Person Trade-off Research Articles

Toward a more universal approach in health valuation

By polling individual responses to hypothetical scenarios, valuation studies estimate population preferences toward health on a quality-adjusted life year (QALY) scale. The scenarios typically involve trade-offs in time (time trade-off (TTO)), risk (standard gamble (SG)), or number of persons affected (person trade-off (PTO)). This paper revisits the QALY assumptions and provides a coherent health econometric approach that unites TTO, SG, and PTO techniques under a common estimator. The proposed approach avoids the use of ratio statistics in QALY estimation and the common convention of arbitrarily changing trade-off responses. As an example, 34% of the TTO responses from the seminal Measurement and Valuation of Health study were changed in the original UK analysis, which led to substantially lower QALY estimates. As a general rule, if the original estimate is less than 0.5 QALYs, add 0.25 QALYs to get the new estimates.

New disability weights for the global burden of disease

Since its launch in the early 1990s, the global burden of disease study has produced a highly influential set of findings on the impacts of different diseases, injuries and risk factors on population health.1–3 One important contribution of the study has been the development of the disability-adjusted life year (DALY) as a new metric for quantifying losses of healthy life, due either to premature mortality or to time lived in a state of reduced health. To account for both types of outcomes in a commensurable way, DALYs require a set of numerical weights attached to the wide array of non-fatal consequences from different diseases and injuries. The conceptualization and measurement of these “disability weights” have prompted considerable commentary and debate.4–6 At present, studies on the global, regional or national burden of disease continue to draw heavily on the set of disability weights derived in the 1996 revision of The global burden of disease.7 For the 1996 study, a panel of health professionals developed these disability weights in a deliberative group exercise. First they evaluated 22 indicator conditions using a method called the “person trade-off”. Based on the resulting values, the 22 conditions were then grouped into seven different classes of severity. For the remainder of the outcomes in the study (referred to as “disabling sequelae”), participants were asked to allocate a set of typical cases for a particular sequela across the seven classes. The indicator conditions within each class supplied a concrete operational definition for the average severity level in that class and thus marked points of reference on the numerical scale ranging from perfect health to death. Critics of the DALY have questioned aspects of this approach. The use of an expert panel has provoked suggestions for a more inclusive measurement exercise to represent the broader perspectives of communities and societies.4 Other researchers have asked whether there may be important variation in assessments of health outcomes across diverse cultures and environments,6,8,9 indicating a need for more cross-country measurement and comparison. Some critics have opposed, on ethical grounds, the use of the person trade-off technique as implemented in the 1996 group exercise.5 Over the past two years, with support from the Bill & Melinda Gates Foundation, a research consortium including the Institute for Health Metrics and Evaluation, Harvard University, Johns Hopkins University, University of Queensland and the World Health Organization has been leading a major revision of the global burden of disease, in collaboration with a network of hundreds of experts from around the world. An essential element in this study is a comprehensive re-estimation of disability weights for the full set of around 230 unique sequelae associated with the array of disease and injury causes in the study. The design of this new study responds to critiques of past efforts in several significant ways. With two major components of survey-based data collection, the new study uses simple paired-comparison questions. In these questions, respondents are asked to consider two hypothetical individuals characterized by different functional limitations, and to indicate which person they would regard as healthier. These simple questions permit input from a broad cross-section of respondents spanning diverse cultural, environmental and demographic circumstances. Straightforward analytic approaches allow inferences about cardinal weights at the population-level based on ordinal responses by individuals.10 The first component of the study comprises new population-based household surveys in six countries (Bangladesh, Indonesia, Peru, South Africa, the United Republic of Tanzania and the United States of America). Data from these surveys will deliver welcome empirical evidence on questions regarding the extent of variation in health assessments across different settings. The second component uses an open-access, internet-based survey, which is intended to expand its global reach. Presently available in English, eventual translation of the survey into other languages will further broaden participation. Design overlap with the household survey will allow direct comparison and evaluation of possible selection effects due to non-random sampling, while the web-based format and unrestricted access allow broader input from voices that are seldom heard in discussions on global health priorities. Reliable, valid and comparable measurement of non-fatal health outcomes is as essential as accurate measurement of mortality and causes of death in assessing the health of populations. Persistent challenges to existing disability weights demand further empirical study through data collection efforts that are broad-reaching and inclusive. This time around, everybody is invited to make his or her view count by participating in the online disability weights measurement survey available at: http://www.gbdsurvey.org

