a computer-based questionnaire. The sampling process consisted of 2 stages: stratified sampling of geographic area, followed by systematic sampling in each area. A hybrid regression model combining composite time trade-off and discrete choice data was used to estimate the potential value sets using main effects as starting point. The comparison between the models was performed using the criteria of logical consistency, goodness of fit, and parsimony.

Results:

Twenty-seven participants from the 1000 were removed following the exclusion criteria. The best-fitted model included 2 significant interaction terms but resulted in marginal improvements in model fit compared to the main effects model. We therefore selected the model results with main effects as a potential value set for this methodological study, based on the parsimony criteria. The results showed that the main effects hybrid model was consistent, with a range of utility values between 1 and −0.224.

Conclusion:

This paper shows the feasibility of using a hybrid approach to estimate a value set for EQ-5D-5L valuation data.

Quality Control Process for EQ-5D-5L Valuation Studies

BACKGROUND

The values of the five-level EuroQol five-dimensional questionnaire (EQ-5D-5L) are elicited using composite time trade-off and discrete choice experiments. Unfortunately, data quality issues and interviewer effects were observed in the first few EQ-5D-5L valuation studies. To prevent these issues from occurring in later studies, the EuroQol Group established a cyclic quality control (QC) process.

OBJECTIVES

To describe this QC process and show its impact on data quality.

METHODS

A newly developed QC tool provided information about protocol compliance, interviewer effects, and mean values by health state severity. In a cyclic process, this information is initially used to evaluate whether new interviewers meet minimal quality requirements and later to provide feedback about how their performance may be improved. To investigate the impact of this cyclic process, we compared the quality of the data in Dutch and Spanish valuation studies that did not have this QC process with that in the follow-up studies in the same countries that used the QC process. Data quality was measured using protocol violations, variability between interviewers, the proportion of inconsistent responders, and clustering of composite time trade-off values.

RESULTS

In Spain, protocol violations were reduced from 87% in the valuation study to 5% in the follow-up study and in the Netherlands from 20% to 8%. In both countries, interviewers performed more homogeneously in the follow-up studies. The number of inconsistent respondents was reduced by 23.2% in Spain and 23.6% in the Netherlands. Values were less clustered in the follow-up studies.

CONCLUSIONS

The implementation of a strict QC process in EQ-5D-5L valuation studies increases interviewer protocol compliance and promotes data quality.

EuroQol Protocols for Time Trade-Off Valuation of Health Outcomes

The time trade-off (TTO) valuation technique is widely used to determine utility values of health outcomes to inform quality-adjusted life-year (QALY) calculations for use in economic evaluation. Protocols for implementing TTO vary in aspects such as the trade-off framework, iteration procedure and its administration model and method, training of respondents and interviewers, and quality control of data collection. The most widely studied and utilized TTO valuation protocols are the Measurement and Valuation of Health (MVH) protocol, the Paris protocol and the EuroQol Valuation Technology (EQ-VT) protocol, all developed by members of the EuroQol Group. The MVH protocol and its successor, the Paris protocol, were developed for valuation of EQ-5D-3L health states. Both protocols were designed for a trained interviewer to elicit preferences from a respondent using the conventional TTO framework with a fixed time horizon of 10 years and an iteration procedure combining bisection and titration. Developed for valuation of EQ-5D-5L health states, the EQ-VT protocol adopted a composite TTO framework and made use of computer technology to facilitate data collection. Training and monitoring of interviewers and respondents is a pivotal component of the EQ-VT protocol. Research is underway aiming to further improve the EuroQol protocols, which form an important basis for the current practice of health technology assessment in many countries.

Underestimation of Variance of Predicted Health Utilities Derived from Multiattribute Utility Instruments

BACKGROUND

Parameter uncertainty in value sets of multiattribute utility-based instruments (MAUIs) has received little attention previously. This false precision leads to underestimation of the uncertainty of the results of cost-effectiveness analyses. The aim of this study is to examine the use of multiple imputation as a method to account for this uncertainty of MAUI scoring algorithms.

