Who should get the scarce ICU bed? The US public's view on triage in the time of COVID-19.

OBJECTIVE: To determine the relative importance members of the US public place on different patient attributes in triage decisions about who should receive the last available intensive care unit (ICU) bed. METHODS: A discrete choice experiment was conducted with a nationally representative sample of 2000 respondents from the YouGov internet panel of US households. Respondents chose which of three hypothetical patients with COVID-19 should receive an ICU bed if only one were available. The three patients differed in age, gender, Alzheimer's-like disability and probability of survival if the patient received the ICU bed. An experimental design varied the values of the four attributes of the three hypothetical patients with COVID-19 that a respondent saw in four choice tasks. RESULTS: The most important patient attribute to respondents was the probability the patient survives COVID-19 if they get the ICU bed (OR CI: 4.41 to 6.91). There was heterogeneity among different age groups of respondents about how much age of the patient mattered. Respondents under 30 years of age were more likely to choose young patients and old patients, and less likely to select patients aged 40-60 years old. For respondents in the age group 30-49 years old, as the age of the patient declined, their preference for saving the patient declined modestly in a linear fashion. CONCLUSIONS: Respondents favoured giving the last ICU bed available to the patient with the highest probability of surviving COVID-19. Public opinion suggests a simple guideline for physician choices based on likelihood of survival as opposed to the number of life-years saved. There was heterogeneity among respondents of different age groups for allocating the last ICU bed, as well as to the importance of the patient having an Alzheimer's-like disability (where religion of the respondent is important) and the gender of the patient (where the gender and racial identity are important).

Testing and Extending Schwartz Refined Value Theory Using a Best-Worst Scaling Approach.

The theory of human values discriminated 10 basic values arrayed in a quasicircular structure. Analyses with several instruments in numerous samples supported this structure. The refined theory of human values discriminates 19 values in the same circle. Its support depends on one instrument, the revised Portrait Values Questionnaire. We introduce a forced choice method, the Best-Worst Refined Values scale (BWVr), to assess the robustness of the refined theory to method of measurement and also assess the distinctiveness and validity of a new animal welfare value. Three studies ( N = 784, 439, and 383) support the theory and the new value. Study 3 also demonstrates the convergent and discriminant validity of the 19 values by comparing the BWVr, the revised Portrait Values Questionnaire, and value-expressive behaviors and confirms the test-retest reliability of BWVr responses. These studies provide further information about the order of values in the value circle.

Evaluation of strategies to communicate harmful and potentially harmful constituent (HPHC) information through cigarette package inserts: a discrete choice experiment.

BACKGROUND: The US Food and Drug Administration (FDA) has regulatory authority to use inserts to communicate with consumers about harmful and potentially harmful constituents (HPHCs) in tobacco products; however, little is known about the most effective manner for presenting HPHC information. METHODS: In a discrete choice experiment, participants evaluated eight choice sets, each of which showed two cigarette packages from four different brands and tar levels (high vs low), accompanied by an insert that included between-subject manipulations (ie, listing of HPHCs vs grouping by disease outcome and numeric values ascribed to HPHCs vs no numbers) and within-subject manipulations (ie, 1 of 4 warning topics; statement linking an HPHC with disease vs statement with no HPHC link). For each choice set, participants were asked: (1) which package is more harmful and (2) which motivates them to not smoke; each with a 'no difference' option. Alternative-specific logit models regressed choice on attribute levels. RESULTS: 1212 participants were recruited from an online consumer panel (725 18-29-year-old smokers and susceptible non-smokers and 487 30-64-year-old smokers). Participants were more likely to endorse high-tar products as more harmful than low-tar products, with a greater effect when numeric HPHC information was present. Compared with a simple warning statement, the statement linking HPHCs with disease encouraged quit motivation. CONCLUSIONS: Numeric HPHC information on inserts appears to produce misunderstandings that some cigarettes are less harmful than others. Furthermore, brief narratives that link HPHCs to smoking-related disease may promote cessation versus communications that do not explicitly link HPHCs to disease.

Application of discrete choice experiments to enhance stakeholder engagement as a strategy for advancing implementation: a systematic review.

