Advertisement

Quantifying the Burden of Hyperphagia in Prader-Willi Syndrome Using Quality-Adjusted Life-years

      Abstract

      Background

      Prader-Willi syndrome (PWS) is a rare disease associated with cognitive impairment, hypotonia, hyperphagia (an insatiable hunger), and obesity. Therapies that target hyperphagia are in development, but understanding the value of these therapies to inform patient-focused drug development (PFDD) requires valid data on disease burden. We estimated disease burden by measuring and comparing quality-adjusted life-years (QALYs) for 3 PWS health states relevant to current PFDD initiatives.

      Methods

      Time trade-off (TTO) and a visual analog scale (VAS) were used to elicit PWS caregivers’ values for 3 fixed health states for a standardized patient described with (1) untreated PWS, (2) PWS with controlled obesity, and (3) PWS with controlled obesity and hyperphagia. We excluded participants who left at least 1 TTO or VAS question blank or incomplete (noncompleters) and respondents who reported the same answer for all TTO scenarios (nontraders). The remaining group of respondents (traders) were used for all primary analyses. We assessed validity and bias of QALY estimates by comparing differences in health state valuations, treatment priorities, and characteristics among respondents who did and did not complete the TTO.

      Results

      A total of 458 respondents completed the survey, including 226 traders, 93 nontraders, and 139 noncompleters. Traders valued untreated PWS at 0.69 QALYs, PWS with controlled obesity at 0.79 QALYs, and controlled hyperphagia/obesity at 0.91 QALY (P < 0.01 for differences among health state values). Reported VAS ratings were similar for traders versus nontraders for untreated PWS (38.64 vs 38.95, P = 0.89) and PWS with controlled obesity (57.36 vs 55.14, P = 0.35) but varied for PWS with controlled obesity and hyperphagia (70.70 vs 64.46, P = 0.02). Exclusion of noncompleters did not introduce obvious bias because traders and noncompleters were similar in treatment priorities and characteristics. The exclusion of nontraders did not meaningfully alter mean or distribution of valuations.

      Conclusions

      This study found that avoiding hyperphagia decreases the burden of PWS and that these results are robust, even once imposing strict inclusion criteria. Use of fixed health states to estimate QALYs addresses many of the complexities of measuring disease burden in rare and pediatric conditions, indicating the potential value of this approach to inform premarket decision makers in identifying outcome importance.

      Key words

      Introduction

      Data on disease burden are routinely used in medical decision making and resource allocation.
      • Noyes J
      • Edwards RT.
      EQ-5D for the assessment of health-related quality of life and resource allocation in children: a systematic methodological review.
      ,
      • Longworth L
      • Yang Y
      • Young T
      • et al.
      Use of generic and condition-specific measures of health-related quality of life in NICE decision-making: a systematic review, statistical modelling and survey.
      Quality-adjusted life-years (QALYs) are most often used as an outcome measure in cost-effectiveness analyses to reflect the impact of a treatment on length of life and disease burden. The disease burden component of a QALY is quantitatively assessed by eliciting individuals’ preferences for a health state or treatment outcome. This preference-based assessment of utility reflects the subjective value or desirability associated with that health state.
      QALYs reflect individual preferences for treatment outcomes and have recently been highlighted as having potential value in informing US Food and Drug Administration (FDA) regulatory decisions as part of patient-focused drug development (PFDD) initiatives.

      Medical Device Innovation Consortium (MDIC). Patient centered benefit-risk project report: a framework for incorporating information on patient preferences regarding benefit and risk into regulatory assessments of new medical technology. Minneapolis, MN. 2015. https://mdic.org/wp-content/uploads/2015/05/MDIC_PCBR_Framework_Web.pdf. Accessed date: 14 May 2021.