METHOD

We fitted a Bayesian model with random effects for respondents and health states to the data from the original US EQ-5D-3L valuation study, thereby estimating the uncertainty in the EQ-5D-3L scoring algorithm. We applied these results to EQ-5D-3L data from the Commonwealth Fund (CWF) Survey for Sick Adults ( n = 3958), comparing the standard error of the estimated mean utility in the CWF population using the predictive distribution from the Bayesian mixed-effect model (i.e., incorporating parameter uncertainty in the value set) with the standard error of the estimated mean utilities based on multiple imputation and the standard error using the conventional approach of using MAUI (i.e., ignoring uncertainty in the value set).

RESULT

The mean utility in the CWF population based on the predictive distribution of the Bayesian model was 0.827 with a standard error (SE) of 0.011. When utilities were derived using the conventional approach, the estimated mean utility was 0.827 with an SE of 0.003, which is only 25% of the SE based on the full predictive distribution of the mixed-effect model. Using multiple imputation with 20 imputed sets, the mean utility was 0.828 with an SE of 0.011, which is similar to the SE based on the full predictive distribution.

CONCLUSION

Ignoring uncertainty of the predicted health utilities derived from MAUIs could lead to substantial underestimation of the variance of mean utilities. Multiple imputation corrects for this underestimation so that the results of cost-effectiveness analyses using MAUIs can report the correct degree of uncertainty.

Comparison of Value Set Based on DCE and/or TTO Data: Scoring for EQ-5D-5L Health States in Japan

BACKGROUND

The valuation study of the five-level version of the EuroQol five-dimensional questionnaire (EQ-5D-5L) involved composite time trade-off (cTTO) and a discrete choice experiment (DCE). The DCE scores must be anchored to the quality-of-life scale from 0 (death) to 1 (full health). Nevertheless, the characteristics of the statistical methods used for converting the EQ-5D-5L DCE results by using TTO information are not yet clearly known.

OBJECTIVES

To present the Japanese DCE value set of the EQ-5D-5L and compare three methods for converting latent DCE values.

METHODS

The survey sampled the general population at five locations in Japan. 1098 respondents were stratified by age and sex. To obtain and compare the value sets of the EQ-5D-5L, the cTTO and DCE data were analyzed by a linear mixed model and conditional logit, respectively. The DCE scores were converted to the quality-of-life scale by anchoring to the worst state using cTTO, mapping DCE onto cTTO, and a hybrid model.

RESULTS

The data from 1026 respondents were analyzed. All the coefficients in the cTTO and DCE value sets were consistent throughout all the analyses. Compared with the cTTO algorithm, the mapping and hybrid methods yielded very similar scoring coefficients. The hybrid model results, however, produced a lower root mean square error and fewer health states with errors exceeding 0.05 than did the other models. The DCE anchored to the worst state overestimated the cTTO scores of almost all the health states.

CONCLUSIONS

Japanese value sets based on DCE were demonstrated. On comparing the observed cTTO scores, we found that the hybrid model was slightly superior to the simpler methods, including the TTO model.

Influenced from the start: anchoring bias in time trade-off valuations

Purpose

The de facto standard method for valuing EQ-5D health states is the time trade-off (TTO), an iterative choice procedure. The TTO requires a starting point (SP), an initial offer of time in full health which is compared to a fixed offer of time in impaired health. From the SP, the time in full health is manipulated until preferential indifference. The SP is arbitrary, but may influence respondents, an effect known as anchoring bias. The aim of the study was to explore the potential anchoring effect and its magnitude in TTO experiments.

Methods

A total of 1249 respondents valued 8 EQ-5D health states in a Web study. We used the lead time TTO (LT-TTO) which allows eliciting negative and positive values with a uniform method. Respondents were randomized to 11 different SPs. Anchoring bias was assessed using OLS regression with SP as the independent variable. In a secondary experiment, we compared two different SPs in the UK EQ-5D valuation study TTO protocol.

Results

A 1-year increase in the SP, corresponding to an increase in TTO value of 0.1, resulted in 0.02 higher recorded LT-TTO value. SP had little impact on the relative distance and ordering of the eight health states. Results were similar to the secondary experiment.

Conclusion

The anchoring effect may bias TTO values. In this Web-based valuation study, the observed anchoring effect was substantial. Further studies are needed to determine whether the effect is present in face-to-face experiments.