BACKGROUND: One of the key strategies to successful implementation of effective health-related interventions is targeting improvements in stakeholder engagement. The discrete choice experiment (DCE) is a stated preference technique for eliciting individual preferences over hypothetical alternative scenarios that is increasingly being used in health-related applications. DCEs are a dynamic approach to systematically measure health preferences which can be applied in enhancing stakeholder engagement. However, a knowledge gap exists in characterizing the extent to which DCEs are used in implementation science. METHODS: We conducted a systematic literature search (up to December 2016) of the English literature to identify and describe the use of DCEs in engaging stakeholders as an implementation strategy. We searched the following electronic databases: MEDLINE, Econlit, PsychINFO, and the CINAHL using mesh terms. Studies were categorized according to application type, stakeholder(s), healthcare setting, and implementation outcome. RESULTS: Seventy-five publications were selected for analysis in this systematic review. Studies were categorized by application type: (1) characterizing demand for therapies and treatment technologies (n = 32), (2) comparing implementation strategies (n = 22), (3) incentivizing workforce participation (n = 11), and (4) prioritizing interventions (n = 10). Stakeholders included providers (n = 27), patients (n = 25), caregivers (n = 5), and administrators (n = 2). The remaining studies (n = 16) engaged multiple stakeholders (i.e., combination of patients, caregivers, providers, and/or administrators). The following implementation outcomes were discussed: acceptability (n = 75), appropriateness (n = 34), adoption (n = 19), feasibility (n = 16), and fidelity (n = 3). CONCLUSIONS: The number of DCE studies engaging stakeholders as an implementation strategy has been increasing over the past decade. As DCEs are more widely used as a healthcare assessment tool, there is a wide range of applications for them in stakeholder engagement. The DCE approach could serve as a tool for engaging stakeholders in implementation science.

Integrating cognitive process and descriptive models of attitudes and preferences.

Discrete choice experiments--selecting the best and/or worst from a set of options--are increasingly used to provide more efficient and valid measurement of attitudes or preferences than conventional methods such as Likert scales. Discrete choice data have traditionally been analyzed with random utility models that have good measurement properties but provide limited insight into cognitive processes. We extend a well-established cognitive model, which has successfully explained both choices and response times for simple decision tasks, to complex, multi-attribute discrete choice data. The fits, and parameters, of the extended model for two sets of choice data (involving patient preferences for dermatology appointments, and consumer attitudes toward mobile phones) agree with those of standard choice models. The extended model also accounts for choice and response time data in a perceptual judgment task designed in a manner analogous to best-worst discrete choice experiments. We conclude that several research fields might benefit from discrete choice experiments, and that the particular accumulator-based models of decision making used in response time research can also provide process-level instantiations for random utility models.

Tasmanian landowner preferences for conservation incentive programs: a latent class approach.

An empirical model of landowners' conservation incentive program choice is developed in which information about landowners' socio-economic and property characteristics and their attitudes, is combined with incentive program attributes. In a Choice survey landowners were presented with the choice of two incentive programs modelled as 'bundles of attributes' mimicking a voluntary choice scenario. Landowner behaviour and decision and the type of conditions and regulations they preferred were analyzed. Based on choice survey data, landowner heterogeneity was accounted for using a latent class approach to estimate the preference parameters. Three latent classes of landowners with different attitudes to the role and outcome of establishing conservation reserves on private land were identified: multi-objective owners; environment owners; and production owners. Only a small proportion of landowners, mostly environment owners, would voluntarily join a program. Although compensation funding contributed to voluntary program choice for multi-objective owners and environment owners, welfare losses were around 4000 AUD per hectare, which is less than the average agricultural land value in Tasmania. Landowners for whom compensation funding contributed to voluntary program choice were also most likely to set aside land for conservation without payment. This raises the possibility that the government's compensation expenditure could potentially be either reduced or re-allocated to landowners who will not voluntarily take conservation action. Increasing participation in conservation incentive programs and minimizing the welfare losses associated with meeting conservation targets may be best achieved by offering programs that allow flexibility in terms of legal arrangements and other program attributes.

Using discrete choice experiments to understand preferences for quality of life. Variance-scale heterogeneity matters.

Health services researchers are increasingly using discrete choice experiments (DCEs) to model a latent variable, be it health, health-related quality of life or utility. Unfortunately it is not widely recognised that failure to model variance heterogeneity correctly leads to bias in the point estimates. This paper compares variance heterogeneity latent class models with traditional multinomial logistic (MNL) regression models. Using the ICECAP-O quality of life instrument which was designed to provide a set of preference-based general quality of life tariffs for the UK population aged 65+, it demonstrates that there is both mean and variance heterogeneity in preferences for quality of life, which covariate-adjusted MNL is incapable of separating. Two policy-relevant mean groups were found: one group that particularly disliked impairments to independence was dominated by females living alone (typically widows). Males who live alone (often widowers) did not display a preference for independence, but instead showed a strong aversion to social isolation, as did older people (of either sex) who lived with a spouse. Approximately 6-10% of respondents can be classified into a third group that often misunderstood the task. Having a qualification of any type and higher quality of life was associated with smaller random component variances. This illustrates how better understanding of random utility theory enables richer inferences to be drawn from discrete choice experiments. The methods have relevance for all health studies using discrete choice tasks to make inferences about a latent scale, particular QALY valuation exercises that use DCEs, best-worst scaling and ranking tasks.