      ,
      • de Bekker-Grob EW
      • Berlin C
      • Levitan B
      • et al.
      Giving patients’ preferences a voice in medical treatment life cycle: the PREFER public–private project.
      PFDD aims to incorporate the perspectives, experiences, and preferences of patients and caregivers into the drug development process.
      Patient-Focused Drug Development Guidance: Collecting Comprehensive and Representative Input
      US Food and Drug Administration.
      The increasing popularity of patient preference methods in the United States are accompanied by an increase in their use internationally and across diverse health policy, public health, and medical settings.
      • Brown L
      • Lee T-h
      • De Allegri M
      • Rao K
      • Bridges JF.
      Applying stated-preference methods to improve health systems in sub-Saharan Africa: a systematic review.
      • Beckham SW
      • Crossnohere NL
      • Gross M
      • Bridges JF.
      Eliciting Preferences for HIV Prevention Technologies: A Systematic Review.
      • Ozawa S
      • Wonodi C
      • Babalola O
      • Ismail T
      • Bridges J.
      Using best-worst scaling to rank factors affecting vaccination demand in northern Nigeria.
      • Ozawa S
      • Grewal S
      • Bridges JF.
      Household size and the decision to purchase health insurance in Cambodia: results of a discrete-choice experiment with scale adjustment.
      • Hashim H
      • Beusterien K
      • Bridges JF
      • Amos K
      • Cardozo L.
      Patient preferences for treating refractory overactive bladder in the UK.
      • Bridges JF
      • Selck FW
      • Gray GE
      • McIntyre JA
      • Martinson NA.
      Condom avoidance and determinants of demand for male circumcision in Johannesburg, South Africa.
      The application of QALYs in PFDD presents several challenges, including uncertainty in their appropriateness for pediatric conditions evaluated by parental proxies,
      • Khadka J
      • Kwon J
      • Petrou S
      • Lancsar E
      • Ratcliffe J.
      Mind the (inter-rater) gap. An investigation of self-reported versus proxy-reported assessments in the derivation of childhood utility values for economic evaluation: A systematic review.
      unknown significance of QALY estimates to premarket decision makers,
      • Brower A.
      Is It Time To Take a Harder Look at the QALY?.
      the ability of QALYs to document small but potentially meaningful differences in treatment outcomes in rare diseases,
      • Crossnohere NL
      • Fischer R
      • Lloyd A
      • Prosser LA
      • Bridges JFP.
      Assessing the Appropriateness of the EQ-5D for Duchenne Muscular Dystrophy: A Patient-Centered Study.
      ,
      • Efthymiadou O
      • Mossman J
      • Kanavos P.
      Health related quality of life aspects not captured by EQ-5D-5L: Results from an international survey of patients.
      and identification of an elicitation method that is appropriate for the health concept of interest.
      Patient-Focused Drug Development Draft Guidance: Methods to Identify What is Important to Patients
      US Food and Drug Administration.
      QALYs can be estimated using several direct elicitation methods, including time trade-off (TTO), standard gamble, and visual analog scale (VAS).
      • Whitehead SJ
      • Ali S.
      Health outcomes in economic evaluation: the QALY and utilities.
      Indirect methods of eliciting QALYs include the EuroQol–5 Dimension (EQ-5D) on which respondents rate 5 domains of health on 3 or 5 levels, and the Short Form–Six Dimensions. More general trade-off techniques can be applied to estimate preferences in PFDD, including contingent valuation to explore monetary valuation and threshold techniques (generalized standard gamble) to explore other trade-offs, such as maximum acceptable risk.
      • Hauber AB
      • Johnson FR
      • Muhlbacher A
      • Marshall D.
      Maximum acceptale risk: Alternative measures for quantifying patients’ tolerance for theraputic benefit. International Society for Pharmacoeconomics and Outcomes Research.
      Our current study was motivated by two factors. First, within the context of Prader-Willi syndrome (PWS), a rare genetic condition that results in cognitive impairment, hypotonia, hyperphagia, and obesity, there has been an active PFDD-type effort to highlight the importance and impact of hyperphagia as a clinical endpoint.
      • Tsai JH
      • Scheimann AO
      • McCandless SE
      • Strong TV
      • Bridges JFP.
      Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best-worst scaling.
      Although hyperphagia is an increasingly common target in drug development
      • Tan Q
      • Orsso CE
      • Deehan EC
      • et al.
      Current and emerging therapies for managing hyperphagia and obesity in Prader-Willi syndrome: A narrative review.
      and can be measured by a validated instrument,
      • Fehnel SE
      • Brown TM
      • Nelson L
      • et al.
      Development of the Hyperphagia Questionnaire for Use in Prader-Willi Syndrome Clinical Trials.
      ,
      • Dykens EM
      • Maxwell MA
      • Pantino E
      • Kossler R
      • Roof E.
      Assessment of hyperphagia in Prader-Willi syndrome.
      evidence is lacking on the impact of hyperphagia on overall disease burden. Second, although the standard gamble and TTO had been highlighted as preference elicitation methods with potential regulatory science applications,

      Medical Device Innovation Consortium (MDIC). Patient centered benefit-risk project report: a framework for incorporating information on patient preferences regarding benefit and risk into regulatory assessments of new medical technology. Minneapolis, MN. 2015. https://mdic.org/wp-content/uploads/2015/05/MDIC_PCBR_Framework_Web.pdf. Accessed date: 14 May 2021.

      these approaches have been seldomly used within the broader PFDD effort and literature. Whether the TTO approach would be a valid and acceptable method for eliciting QALYs in PWS was of particular interest in this study because the FDA expressed interest in data indicating whether caregivers would accept early death in exchange for hyperphagia treatment benefits.

      Kayadjanian N. PWS Clinical Trials Consortium Meets with FDA for a Critical Path Innovation Meeting: Foundation for Prader-Willi Research; 2019. https://www.fpwr.org/blog/pws-clinical-trials-consortium-meets-with-fda-for-a-critical-path-innovation-meeting Accessed 14 May 2021.