How different are composite and traditional TTO valuations of severe EQ-5D-5L states?

OBJECTIVE

Different variants of time trade-off (TTO) have been employed to elicit health state preferences and to create value sets for preference-based instruments. We compared composite TTO (cTTO) with traditional TTO (tTTO) in valuing severe EQ-5D-5L health states.

METHODS

cTTO uses tTTO to elicit values for health states better than dead and the lead-time TTO for states worse than dead. Eighteen severe states were valued using both cTTO and tTTO. Participants meeting predefined inconsistency criteria were excluded from the analyses. Histograms were used to examine the distributions of cTTO and tTTO values. Mean difference between the cTTO and tTTO values was calculated. Bland-Altman plots were used to examine the agreement between the cTTO and tTTO values for each health state. We used a logistic mixed effects model with random intercepts to identify variables that were associated with the directional change between the two TTO values.

RESULTS

A total of 1024 participants were included in the analysis with the mean age (SD) being 47.1 (17.4) years and 54.9 % female. For cTTO, 25 % of the values clustered at zero and there were few values between 0 and -0.5. In contrast, tTTO had fewer values at zero and more falling between -0.5 and 0. The distribution of positive values was similar between cTTO and tTTO. For worse than dead health states, the cTTO values tended to be higher than the tTTO values. In the logistic mixed effects model, those who did not agree that it was easy to understand the cTTO questions more likely changed from positive values in cTTO to zero or negative values in tTTO or change from zero cTTO values to negative values in tTTO compared with those who agreed (odds ratio 1.314, p = 0.037).

CONCLUSION

cTTO is an appealing technique in eliciting health state preferences, but further evidence is needed for its performance in valuing EQ-5D health states on a wide spectrum of health state severity.

An Empirical Study of Two Alternative Comparators for Use in Time Trade-Off Studies

BACKGROUND

Studies to produce value sets for preference-based measures of health require definition of a full health upper anchor if the values are to be used to calculate quality-adjusted life years. Recent value sets derived for the EQ-5D-5L instrument have described the upper anchor as "full health," whereas older valuation studies for the EQ-5D used "best health state" in the descriptive system (11111). It is unclear whether this change could have led to differences in the values obtained. The objective of this study was to assess differences in time trade-off (TTO) valuations using two different comparators (full health and 11111).

METHODS

Preferences for EQ-5D-5L health states were elicited from a broadly representative sample of the UK general public. TTO data were collected by using computer-assisted personal interviews. Respondents were randomly allocated to one of two arms, each using a different comparator health state. Respondents completed 10 or 11 TTO valuations and a series of follow-up questions examining their interpretations of the term "full health."

RESULTS

Interviews with 443 respondents were completed in 2014. The differences in mean values across arms are mostly small and nonsignificant. The two arms produced data of similar quality. There is evidence of interviewer effects. Health state 11111 was given a value of 1 by 98.2% of the respondents who valued it.

CONCLUSIONS

EQ-5D-5L values elicited by using the composite TTO approach are not greatly affected by the use of full health or 11111 as the comparator health state.

Values for health states with different life durations

BACKGROUND

Recent research suggests that the value of health states may depend on their duration, implying a more complex relationship between quality and quantity of life than the standard linear relationship assumed in the quality-adjusted life years (QALY) model.

OBJECTIVE

This study models how duration affects the value of health states, using discrete choice data.

METHODS

A study was conducted comprising health profiles including various durations. Health states were described with the EQ-5D. Duration was introduced as a sixth domain with six levels (1, 5, 10, 15, 30, 50 years). Sixty pairs of health profiles were selected, applying a Bayesian approach. Two hundred eight university students self-completed the computerized response tasks. Data were analyzed with a conditional logistic regression model.

RESULTS

There is clear evidence for non-linear values for duration, that is, preferences for duration are more accurately described by a logarithmic function.

CONCLUSIONS

In discrete choice experiments using health states with different durations, these data present the first evidence for decreasing marginal utility for duration. Our results suggest that refinement of the standard QALY framework may be needed for states with a long duration.