Using best-worst scaling choice experiments to measure public perceptions and preferences for healthcare reform in australia.

BACKGROUND: : One of the greatest difficulties in evaluating healthcare system reform in any country is that governments often do not clearly articulate what it is they are attempting to do. In Australia, a recent inquiry set out 15 principles to guide the reform process, but it remains unclear how the Australian public values the principles, how such values vary across the country, and, more fundamentally, if Australians understand the principles. OBJECTIVES: : To evaluate the Australian healthcare reform principles from the perspective of the Australian public, to test if such preferences are valued consistently across geographic and socioeconomic strata, and to test for the degree of understanding of the principles among the public. METHODS: : We employed best-worst scaling (BWS), a stated-preference method grounded in random utility theory, to elicit public preference for 15 healthcare reform principles. The BWS tasks were incorporated into an online survey that also gathered geographic and socioeconomic information and included questions relating to the understanding of the reform principles. Respondents were a geographically diverse set of Australians who were randomized to receive one of two versions of the survey, each containing a block of 15 choice tasks. Tasks in block one contained a subset of the choice tasks containing subsets of seven principles based on a balanced incomplete block design, while tasks in block two contained tasks with eight principles defined by the complement of the former.In each BWS task, respondents were simply asked to identify the most and least important principle. Analysis of preference was based on assigning the most valued principles a '1' and the least valued principles '-1', and with each item appearing eight times in each block, preferences were analyzed over a cardinal utility scale bounded by -8 and +8. Analysis was based on simple summary statistics and stratified by geographic and socioeconomic measures. RESULTS: : A sample of 204 respondents participated in the survey (a participation rate of 85%). Quality and safety was the most important principle and a culture of reflective improvement and innovation was the least important. Public voice and community engagement was the second least important principle and was also understood by barely half the respondents. CONCLUSIONS: : This research demonstrates how random-utility-based methods can be used to provide estimates of the importance of reform principles that have known statistical properties. The BWS task used forced respondents to discriminate between the principles on offer, unlike rating scales. Researchers and practitioners in healthcare should consider using BWS tasks in preference to rating scales.

Choice experiments in health: the good, the bad, the ugly and toward a brighter future.

Compared to many applied areas of economics, health economics has a strong tradition in eliciting and using stated preferences (SP) in policy analysis. Discrete choice experiments (DCEs) are one SP method increasingly used in this area. Literature on DCEs in health and more generally has grown rapidly since the mid-1990s. Applications of DCEs in health have come a long way, but to date few have been 'best practice', in part because 'best practice' has been somewhat of a moving target. The purpose of this paper is to briefly survey the history of DCEs and the state of current knowledge, identify and discuss knowledge gaps, and suggest potentially fruitful areas for future research to fill such gaps with the aim of moving the application of DCEs in health economics closer to best practice.

Estimating preferences for a dermatology consultation using Best-Worst Scaling: comparison of various methods of analysis.

BACKGROUND: Additional insights into patient preferences can be gained by supplementing discrete choice experiments with best-worst choice tasks. However, there are no empirical studies illustrating the relative advantages of the various methods of analysis within a random utility framework. METHODS: Multinomial and weighted least squares regression models were estimated for a discrete choice experiment. The discrete choice experiment incorporated a best-worst study and was conducted in a UK NHS dermatology context. Waiting time, expertise of doctor, convenience of attending and perceived thoroughness of care were varied across 16 hypothetical appointments. Sample level preferences were estimated for all models and differences between patient subgroups were investigated using covariate-adjusted multinomial logistic regression. RESULTS: A high level of agreement was observed between results from the paired model (which is theoretically consistent with the 'maxdiff' choice model) and the marginal model (which is only an approximation to it). Adjusting for covariates showed that patients who felt particularly affected by their skin condition during the previous week displayed extreme preference for short/no waiting time and were less concerned about other aspects of the appointment. Higher levels of educational attainment were associated with larger differences in utility between the levels of all attributes, although the attributes per se had the same impact upon choices as those with lower levels of attainment. The study also demonstrated the high levels of agreement between summary analyses using weighted least squares and estimates from multinomial models. CONCLUSION: Robust policy-relevant information on preferences can be obtained from discrete choice experiments incorporating best-worst questions with relatively small sample sizes. The separation of the effects due to attribute impact from the position of levels on the latent utility scale is not possible using traditional discrete choice experiments. This separation is important because health policies to change the levels of attributes in health care may be very different from those aiming to change the attribute impact per se. The good approximation of summary analyses to the multinomial model is a useful finding, because weighted least squares choice totals give better insights into the choice model and promote greater familiarity with the preference data.