      The aim of this study was to document whether eliminating hyperphagia in people with PWS would be considered (ie, subjectively valued) as affecting disease burden. However, several uncertainties exist in applying a preference elicitation approach to assess PWS-related quality of life in this regulatory context. First, whether use of the TTO approach is sensitive enough to detect a statistical and clinical difference in utilities across PWS health states is not known. Hyperphagia and other symptoms of PWS are variable across the natural history of PWS. Therefore, cross-sectional approaches using caregiver proxy report of a care recipient's current health state would be confounded by a range of other child and family factors, including clinical, demographic, and socioeconomic characteristics. Second, in pretesting caregivers objected to the use of direct elicitation methods that applied to their own child.
      To overcome some of these barriers, we use a third-person fixed-vignette approach to describe the health state. With this approach, respondents are presented with profiles of a hypothetical person with PWS and asked to evaluate their health in different standardized scenarios with and without the symptoms of obesity and hyperphagia, using the TTO method. This method avoids the inherent heterogeneity of valuing patients with a range of clinical characteristics, demographic characteristics, and treatment histories, hence reducing measurement error. It also avoids biases associated with having the caregivers value their own child's health.
      • Albrecht GL
      • Devlieger PJ.
      The disability paradox: high quality of life against all odds.
      In addition, evaluation of third-person vignettes has proved beneficial in other preference studies across a range of health conditions, including rare diseases,
      • Simon NJ
      • Richardson J
      • Ahmad A
      • et al.
      Health utilities and parental quality of life effects for three rare conditions tested in newborns.
      • Pyne JM
      • Fortney JC
      • Tripathi S
      • Feeny D
      • Ubel P
      • Brazier J.
      How bad is depression? Preference score estimates from depressed patients and the general population.
      • Bridges JF
      • Kinter ET
      • Schmeding A
      • Rudolph I
      • Mühlbacher A.
      Can patients diagnosed with schizophrenia complete choice-based conjoint analysis tasks?.
      and the method is more closely aligned with the underlying approach to health state valuation.
      The EuroQol Group
      EuroQol–a new facility for the measurement of health-related quality of life.
      Through the use of intensive stakeholder engagement with the PWS community, we anticipated that a TTO preference elicitation approach would facilitate tradeoffs focused on the symptoms of obesity and hyperphagia.
      We assessed the generalizability of results by exploring the impact of 2 sources of censoring in the TTO: a low response rate and nontrading participants. This research indicates that QALYs could have relevance to both premarket decision makers in identifying how much outcomes matter to disease burden and postmarket decision makers in specifying the value of improving those outcomes. These postmarket applications could include informing future cost-effectiveness analysis, informing reimbursement decisions, and determining how the benefits of future therapies are communicated and marketed to physicians and caregivers. Although our study is predominantly focused on regulatory science applications in the United States,

      Medical Device Innovation Consortium (MDIC). Patient centered benefit-risk project report: a framework for incorporating information on patient preferences regarding benefit and risk into regulatory assessments of new medical technology. Minneapolis, MN. 2015. https://mdic.org/wp-content/uploads/2015/05/MDIC_PCBR_Framework_Web.pdf. Accessed date: 14 May 2021.

      ,
      Patient-Focused Drug Development Guidance: Collecting Comprehensive and Representative Input
      US Food and Drug Administration.
      ,
      Patient-Focused Drug Development Draft Guidance: Methods to Identify What is Important to Patients
      US Food and Drug Administration.
      ,
      • Perfetto EM
      • Burke L
      • Oehrlein EM
      • Epstein RS.
      Patient-focused drug development: a new direction for collaboration.
      the evidence on the burden of this rare disease could have implications globally.
      • Hsu JC
      • Wu H-C
      • Feng W-C
      • Chou C-H
      • Lai EC-C
      • Lu CY.
      Disease and economic burden for rare diseases in Taiwan: A longitudinal study using Taiwan's National Health Insurance Research Database.