Population preference values for health states in relapsed or refractory B-precursor acute lymphoblastic leukemia in the United Kingdom

BACKGROUND

To date, reliable and comprehensive health-related quality of life data for patients with relapsed or refractory B-precursor acute lymphoblastic leukemia (ALL) have not been collected in clinical trials of the disease, and no utility studies have been published. The purpose of this study was to define and validate health states experienced by adults with relapsed/refractory B-precursor ALL, and to assign utility values to these health states using time-trade off methodology.

METHODS

This study was conducted in the UK in three phases. In the first phase, five health state descriptions were developed based on a recent clinical trial. The second phase validated the health state descriptions with clinicians and patients with experience of relapsed/refractory B-precursor ALL. The third phase involved prospective health state valuation using time-trade off methodology in a sample of the general public. The study was approved by the UK National Health Service Research Ethics Committee.

RESULTS

In total, 123 participants were recruited and included in the final analysis; all participants gave written, informed consent. Complete remission was the most preferred health state (mean utility [SEM], 0.86 [0.01]), followed by complete remission with partial hematological recovery (with minimal risk of bleeding or developing infection) (0.75 [0.02]); aplastic bone marrow (0.59 [0.02]); partial remission (0.50 [0.03]); and progressive disease (0.30 [0.04]).

CONCLUSIONS

This is the first study to report utility values for health states associated with relapsed/refractory B-precursor ALL. It was designed and conducted to align with NICE guidance on alternative methods for generating health state utility values when EQ-5D data are either unavailable or inappropriate. These utilities can be applied in future cost-effectiveness analyses of treatment for relapsed/refractory B-precursor ALL.

An Investigation of the Feasibility and Cultural Appropriateness of Stated Preference Methods to Generate Health State Values in the United Arab Emirates

BACKGROUND

No five-level EuroQol five-dimensional questionnaire (EQ-5D-5L) value sets are currently available in the Middle East to inform decision making in the region's health care systems.

OBJECTIVES

To test the feasibility of eliciting EQ-5D-5L values from a general public sample in the United Arab Emirates (UAE) using the EuroQol Group's standardized valuation protocol.

METHODS

Values were elicited in face-to-face computer-assisted personal interviews. Adult Emiratis were recruited in public places. Respondents completed 10 time trade-off tasks and 7 discrete choice experiment tasks, followed by debriefing questions about their experience of completing the valuation tasks. Descriptive analyses were used to assess the face validity of the data.

RESULTS

Two hundred respondents were interviewed in December 2013. The face validity of the data appears to be reasonably high. Mean time trade-off values ranged from 0.81 for the mildest health state (21111) to 0.19 for the worst health state in the EQ-5D-5L descriptive system (55555). Health states were rarely valued as being worse than dead (6.2% of all observations; 10% of all valuations of 55555). In a rationality check discrete choice experiment task whereby a health state (55554) was compared with another that logically dominated it (55211), 99.5% of the respondents chose the dominant option. Most of the respondents stated that their religious beliefs influenced their responses to the valuation tasks.

CONCLUSIONS

Our results suggest that it is feasible to generate meaningful health-state values in the UAE, though some adaptation of the methods may be required to improve their acceptability in the UAE (and other countries with predominantly Arab and/or Muslim populations).

A framework for estimating health state utility values within a discrete choice experiment: modeling risky choices

BACKGROUND

There has been recent interest in using the discrete choice experiment (DCE) method to derive health state utilities for use in quality-adjusted life year (QALY) calculations, but challenges remain.

OBJECTIVES

We set out to develop a risk-based DCE approach to derive utility values for health states that allowed 1) utility values to be anchored directly to normal health and death and 2) worse than dead health states to be assessed in the same manner as better than dead states. Furthermore, we set out to estimate alternative models of risky choice within a DCE model.

METHOD

A survey was designed that incorporated a risk-based DCE and a "modified" standard gamble (SG). Health state utility values were elicited for 3 EQ-5D health states assuming "standard" expected utility (EU) preferences. The DCE model was then generalized to allow for rank-dependent expected utility (RDU) preferences, thereby allowing for probability weighting. A convenience sample of 60 students was recruited and data collected in small groups.