Test-retest reliability of health state valuation techniques: the time trade off and person trade off

Economic analysis is increasingly being employed in formal resource allocation decision-making processes in health care. As a consequence, the methods employed by economic analysts are increasingly subject to close scrutiny. One such area of methodology concerns the instruments used to elicit preferences for various health states for use in the construction of quality-adjusted life years. There are a number of techniques which may be used to elicit preferences and different techniques produce different results. The objective of this study was to explore the test-retest reliability of two techniques: Time Trade Off (TTO) and Person Trade Off (PTO) valuations were collected by a general population postal survey. A total of 798 respondents returned questionnaires. The intra class correlation coefficients ranged from 0.40 to 0.88 for TTO and, -0.17 to 0.82 for PTO, with the majority of coefficients being >0.50. The reliability coefficients varied between techniques and health states, with the TTO technique tending to produce higher coefficients. While the reliability results for TTO were generally positive, the reliability results for PTO are less clear.

Severity as an independent determinant of the social value of a health service

This paper has two objectives, first to review the relevant literature concerning the social importance of severity of pre-treatment condition, and second to present the results of a new analysis of the relationship between social value, individual assessment of health improvement and the severity of illness. The present study differs methodologically from others reported in the literature. The underlying hypothesis is that members of the public have an aversion to patients being in a severe health state irrespective of the reason for their being there, and that this aversion will affect the social valuation of a health program after taking account of the magnitude of the health improvement. This effect will be observable in a program which (compared to another) takes a person out of a severe health state--the usual case discussed in the literature--or in a program which (compared to another) leaves a person in a severe health state. The present study tests this second implication of the hypothesis. We present data consistent with the view that after taking account of health improvement, health programs are preferred which do not leave people in severe health states. Alternative explanations are considered and particularly the possibility that data reflect a social preference for individuals achieving their health potential. Both explanations imply the need to reconsider the rules for prioritizing programs. In this analysis, Person Trade-Off (PTO) scores are used to measure social preferences ('value' or 'social utility') and Time Trade-Off (TTO) scores are used to measure individual assessments of health improvement and initial severity. Econometric results suggest that severity is highly significant and may more than double the index of social value of a health service.

Quantifying the Value of Stroke Disability Outcomes

Background and Purpose— The modified Rankin Scale (mRS) categorizes poststroke disability among 7 broad, ordinal grades, but the interval distances between these levels are spaced along the disability spectrum have not been previously investigated. Methods— We used the person trade-off procedure developed by the World Health Organization Global Burden of Disease Project (WHO-GBDP) to generate disability weights (DWs) ranging from 0 (normal) to 1 (dead) for each of 7 mRS grades. The ratings of an international, 9-member panel of stroke experts were combined by a modified Delphi process. Results— DWs (95% CI) were 0 for mRS 0, 0.046 (0.004 to 0.088) for mRS 1, 0.212 (0.175 to 0.250) for mRS 2, 0.331 (0.292 to 0.371) for mRS 3, 0.652 (0.625 to 0.678) for mRS 4, 0.944 (0.873 to 1.015) for mRS 5, and 1.0 for mRS 6. DWs of adjacent mRS levels were significantly different ( P <0.001 for all). Coefficients of variation showed a high degree of consensus for DWs among panel members. DWs placed each of the 5 intermediate mRS states in different disability class levels of the WHO-GBDP anchor conditions and identified natural clusters to use when reducing the mRS to fewer categories. Conclusions— Formal DW assignment confirms that the mRS is an ordered but unequally spaced scale. The availability of DWs for each mRS level now permits direct comparison of each poststroke outcome state with the outcomes of hundreds of other diseases in the WHO-GBDP and the expression of stroke burden in different populations by using the uniform metric of disability-adjusted life-years lost.

When are person tradeoffs valid?