      Participants and Methods

      PWS is a rare neurodevelopmental genetic disorder with no known cure and an unmet need for effective therapies.
      • Tsai JH
      • Scheimann AO
      • McCandless SE
      • Strong TV
      • Bridges JFP.
      Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best-worst scaling.
      We developed an online patient preference survey in partnership with the Foundation for Prader Willi Research and the PWS Clinical Trial Consortium, a not-for-profit international partnership of industry, academic, patient advocacy, and governmental agencies. Such organizations are increasingly playing a role in facilitating drug development for rare diseases.
      • Wästfelt M
      • Fadeel B
      • Henter JI.
      A journey of hope: lessons learned from studies on rare diseases and orphan drugs.
      We used a community-engaged approach in accordance with the research team's previous experience
      • Hollin IL
      • Young C
      • Hanson C
      • Bridges JF
      • Peay H.
      Developing a patient-centered benefit-risk survey: a community-engaged process.
      and current regulatory recommendations on patient group engagement.
      Patient-Focused Drug Development Draft Guidance: Methods to Identify What is Important to Patients
      US Food and Drug Administration.
      A community advisory board composed of patients with PWS, parents, clinicians, and researchers guided the survey development process, as previously described.
      • Tsai JH
      • Scheimann AO
      • McCandless SE
      • Strong TV
      • Bridges JFP.
      Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best-worst scaling.
      Caregivers were recruited via e-mail to participate in the online survey through the PWS Clinical Trial Consortium, Prader-Willi Syndrome Association, and Foundation for Prader-Willi Research. These groups are predominantly based in the United States but have some global reach. Adults with PWS were not eligible to participate because of the cognitive challenges with completing the survey. Caregivers were eligible to participate if (1) they were a parent, grandparent, sibling, or legal guardian of a person with PWS; (2) they were involved in medical decision making for the person with PWS and (3) the person with PWS was 4 years or older. Caregivers of children 3 years and younger were ineligible because hyperphagia onset is typically after 4 years of age.
      • Miller JL
      • Lynn CH
      • Driscoll DC
      • et al.
      Nutritional phases in Prader-Willi syndrome.
      The survey included 5 sections: background information, treatment preference elicitation using a discrete-choice experiment, treatment priority elicitation using a best-worst scaling (BWS) method, disease burden assessment using direct utility elicitation methods, and overall survey feedback. In the disease burden assessment, we presented respondents with 3 hypothetical fixed vignettes to assess burden, with each describing an 18-year-old person with PWS. Each vignette varied regarding the presence or absence of 2 typical PWS symptoms: hyperphagia and obesity. The first vignette presented a patient with PWS and hyperphagia and obesity, the second presented a patient with hyperphagia but not obesity, and the third presented a patient with neither hyperphagia nor obesity. These 3 vignettes were developed together with the community advisory board. Given the FDA's interest in the topic of exploring tradeoffs between a child's early death and reductions in hyperphagia, the advisory board identified the TTO method as the best approach to capture these tradeoffs. The advisory board developed 3 scenarios describing the experiences of diverse patients with PWS for use in the TTO. Finally, the vignettes were modified for clarity after a pretesting of the instrument at a PWS community-based event.
      • Tsai JH
      • Janssen E
      • Bridges JF.
      Research as an event: a novel approach to promote patient-focused drug development.
      To highlight the disease burden associated with hyperphagia in PWS, we valued 3 heath states. The first related to the natural history of PWS and was the historic presentation of the condition of both hyperphagia and obesity. The second reflects the contemporary presentation of PWS for which early diagnosis, food-restricted environments, and the use of human growth hormone has helped reduce obesity but hyperphagia remains. Third, we studied a hypothetical health state related to the targeted product profile of a number of investigations therapies where neither obesity nor hyperphagia was present. To prevent additional burden, we did not consider the health state where obesity and hyperphagia were present but are exploring obesity-hyperphagia tradeoffs in a separate study using a discrete-choice experiment.
      Survey participants rated the health state in each vignette using a VAS ranging from worst (score of 0) to best (score of 100) imaginable health state (Figure 1). Survey participants then completed a TTO exercise in which they indicated whether it would be better for the person described in the vignette to live in their current condition as described for their remaining life expectancy of 20 years or to live fewer years completely healthy and without symptoms described in the scenario. If they chose the latter, participants were asked to indicate how many years of life the person in the scenario should give up in exchange for living the remaining years in complete health. If the number of years traded by survey participants exceeded 20, we set the value back to 20. We expected that participants would trade the most years to restore the health of the patient with both hyperphagia and obesity and the least years to restore health for the patient without hyperphagia and obesity.
      Figure 1
      Figure 1Examples of visual analog scale and time trade-off task.
      In the final sample, we excluded respondents who did not fully complete the BWS portion of the survey and did not begin answering the TTO questions. To maintain the internal consistency of our preference data, we also excluded participants who left at least 1 TTO or VAS question blank or indicated that they would trade years of life in the TTO question but did not indicate how many years (noncompleters). We also excluded respondents who reported the same answer for all TTO scenarios (nontraders) after consultation with the study's community advisory board. The advisors thought that the disease burden was different across the 3 health state vignettes and therefore believed that the responses were not a valid reflection of the impact of hyperphagia and obesity. The remaining group of respondents (traders) were used for all primary analyses, and nontraders were included in the sample in sensitivity analysis.

      Statistical Analysis

      We calculated QALYs by dividing the number of years traded in the TTO exercise by 20 years of life expectancy and subtracted this estimate from 1. Although a QALY reflects both the time spent in a health state and the preference-based quality of life of that health state, we use the term QALY in this article to reflect the latter concept only because this term is consistent with how patient groups have used the term QALY in PFDD. We summarized unadjusted TTO, QALY, and VAS estimates for each health state. We compared VAS estimates between traders and nontraders.
      We used generalized estimating equations to evaluate associations between years traded in the TTO questions (dependent variable) and features of the respondent and scenario (independent variables). To determine the appropriate model for the TTO count data, we first assessed the dispersion of the data and used a negative binomial regression model to account for the observed overdispersion.
      • Rodrıguez G.
      Models for count data with overdispersion.
      ,
      • Hanley JA
      • Negassa A
      • Edwardes MDd
      • Forrester JE.
      Statistical analysis of correlated data using generalized estimating equations: an orientation.
      A systematic model-building process examined associations between the years traded in the TTO question and (1) caregiver demographic characteristics, (2) clinical characteristics of patients with PWS, and (3) 5 self-reported personality features reported by caregivers. This process led to the development of a final model examining years traded in the TTO exercise as a function of significant independent variables.
      Survey respondents evaluated the TTO exercise using debriefing questions assessing whether the instrument was easy to understand, easy to answer, and consistent with their preferences.
      • Bridges JF
      • Tsai J-H
      • Janssen E
      • Crossnohere NL
      • Fischer R
      • Peay H.
      How do members of the Duchenne and Becker muscular dystrophy community perceive a discrete-choice experiment incorporating uncertain treatment benefit? An application of research as an event.
      Participants rated their agreement using a 5-point scale ranging from strongly agree to strongly disagree, and we recoded answers as strongly agree/agree versus do not agree. Differences in percentage of agreement were assessed across trader and nontrader groups. The same evaluative questions were included for the BWS, and we compared the endorsement rates of the BWS and the TTO.