RESULTS

Under the assumption of "standard" EU preferences, the utility values derived within the DCE corresponded fairly closely to the mean results from the modified SG. Under the assumption of RDU preferences, the utility values estimated are somewhat lower than under the assumption of standard EU, suggesting that the latter may be biased upward.

CONCLUSION

Applying the correct model of risky choice is important whether a modified SG or a risk-based DCE is deployed. It is, however, possible to estimate a probability weighting function within a DCE and estimate "unbiased" utility values directly, which is not possible within a modified SG. We conclude by setting out the relative strengths and weaknesses of the 2 approaches in this context.

The better than dead method: feasibility and interpretation of a valuation study

BACKGROUND

Traditionally, the valuation of health states worse than being dead suffers from two problems: [1] the use of different elicitation methods for positive and negative values, necessitating arbitrary transformations to map negative to positive values; and [2] the inability to quantify that values are time dependent. The Better than Dead (BTD) method is a health-state valuation method where states with a certain duration are compared with being dead. It has the potential to overcome these problems.

OBJECTIVES

To test the feasibility of the BTD method to estimate values for the EQ-5D system.

METHODS

A representative sample of 291 Dutch respondents (aged 18-45 years) was recruited. In a web-based questionnaire, preferences were elicited for a selection of 50 different health states with six durations between 1 and 40 years. Random-effects models were used to estimate the effects of socio-demographic and experimental variables, and to estimate values for the EQ-5D. Test-retest reliability was assessed in 41 respondents.

RESULTS

Important determinants for BTD were a religious life stance [odds ratio 4.09 (2.00-8.36)] and the educational level. The fastest respondents more often preferred health-state scenarios to being dead and had lower test-retest reliability (0.45 versus 0.77 and 0.84 for fast, medium and slow response times, respectively). The results showed a small number of so-called maximal endurable time states.

CONCLUSION

Valuating health states using the BTD method is feasible and reliable. Further research should explore how the experimental setting modifies how values depend on time.

The lead time tradeoff: the case of health states better than dead

BACKGROUND

Lead time tradeoff (L-TTO) is a variant of the time tradeoff (TTO). L-TTO introduces a lead period in full health before illness onset, avoiding the need to use 2 different procedures for states better and worse than dead. To estimate utilities, additive separability is assumed. We tested to what extent violations of this assumption can bias utilities estimated with L-TTO.

METHODS

A sample of 500 members of the Spanish general population evaluated 24 health states, using face-to-face interviews. A total of 188 subjects were interviewed with L-TTO and the rest with TTO. Both samples evaluated the same set of 24 health states, divided into 4 groups with 6 health states per set. Each subject evaluated 1 of the sets. A random effects regression model was fitted to our data. Only health states better than dead were included in the regression since it is in this subset where additive separability can be tested clearly.

RESULTS

Utilities were higher in L-TTO in relation to TTO (on average L-TTO adds about 0.2 points to the utility of health states), suggesting that additive separability is violated. The difference between methods increased with the severity of the health state. Thus, L-TTO adds about 0.14 points to the average utility of the less severe states, 0.23 to the intermediate states, and 0.28 points to the more severe estates.

CONCLUSIONS

L-TTO produced higher utilities than TTO. Health problems are perceived as less severe if a lead period in full health is added upfront, implying that there are interactions between disjointed time periods. The advantages of this method have to be compared with the cost of modeling the interaction between periods.

A program of methodological research to arrive at the new international EQ-5D-5L valuation protocol

OBJECTIVES

To describe the research that has been undertaken by the EuroQol Group to improve current methods for health state valuation, to summarize the results of an extensive international pilot program, and to outline the key elements of the five-level EuroQol five-dimensional (EQ-5D-5L) questionnaire valuation protocol, which is the culmination of that work.

METHODS

To improve on methods of health state valuation for the EQ-5D-5L questionnaire, we investigated the performance of different variants of time trade-off and discrete choice tasks in a multinational setting. We also investigated the effect of three modes of administration on health state valuation: group interviews, online self-completion, and face-to-face interviews.

RESULTS

The research program provided the basis for the EQ-5D-5L questionnaire valuation protocol. Two different types of tasks are included to derive preferences: a newly developed composite time trade-off task and a forced-choice paired comparisons discrete choice task. Furthermore, standardized blocked designs for the selection of the states to be valued by participants were created and implemented together with all other elements of the valuation protocol in a digital aid, the EuroQol Valuation Technology, which was developed in conjunction with the protocol.