The person tradeoff (PTO) is commonly used in health economic applications. However, to date it has no theoretical basis. The purpose of this paper is to provide this basis from a set of assumptions that together justify the most common applications of the PTO method. Our analysis identifies the central assumptions in PTO measurements. We test these assumptions in an experiment, but find only limited support for the validity of the PTO.

Aversion to geographic inequality and geographic variation in preferences in the context of healthcare

It is now widely accepted that social values have a role to play in terms of healthcare resource allocation decisions. Furthermore, with the recent focus on decentralization, geographic variation in preferences has become a pertinent issue. However, the geographic dimension of inequality in the distribution of health gains across different regions has received little attention in the literature. To elicit the general public's preferences regarding inequality in the distribution of health gains across regions and to test the assumption of variation in preferences across regions. A questionnaire was developed to elicit social and personal preferences. The first preference questions focused on the criteria of maximization of health gain and geographic equality of health gain. Person trade-off questions were used to assess trade-offs and inequality aversion. The latter preference questions focused on attributes of healthcare delivery. The questionnaire was administered in two Portuguese municipalities. Samples from both localities were defined by quotas based on the composition of local populations. Overall, 70 respondents (half from each municipality) self-completed the questionnaire. Respondents agreed or strongly agreed that both maximization of health gain and geographic equality of health gain should be considered in resource allocation decisions; when they were asked to make trade-offs, 70-80% were prepared to do so. Of those making trade-offs, most people were willing to forego 10-30% of total health gain to keep geographic equality of health gain in return. Regarding inequality aversion, the results indicate that there was some aversion to inequality in the sample. Regarding preference variation, the results do not corroborate the hypothesis of variation in preferences across regions. These results suggest that the general public supports the principles of maximization of health gain and geographic equality in the distribution of health gain. The results further suggest the existence of aversion to inequality, subject to a maximum opportunity cost. Further research is required into preference variation; in particular, in larger samples, and in other regions of Portugal.

The lack of theoretical support for using person trade-offs in QALY-type models

Considerable support for the use of person trade-off methods to assess the quality-adjustment factor in quality-adjusted life years (QALY) models has been expressed in the literature. The WHO has occasionally used similar methods to assess the disability weights for calculation of disability-adjusted life years (DALYs). This paper discusses the theoretical support for the use of person trade-offs in QALY-type measurement of (changes in) population health. It argues that measures of this type based on such quality-adjustment factors almost always violate the Pareto principle, and so lack normative justification.

Eliciting people's preferences for the distribution of health: A procedure for a more precise estimation of distributional weights

In order to incorporate distributional concerns into cost-effectiveness analysis, it would be useful to elicit distributional weights that express people's valuation of marginal health gains at various levels of health. Distributional preferences are commonly elicited either through a person trade off (PTO) or a gain trade off (GTO) technique. An inherent problem of the GTO is that it is based on the valuation of non-marginal health gains. In practice, many contributions using the PTO also focus on non-marginal health gains. This paper demonstrates that the failure to distinguish appropriately between marginal and non-marginal health gains may lead to seriously misleading estimates of distributional weights. Moreover, the paper proposes a methodology for utilising information obtained through non-marginal analysis more efficiently in order to obtain more reliable estimates of distributional weights. The methodology was successfully applied in an empirical study of age weights.

Why People Refuse to Make Tradeoffs in Person Tradeoff Elicitations: A Matter of Perspective?

Person tradeoff (PTO) elicitations assess people's values for health states by asking them to compare the value of treatment programs. For example, people might be asked how many patients need to be cured of health condition X to equal the benefit of curing 100 people of condition Y. However, when faced with PTO elicitations, people frequently refuse to make quantifiable tradeoffs, exhibiting 2 kinds of refusals: 1) They say that 2 treatment programs have equal value, that curing 100 of X is just as good as curing 100 of Y, even if X is a less serious condition than Y, or 2) they say that the 2 programs are incomparable, that millions of people need to be cured of X to be as good as curing 100 of Y. The authors explore whether people would be more willing to make tradeoffs if the focus was changed from trading off groups of patients to choosing the best decision or evaluating treatment benefits. . Two randomized trials used diverse samples (N=2400) via the Internet to test for the effect of perspective on refusal rates. The authors predicted that perspectives that removed people from decision-making roles would increase their willingness make tradeoffs. Contrary to expectations, refusal rates increased when people were removed from decision-making roles. In fact, the more pressure put on people to make a decision, the less likely they were to refuse to make tradeoffs. To reduce PTO refusals, it is best to adopt a decision-maker perspective.