      Data Visualization

      We sought to visualize health status associated with different health states in PWS and the heterogeneity in the evaluation of these health status. Heterogeneity assessment is a recommended quality of good preference research as per the FDA.
      Patient Preference Information–Voluntary Submission, Review in Premarket Approval Applications, Humanitarian Device Exemption Applications, and De Novo Requests, and Inclusion in Decision Summaries and Device Labeling
      US Food and Drug Administration.
      We created a graph with QALYs (range, 0–1) and percentage of respondents endorsing the health state as at least that QALY or greater. This graph was inspired by Fries’ increasingly rectangular survival curve, which describes how changes in population aging over time resulted in an increasingly more rectangular survival curve.
      • Fries JF.
      Aging, natural death, and the compression of morbidity.
      Complete rectangularization in this graph is achieved when 100% of the population survives until the natural limit of the life span. In our graph, maximum rectangularization is achieved when 100% of respondents value a health state as full health (ie, 1).
      Figure 2 demonstratets this graph in the context of our study. Panel A displays the maximum rectangularization scenario wherein all respondents value the health state as equivalent to full health. Panel A achieves maximum rectangularization. Panel B describes an impaired health state that is consistently valued by all respondents as 0.6 QALYs. Although the figure is rectangular because there is no heterogeneity in responses, it does not reflect the maximum rectangularization because the health state is impaired relative to full health. Panel C describes an impaired health state for which there is an inconsistent valuation of that health state. The median valuation is 0.6 QALYs, with 50% of respondents indicating the health state is >0.6 QALYs and 50% of respondents indicating the health state is <0.6 QALYs.
      Figure 2
      Figure 2Three different distributions of health state valuations that could be observed in a sample. (A) A total of 100% of the respondents endorse the given health state as 1 quality-adjusted life-year (QALY) (full health); (B) 100% of the respondents endorse the given health state as 0.6 QALY; (C) 50% of the respondents endorse the given health state as <0.6 QALY and 50% of the respondents endorse the given health state as >0.6 QALY.

      Sensitivity Analyses

      We compared BWS results across traders, nontraders, and noncompleters to understand whether their treatment priorities varied. The BWS protocol was previously published.
      • Tsai JH
      • Scheimann AO
      • McCandless SE
      • Strong TV
      • Bridges JFP.
      Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best-worst scaling.
      Survey respondents evaluated treatment priorities including hyperphagia, temper outbursts, intellectual functioning, overweight, anxiety, skin picking, and gastrointestinal problems. Preference models were estimated using conditional logistic regression and probability rescaled.

      The MaxDiff System Technical Paper. Sawtooth Software Technical Paper Series. 2020. https://sawtoothsoftware.com/uploads/sawtoothsoftware/originals/0114c81d-43ca-4339-bd30-fbf45146e8b2.pdf Accessed 14 May 2021.

      Analysis of variance was used to assessed differences in the relative importance of each outcome across groups. All analysis in the current study were performed in Stata software, version 16 (StataCorp, College Station, Texas). This study was approved as exempt human subject research by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board.

      Results

      Among the 458 participants who completed the BWS exercise in the first part of the survey, 67 left at least 1 TTO or VAS question blank and 72 did not indicate how many years they would trade (139 noncompleters) (Figure 3). Ninety-four respondents reported the same answer for all 3 TTO scenarios (94 nontraders), and 226 completed all TTO and VAS questions (226 traders). The VAS scores for nontraders were significantly different across health states (ranging from 38.35 for hyperphagia and obesity to 64.46 for no hyperphagia and no obesity; P < 0.01) and generally consistent with the VAS scores for traders, prompting us to consider the repetitive TTO answers from nontraders invalid (Supplemental Table I). All results are from 226 traders unless otherwise stated.
      Figure 3
      Figure 3Flowchart of participation in the study. BWS = best-worst scaling; PWS = Prader-Willi syndrome.; QALY = quality-adjusted life-year; TTO = time trade-off; VAS = visual analog scale.
      Table I describes characteristics of the 3 groups. The mean age of the 226 traders was 50 years (range, 22–83 years). Most were female (80%), the parent of a child with PWS (96%), and white (90%). Sixty-four percent had a household income >$100,000 per year. Most caregivers reported being optimistic about the future (76%), actively working to improve their health (84%), being good with numbers (72%), and having a lot of self-control (75%). Caregivers reported a mean age of 16 years (range, 4–54 years) for their family members with PWS. More than half of the PWS individuals had symptoms of hyperphagia (64%), emotional tantrums (69%), and cognitive difficulties (66%). Fifty percent of individuals with PWS had overweight or obesity. The groups did not differ except with regard to self-rated optimism, with traders being more likely to endorse themselves as optimistic compared with noncompleters or nontraders (76% of traders vs 68% of noncompleters and nontraders, P = 0.043). Among the 458 individuals who engaged in the survey overall, most came from the United States (89%) and Canada (5%), with 3% of the sample coming from Europe and 3% from other countries.
      Table ICaregiver and patient characteristics.
      CharacteristicFull Sample (N = 458)Traders
      Traders (respondents) are those who completed all 3 time trade-off scenarios and gave varied answers across the different scenarios.