Rescaling quality of life values from discrete choice experiments for use as QALYs: a cautionary tale.

BACKGROUND: Researchers are increasingly investigating the potential for ordinal tasks such as ranking and discrete choice experiments to estimate QALY health state values. However, the assumptions of random utility theory, which underpin the statistical models used to provide these estimates, have received insufficient attention. In particular, the assumptions made about the decisions between living states and the death state are not satisfied, at least for some people. Estimated values are likely to be incorrectly anchored with respect to death (zero) in such circumstances. METHODS: Data from the Investigating Choice Experiments for the preferences of older people CAPability instrument (ICECAP) valuation exercise were analysed. The values (previously anchored to the worst possible state) were rescaled using an ordinal model proposed previously to estimate QALY-like values. Bootstrapping was conducted to vary artificially the proportion of people who conformed to the conventional random utility model underpinning the analyses. RESULTS: Only 26% of respondents conformed unequivocally to the assumptions of conventional random utility theory. At least 14% of respondents unequivocally violated the assumptions. Varying the relative proportions of conforming respondents in sensitivity analyses led to large changes in the estimated QALY values, particularly for lower-valued states. As a result these values could be either positive (considered to be better than death) or negative (considered to be worse than death). CONCLUSION: Use of a statistical model such as conditional (multinomial) regression to anchor quality of life values from ordinal data to death is inappropriate in the presence of respondents who do not conform to the assumptions of conventional random utility theory. This is clearest when estimating values for that group of respondents observed in valuation samples who refuse to consider any living state to be worse than death: in such circumstances the model cannot be estimated. Only a valuation task requiring respondents to make choices in which both length and quality of life vary can produce estimates that properly reflect the preferences of all respondents.

Valuing the ICECAP capability index for older people.

This paper reports the first application of the capabilities approach to the development and valuation of an instrument for use in the economic evaluation of health and social care interventions. The ICECAP index of capability for older people focuses on quality of life rather than health or other influences on quality of life, and is intended to be used in decision making across health and social care in the UK. The measure draws on previous qualitative work in which five conceptual attributes were developed: attachment, security, role, enjoyment and control. This paper details the innovative use within health economics of further iterative qualitative work in the UK among 19 informants to refine lay terminology for each of the attributes and levels of attributes used in the eventual index. For the first time within quality of life measurement for economic evaluation, a best-worst scaling exercise has been used to estimate general population values (albeit for the population of those aged 65+ years) for the levels of attributes, with values anchored at one for full capability and zero for no capability. Death was assumed to be a state in which there is no capability. The values obtained indicate that attachment is the attribute with greatest impact but all attributes contribute to the total estimation of capability. Values that were estimated are feasible for use in practical applications of the index to measure the impact of health and social care interventions.

Using discrete choice experiments to investigate subject preferences for preventive asthma medication.

BACKGROUND AND OBJECTIVE: Long-term adherence to inhaled corticosteroids is poor despite the crucial role of preventer medications in achieving good asthma outcomes. This study was undertaken to explore patient preferences in relation to their current inhaled corticosteroid medication, a hypothetical preventer or no medication. METHODS: A discrete choice experiment was conducted in 57 adults with mild-moderate asthma and airway hyper-responsiveness, who were using inhaled corticosteroid

Best--worst scaling: What it can do for health care research and how to do it.

Statements like "quality of care is more highly valued than waiting time" can neither be supported nor refuted by comparisons of utility parameters from a traditional discrete choice experiment (DCE). Best--worst scaling can overcome this problem because it asks respondents to perform a different choice task. However, whilst the nature of the best--worst task is generally understood, there are a number of issues relating to the design and analysis of a best--worst choice experiment that require further exposition. This paper illustrates how to aggregate and analyse such data and using a quality of life pilot study demonstrates how richer insights can be drawn by the use of best--worst tasks.

What influences participation in genetic carrier testing? Results from a discrete choice experiment.

This study explores factors that influence participation in genetic testing programs and the acceptance of multiple tests. Tay Sachs and cystic fibrosis are both genetically determined recessive disorders with differing severity, treatment availability, and prevalence in different population groups. We used a discrete choice experiment with a general community and an Ashkenazi Jewish sample; data were analysed using multinomial logit with random coefficients. Although Jewish respondents were more likely to be tested, both groups seem to be making very similar tradeoffs across attributes when they make genetic testing choices.