Person tradeoffs and the problem of risk

Recently, the person tradeoff method has received much attention because it directs one to value health improvements of those other than one’s self. As a result, several researchers have suggested that the person tradeoff measures preferences in pharmaceutical and health economic analysis more accurately than the standard gamble, which many consider the gold standard approach. However, the person tradeoff exercise as it is currently practiced assumes a riskless world where policy decisions lead to certain societal outcomes. In reality, health policy decisions are made under risk or uncertainty. Evidence suggests riskless and risky values differ for health improvements over persons. A problem with using the riskless person tradeoff method in economic evaluation of health is that the method conflates a change in the objects of measurement (i.e., a change from health gains with respect to self to gains with respect to others) with a change in the measurement context (i.e., a change from a risk context to a risk-free one). Therefore, it is wise to examine what is sacrificed by ignoring risk and uncertainty inherent in health policy decision making. Given that the choices we make in the context of risk are often different than those we make under certainty, the authors argue that if one assumes context is important, the measurement of health improvements over persons should then be carried out in the context of risk. The authors propose the application of standard methods from utility theory to carry out elicitations for valuing health improvements of others within the context of risky choices. In addition, an alternative perspective of utility is discussed that in comparison puts forth a context free notion of utility.

The impact of considering adaptation in health state valuation

Patients with chronic health conditions often rate their quality of life (QoL) significantly higher than non-patients. One explanation for this discrepancy is that non-patients focus on the negative aspects of the onset of a condition, especially the early difficulties people face when they first experience a debilitating condition, without considering that patients can adapt to it over time. To test this hypothesis, we had 359 people perform person tradeoff (PTO) elicitations in an online survey, varying whether the treatment programs under consideration saved the lives of patients (a) with pre-existing paraplegia; or (b) who would experience new onset of paraplegia. Half of each group completed an “adaptation exercise” which encouraged them to consider their own ability to emotionally adapt to negative events in general and specifically to having paraplegia. The adaptation manipulation increased the value participants placed on pre-existing paraplegia ( p=0.03) and on new onset paraplegia ( p=0.05), relative to saving healthy lives. Moreover, the adaptation exercise dramatically reduced the differences between evaluations of pre-existing and new onset paraplegia to values within 2% of each other. Our findings suggest that asking non-patients to do an adaptation exercise before giving QoL ratings may help close the gap in ratings between patients and citizen non-patients.

Measuring the health of populations: the veil of ignorance approach.

We report the results from two surveys designed to explore whether an application of Harsanyi's principle of choice form behind a veil of ignorance (VEI) can be used in order to measure the health of populations. This approach was tentatively recommended by Murray et al. (Bull. World Health Organ 2000; 78: 981-994; Summary Measures of population health: Concepts, Ethics, Measurement and Applications, WHO, 2002.) as an appropriate way of constructing summary measures of population health (SMPH) for comparative purposes. The operationalization of the VEI approach used in this paper was suggested by Nord (Summary Measures of Population Health: Concepts, Ethics, Measurement and Applications, WHO, 2002.). We test if VEI and person trade-off (PTO) methods generate similar quality-of-life weights. In addition, we compare VEI and PTO weights with individual utilities estimated by means of the conventional standard gamble (SG) and a variation of it we call double gamble. Finally, psychometric properties like feasibility, reliability, and consistency are examined. Our main findings are next: (1) VEI and PTO approaches generate very different weights; (2) it seems that differences between PTO and VEI are not due to the 'rule of rescue'; (3) the VEI resembled more a DG than a classical SG; (4) PTO, VEI, and DG exhibited good feasibility, reliability and consistency.