      (n = 226)
      Nontraders
      Nontraders are those who completed the activity but traded the same number of years in each of the 3 scenarios.
      (n = 93)
      Noncompleters
      Noncompleters are those who were missing a response to at least 1 time trade-off scenario but traded the same number of years.
      (n = 139)
      P
      Comparing traders, nontraders, and noncompleters.
      Caregiver characteristics
       Age, mean (range), y49.2 (22–83)49.7 (22–83)47.7 (28–68)50.0 (25–75)0.19
       Female, %83.479.985.089.80.10
       Parent of child with PWS, %97.496.097.8100.00.12
       Race, %0.25
        White86.090.382.081.7
        Black/African American2.00.92.93.2
         Latinx4.13.55.04.3
        Other7.95.310.110.8
       Private insurance, %69.069.570.565.60.71
       Educational level of BA degree or higher, %71.372.668.872.00.74
       Household income, %0.12
        <$50,00014.310.316.220.9
        $50-$100,00026.625.627.427.9
        >$100,00059.164.056.451.2
       Personality assessment, %
        Active health improvement83.583.580.488.20.30
        Optimism72.176.463.874.20.03
        Self-control71.074.767.966.70.24
        Good with numbers69.272.069.362.40.23
        Risk taker28.427.126.534.40.39
      Patient characteristics
       Age, mean (range), y15.6 (4–54)16.1 (4–54)14.2 (4–40)16.6 (4–42)0.13
       Symptoms, %
        Cognitive difficulties96.796.998.693.50.11
        Difficult behavior84.788.992.889.20.46
        Hyperphagia67.767.764.073.10.38
        Overweight/obesity50.749.648.955.90.52
      BA = bachelor of arts; PWS = Prader-Willi syndrome.
      low asterisk Traders (respondents) are those who completed all 3 time trade-off scenarios and gave varied answers across the different scenarios.
      Nontraders are those who completed the activity but traded the same number of years in each of the 3 scenarios.
      Noncompleters are those who were missing a response to at least 1 time trade-off scenario but traded the same number of years.
      § Comparing traders, nontraders, and noncompleters.

      Disease Burden Assessment

      Traders reported that a hypothetical individual with PWS and both hyperphagia and obesity symptoms should give up a mean of 6.12 of 20 years to live their remaining years in complete health (0.69 QALYs) (Supplemental Table I). Participants reported that individuals with PWS and hyperphagia but no obesity should give up 4.26 years to live their remaining years in full health (0.79 QALYs). An individual with PWS but no hyperphagia and no obesity should give up 1.82 years to live in full health (0.91 QALYs). The PWS VAS scores were lower than QALY weights derived from TTO amounts for the same health state: 38.64 of 100 for hyperphagia and obesity, 57.36 for hyperphagia but no obesity, and 70.70 for no hyperphagia and no obesity.
      Table II reports years traded and QALYs associated with PWS health states adjusted for significant respondent and child characteristics including caregiver sex, race, age >50 or <50 years, health seeking, and PWS symptoms of obesity/overweight and hyperphagia (see Supplemental Table II for model-building process). Compared with a scenario where apatient is living with PWS, hyperphagia, and obesity, a scenario where a patients had hyperphagia but no obesity was significantly associated with 1.95 fewer years traded to live completely healthy (0.10 additional QALYs), and a scenario where a patient had no hyperphagia and no obesity was significantly associated with 4.48 fewer years traded (0.22 additional QALYs; P < 0.01 for both comparisons). The caregiver reported that obesity/overweight and hyperphagia symptoms of the person with PWS did not impact the TTO responses. Caregivers who were Hispanic were willing to trade less time (P < 0.05), and those with health seeking behavior were willing to trade more time (P < 0.05) to improve health.
      Table IIEffect of scenario, caregiver, and child characteristics on years traded and associated QALYs.
      VariableYears Traded,
      Years traded derived from negative binomial coefficients converted using recycled prediction methods
      Mean (SE)

      (n = 226)
      QALYs
      QALY estimates derived by dividing all years traded estimates by 20 (maximum number of years); fewer years traded equate greater QALY estimates; more years traded equate lower QALYs.