CONCLUSIONS

The EuroQol Group has developed a standard protocol, with accompanying digital aid and interviewer training materials, that can be used to create value sets for the EQ-5D-5L questionnaire. The use of a well-described, consistent protocol across all countries enhances the comparability of value sets between countries, and allows the exploration of the influence of cultural and other factors on health state values.

Deriving time discounting correction factors for TTO tariffs

The Time Trade-off (TTO) method is a popular method for valuing health state utilities and is frequently used in economic evaluations. However, this method produces utilities that are distorted by several biases. One important bias entails the failure to incorporate time discounting. This paper aims to measure time discounting for health outcomes in a sample representative for the general population. In particular, we estimate TTO scores alongside time discounting in order to derive a set of correction factors that can be employed to correct raw TTO scores for the downward bias caused by time discounting. We find substantial positive correction factors, which are increasing with the severity of the health state. Furthermore, higher discounting is found when using more severe health states in the discounting elicitation task. More research is needed to further develop discount rate elicitation procedures and test their validity, especially in general public samples. Moreover, future research should investigate the correction of TTO values for other biases as well, such as loss aversion, and to develop a criterion to test the external validity of TTO scores.

Eliciting preferences to the EQ-5D-5L health states: discrete choice experiment or multiprofile case of best-worst scaling?

Choice-based methods have been used widely in assessing healthcare programs. This study compared the binary discrete choice experiment (DCE) and the multiprofile case of best-worst scaling (BWS) in eliciting preferences for the EQ-5D-5L. Forty-eight EQ-5D-5L health states were selected using a Bayesian efficient design and grouped into 24 pairs for the DCE tasks and 8 sets for the BWS tasks (each set has three health states). A total of 100 participants completed 12 pairs and 8 sets in a random order. A probit regression model and ranked order logistic regression model were used to estimate the latent utilities from the DCE and BWS, respectively. Both tasks were well understood by the majority of participants. The DCE tasks were relatively easier and took a shorter time to complete. The intraclass correlation coefficient (ICC) of the DCE was higher than that of the BWS. The variances associated with the latent utilities estimated from the DCE were larger than those from the BWS. The DCE is more feasible and reliable than the BWS in valuing the EQ-5D-5L. Future studies could focus on comparing the consistency and accuracy of these techniques in predicting the health utilities of the EQ-5D-5L.

Qualitative and Quantitative Analysis of Interviewer Help Answering the Time Tradeoff

OBJECTIVES

The time tradeoff (TTO) is an important method to directly obtain health utilities. Challenges of the TTO are, among others, "nontraders" and illogical answers. In TTO interviews, these challenges are resolved by the interviewer. In web-based TTOs, training procedures and logical checks are used based on the views of the researchers. As web-based TTOs will be used more often in the future, we investigated how respondents arrive at their ratings to determine the help they require.

METHODS

In 2 earlier studies performed by this research group, respondents valued 6 EQ-5D states on a TTO. Respondents were asked to think out loud, and all interviews were audiotaped. A random selection of these interviews were transcribed and double-coded by two independent raters, using a priori and inductive coding until saturation was reached. Based on the retrieved mistakes and comments, a list of frequently asked questions (FAQ) was developed.

RESULTS

In total, 91 interviews were coded. In all, 85% made at least 1 mistake, 41% made a misreading/miscalculation, 70% misunderstood the tradeoff, 27% misunderstood the EQ-5D dimensions, 29% misunderstood the scenario, 45% made a comment about the TTO, and 43% expressed frustration. More misunderstandings were reported in the Peeters study, which was performed in a realistic setting, whereas the van Osch study was conducted in a more ideal setting. Misunderstandings of the scenario were mosly reported by patients.

CONCLUSIONS

Almost all respondents need interviewer help. This may have implications for the validity of interviewer-based TTO elicitations when social acceptability bias is an issue or with explicit hypothesis and the interviewer is not blinded. The FAQ list can be used to standardize interviewer help or as a help function in a web-based TTO.