Valuing health care using willingness to pay: a comparison of the payment card and dichotomous choice methods

This paper compares willingness to pay (WTP) estimates generated from the dichotomous choice (DC) and payment card (PC) approaches. In a split-sample WTP experiment concerned with allocating scarce health care resources across three health care interventions, the DC approach is shown consistently to generate larger welfare estimates than the PC. Observed difference between PC and DC experiments cannot be explained by the inclusion of non-demanders or methods of statistical analysis but may be partly explained by “yea-saying”. No evidence of range bias or mid-point bias was found with PC responses. Data were also collected on respondents’ ordinal rankings of the three interventions and person-trade-offs (PTOs). Neither of these approaches converged with WTP. Future work must address the decision heuristics individuals employ when responding to valuation experiments.

Cross-national agreement on disability weights: the European Disability Weights Project.

BackgroundDisability weights represent the relative severity of disease stages to be incorporated in summary measures of population health. The level of agreement on disability weights in Western European countries was investigated with different valuation methods.MethodsDisability weights for fifteen disease stages were elicited empirically in panels of health care professionals or non-health care professionals with an academic background following a strictly standardised procedure. Three valuation methods were used: a visual analogue scale (VAS); the time trade-off technique (TTO); and the person trade-off technique (PTO). Agreement among England, France, the Netherlands, Spain, and Sweden on the three disability weight sets was analysed by means of an intraclass correlation coefficient (ICC) in the framework of generalisability theory. Agreement among the two types of panels was similarly assessed.ResultsA total of 232 participants were included. Similar rankings of disease stages across countries were found with all valuation methods. The ICC of country agreement on disability weights ranged from 0.56 [95% CI, 0.52–0.62] with PTO to 0.72 [0.70–0.74] with VAS and 0.72 [0.69–0.75] with TTO. The ICC of agreement between health care professionals and non-health care professionals ranged from 0.64 [0.58–0.68] with PTO to 0.73 [0.71–0.75] with VAS and 0.74 [0.72–0.77] with TTO.ConclusionsOverall, the study supports a reasonably high level of agreement on disability weights in Western European countries with VAS and TTO methods, which focus on individual preferences, but a lower level of agreement with the PTO method, which focuses more on societal values in resource allocation.

A multi-method approach to measuring health-state valuations.

Existing techniques for eliciting health-state valuations incorporate both strength of preferences for health states and other values such as risk aversion or time preference. This paper presents a new methodological approach that allows estimation of a set of core underlying health-state values based on responses elicited through multiple measurement techniques. A study was undertaken in which respondents completed the visual analogue (VAS) scale, time trade-off (TTO), standard gamble (SG) and person trade-off (PTO) for a range of states. By specifying flexible parametric functions to explain responses on each measurement technique, we estimated both the underlying strength of preference values for the health states in the study and the values for a set of auxiliary parameters characterising risk attitudes, discount rates, distributional concerns and scale distortion effects in the group of respondents. This study demonstrates that it is possible to understand responses on these four different measurement techniques based on a consistent set of core values. The approach presented here can provide insights into different sources of observed variation in VAS, TTO, SG and PTO responses and facilitate appropriate adjustment of valuations elicited through different methods for use in summary health measures and economic analyses.

Lack of multiplicative transitivity in person trade-off responses.

The person trade-off (PTO) is a technique for eliciting preferences for resource allocation across patient groups. In principle PTO responses should satisfy a requirement of multiplicative transitivity, i.e. that if people consider treatment of 1 in state A to be equivalent to treating 10 in state B, and 1 in state B to be equivalent to 10 in state C, then they should find 1 in state A equivalent to 100 in state C. Earlier studies addressing labelled diseases (specific diagnoses), have shown multiplicative intransitivity of the PTO responses. Our purpose was to test multiplicative transitivity in the case of health states described with the EuroQol instrument only and to find a possible framing effect such as the number of persons in the reference intervention. Forty-four master degree students were asked to fill in a questionnaire addressing four chronic health states. Their task consisted in (1). ranking the states by severity, (2). valuing each of them by the means of the time trade-off, and (3). doing the PTO for all the 10 possible pairwise combinations of the four chronic states plus a fatal one. In a subsequent questionnaire the number of persons in the reference intervention in the PTO was increased from 10 to 100. Multiplicative transitivity was studied in subjects who demonstrated a willingness to trade off and consistency in ranking individual values. None of the 39 subjects included satisfied a minimum multiplicative transitivity requirement in PTO responses. Internal consistency was not improved when the PTO involved health states close to each other in terms of severity, nor when the prevention of death was not the reference intervention. For the 22 subjects having answered both types of questionnaire, increasing the number of persons in the reference intervention did not improve multiplicative transitivity. The PTO holds promise as a useful method for determining social preferences for priority setting, inasmuch as it captures distributive concerns that individual utility techniques such as the time trade-off do not address. But the lack of multiplicative transitivity in PTO responses is unsatisfactory, and ways to reduce this problem need to be explored.