      (n = 226)
      Scenario
      All scenario coefficients are significantly different from one another.
       Hyperphagia and obesityReferenceReference
       Hyperphagia, no obesity−1.95
      P < 0.01.
      (0.03)
      0.10
       No hyperphagia, no obesity−4.48
      P < 0.01.
      (0.08)
      0.22
      Caregiver characteristics
       Female sex0.11 (0.01)−0.01
       Race
        WhiteReference
        Black/African American0.09 (0.01)−0.005
        Latinx−1.77
      P < 0.05.
      (0.06)
      0.09
        Other−0.96 (0.06)0.05
      Age ≥50 years−0.48 (0.02)0.02
      Health seeking1.30
      P < 0.05.
      (0.04)
      −0.07
      Patient characteristics
       Hyperphagia1.01 (0.03)−0.05
       Obesity/overweight0.62 (0.04)−0.03
      QALY = quality-adjusted life-year.
      low asterisk Years traded derived from negative binomial coefficients converted using recycled prediction methods
      QALY estimates derived by dividing all years traded estimates by 20 (maximum number of years); fewer years traded equate greater QALY estimates; more years traded equate lower QALYs.
      All scenario coefficients are significantly different from one another.
      § P < 0.01.
      || P < 0.05.

      Sample Distribution of QALYs Across Scenarios

      Figure 4 shows the distribution in health state valuations observed in our sample for each health state and illustrates the percentage of people who value a health state as greater than or equal to a given QALY. The less severe the health state described, the more skewed the distribution of the sample responses was toward a QALY of 1 (Figure 4 and Supplemental Figure 1). For the health state describing PWS without hyperphagia or obesity (mean [SD] QALYs, 0.91 [0.3]), 80% of respondents valued the health state QALY at ≥0.90, and 68% of respondents valued the health state at 1 QALY. For individuals with PWS with hyperphagia but no obesity (mean [SD] QALYs, 0.79 [0.2]), 40% of respondents valued the health state QALY at ≥0.90, and 22% valued it at 1. For individuals with PWS with hyperphagia and obesity (mean [SD] QALYs, 0.69 [0.3]), 10% of respondents valued the health state QALY at ≥0.90, and 0% valued it at 1. For all 3 health states, few respondents valued the health state with a QALY <0.5.
      Figure 4
      Figure 4Sample quality-adjusted life-year (QALY) distributions for each health state among traders only and traders and nontraders.
      QALY distributions for all 3 health states are variations of the example educational illustration shown in Figure 2, which depicts variation across the sample of health state values reported. Among the 3 health states, the least severe health state with no obesity or hyperphagia has a distribution of values most strongly skewed toward a QALY of 1 and most closely resembles the distribution shown in Figure 2A wherein 100% of respondents in a sample agree that a health state value is equal to full health.

      Sensitivity Analyses

      Nontraders traded the same number of years for all 3 scenarios (mean QALY, 0.90) (Supplemental Table III). This was significantly higher than the QALY scores of traders for the hyperphagia/obesity scenario (P < 0.001) and hyperphagia/no obesity scenario (P < 0.001). No difference was found between the QALY scores of traders and nontraders in the no hyperphagia/no obesity scenario (P = 0.90).
      Including both nontraders and traders in our sample (total n = 319), we found that the QALY distributions of the scenarios were more closely aligned (Figure 4). In adjusted analyses, inclusion of nontraders reduced the magnitude of the differences in years traded across the scenarios, but differences were still statistically significant (Supplemental Table IV) (P < 0.01). Similarly, TTO amounts remained significantly lower for Hispanic individuals and higher for health-seeking individuals, but the magnitude of these differences was reduced (Supplemental Table IV).
      In comparing groups based on their treatment priorities elicited in BWS, we found that hyperphagia was the most important treatment target for all 3 groups (relative importance: 19.21, 95% CI, 16.6–21.8 for traders; 15.91; 95% CI, 12.2-19.6 for nontraders; and 18.16, 95% CI, 15.0–21.4 for noncompleters) (Supplemental Table V). No significant differences were found in the relative importance of any of the attributes across the 3 groups (all between-group P > 0.05).

      Acceptability of TTO Exercise

      Nontraders were significantly less likely than traders to report that the TTO questions were easy to understand (73% vs 81%; P = 0.01) (Supplemental Table VI) and consistent with their preferences (57% vs 74%; P < 0.01). Noncompleters who left QALY questions blank reported fairly high satisfaction with the earlier BWS questions and similar BWS satisfaction compared with those who completed the QALY section, indicating that their noncompletion of the QALY question was likely attributable to their dissatisfaction with the QALY section rather than the BWS.