Preferences and person trade-offs: forcing consistency or inconsistency in health-related quality of life measures?

We consider assumptions about preferences implicit in the person trade-off exercises used to derive health-related quality of life measures for the Global and US Burden of Disease Projects. Because these methods and their results have the potential of being adopted by other researchers, a critical review of this methodology and its assumptions is warranted. Exercise participants are told that quality of life valuation is approached using two different questions to reveal logical inconsistencies in each person's responses. An inconsistency is claimed to exist if a participant's two responses violate a particular mathematical relationship, and participants are forced to modify their responses to satisfy that relationship. We demonstrate that this supposed 'logical' relationship need not hold for logically consistent, rational individuals, and we prove that the relationship will in fact hold only for a particular class of social value functions exhibiting two characteristics that may not be consistent with the preferences of some participants. These results imply that the forced modification may invalidate some responses, as it may require some participants to provide final answers that are inconsistent with their true preferences. We then discuss preference characterizations in the existing person trade-off literature, from which this relationship may have been derived.

The Rule of Rescue

Jonsen coined the term “Rule of Rescue”(RR) to describe the imperative people feel to rescue identifiable individuals facing avoidable death. In this paper we attempt to draw a more detailed picture of the RR, identifying its conflict with cost-effectiveness analysis, the preference it entails for identifiable over statistical lives, the shock–horror response it elicits, the preference it entails for lifesaving over non-lifesaving measures, its extension to non-life-threatening conditions, and whether it is motivated by duty or sympathy. We also consider the measurement problems it raises, and argue that quantifying the RR would probably require a two-stage procedure. In the first stage the size of the individual utility gain from a health intervention would be assessed using a technique such as the Standard Gamble or the Time Trade-Off, and in the second the social benefits arising from the RR would be quantified employing the Person Trade-Off. We also consider the normative status of the RR. We argue that it can be defended from a utilitarian point of view, on the ground that rescues increase well-being by reinforcing people's belief that they live in a community that places great value upon life. However, utilitarianism has long been criticised for failing to take sufficient account of fairness, and the case is no different here: fairness requires that we do not discriminate between individuals on morally irrelevant grounds, whereas being “identifiable” does not seem to be a morally relevant ground for discrimination.

Types of inconsistency in health-state utility judgments

In making judgments of health-related quality of life, respondents often compare the relative magnitude of two intervals between health states, such as the interval between normal health and blindness compared to that between normal health and death. We examined two ways of comparing such intervals: person-trade-off (PTO)—in which the judgment concerns matching numbers of people so that two changes are equivalent—and direct judgment of the ratio. Both measures showed ratio inconsistency (a ratio that should be the product of two other ratios is too high) and superadditivity (two ratios that should add to 1 are too high). Some responses in both methods implied that two intervals which should have been different (because they shared a top or bottom point, but differed on the other point) were nevertheless viewed by subjects as being of equal size. These equality responses were more common when death was the bottom (worse end) of both intervals being compared (e.g., the interval between death and blindness is perceived as being the same size as the interval between death and normal health) than when any other condition was at the bottom or when the condition common to the two intervals was at the top. A second experiment indicated that subjects really do consider the intervals to be equal. Our findings argue for giving subjects a chance to reflect on such apparent inconsistencies in practical utility elicitation.