      Discussion

      This is the first study to find that improvements in hyperphagia and obesity for PWS can be measured using TTO methods and that the value of these changes in health symptoms can be measured with the use of QALYs. As a previous study
      • Matza LS
      • Boye KS
      • Feeny DH
      • Johnston JA
      • Bowman L
      • Jordan JB.
      Impact of caregiver and parenting status on time trade-off and standard gamble utility scores for health state descriptions.
      that elicited parent TTO values for health states related to their child's illness, there was a low TTO response rate. Approximately half of eligible survey participants attentively completed the TTO questions. Despite this censoring, group characteristics and preferences based on BWS values of those that completed the TTO questions were consistent with the overall survey sample, indicating that the TTO responses obtained are generalizable to a broader group of caregivers of patients with PWS.
      Previous studies
      • Chevreul K
      • Berg Brigham K
      • Clement MC
      • Poitou C
      • Tauber M
      Economic burden and health-related quality of life associated with Prader-Willi syndrome in France.
      ,
      • Lopez-Bastida J
      • Linertova R
      • Oliva-Moreno J
      • et al.
      Social/economic costs and health-related quality of life in patients with Prader-Willi syndrome in Europe.
      have found that 1 year spent with PWS is equivalent to a QALY between 0.40 and 0.81. Chevreul et al
      • Chevreul K
      • Berg Brigham K
      • Clement MC
      • Poitou C
      • Tauber M
      Economic burden and health-related quality of life associated with Prader-Willi syndrome in France.
      estimated that PWS was associated with a utility value of 0.41 by having a mixed group of child and adult patients with PWS or caregivers in France complete the EQ-5D. Lopez-Bastida et al
      • Lopez-Bastida J
      • Linertova R
      • Oliva-Moreno J
      • et al.
      Social/economic costs and health-related quality of life in patients with Prader-Willi syndrome in Europe.
      similarly used the EQ-5D to elicit utility values for PWS from patients and caregivers throughout Europe and found country-level PWS utility values ranging from 0.40 and 0.81. The acceptability of using preference-based generic instruments, such as the EQ-5D, for estimating QALYs in rare diseases is currently being debated.
      • Crossnohere NL
      • Fischer R
      • Lloyd A
      • Prosser LA
      • Bridges JFP.
      Assessing the Appropriateness of the EQ-5D for Duchenne Muscular Dystrophy: A Patient-Centered Study.
      ,
      • Efthymiadou O
      • Mossman J
      • Kanavos P.
      Health related quality of life aspects not captured by EQ-5D-5L: Results from an international survey of patients.
      ,
      • Sampson C
      • Garau M.
      How Should We Measure Quality of Life Impact in Rare Disease? Recent Learnings in Spinal Muscular Atrophy: Office of Health Economics.
      In some instances, members of the rare disease community have indicated that generic measures of health status fail to incorporate all domains of health that are important to them.
      • Efthymiadou O
      • Mossman J
      • Kanavos P.
      Health related quality of life aspects not captured by EQ-5D-5L: Results from an international survey of patients.
      The EQ-5D evaluates health based on 5 domains: mobility, self-care, usual activity, pain/discomfort, and anxiety/depression.
      The EuroQol Group
      EuroQol–a new facility for the measurement of health-related quality of life.
      Although these domains are associated with many symptoms of PWS, they do not directly capture and may not be sensitive to capture burdens from hyperphagia and obesity, outcomes of particular importance in patient-focused regulatory contexts when evaluating clinical therapies. Other studies, however, have found that the EQ-5D may be an appropriate measure of health status in certain rare diseases.
      • Crossnohere NL
      • Fischer R
      • Lloyd A
      • Prosser LA
      • Bridges JFP.
      Assessing the Appropriateness of the EQ-5D for Duchenne Muscular Dystrophy: A Patient-Centered Study.
      Our study builds on these previous studies by eliciting utility values not on actual patients with PWS but with the use of fixed vignettes that describe hypothetical patients with PWS that varied only on the presence or absence of obesity and hyperphagia. Few previous studies
      • Carroll AE
      • Downs SM.
      Improving decision analyses: parent preferences (utility values) for pediatric health outcomes.
      ,
      • Lloyd A
      • Hodgkins P
      • Sasane R
      • et al.
      Estimation of utilities in attention-deficit hyperactivity disorder for economic evaluations.
      have used fixed vignettes with a TTO exercise to elicit utility values for distinct child health states that vary solely based on severity of disease or the presence or absence of symptoms. In this study, these vignettes were successful in reducing variability of other symptoms experienced by individuals with PWS and helped to elicit 3 precise health state values from parents that differed significantly based on the presence or absence of obesity and hyperphagia. These values were also not influenced by specific caregiver-reported symptoms in their family member with PWS. This method is an important contribution to the field of child health utility elicitation.
      • Prosser LA
      • Hammitt JK
      • Keren R.
      Measuring health preferences for use in cost-utility and cost-benefit analyses of interventions in children.
      The significant differences between health states provide strong evidence of the value that caregivers place on treatments that produce symptom improvement. This evidence can be used both in regulatory contexts and in future economic evaluations of treatments. Some previous research has failed to find significant differences in child health utility values stratified by severity level when using caregiver proxy report of a child's current health state
      • Lavelle TA
      • Weinstein MC
      • Newhouse JP
      • Munir K
      • Kuhlthau KA
      • Prosser LA.
      Parent Preferences for Health Outcomes Associated with Autism Spectrum Disorders.
      because of respondent and child heterogeneity. Moreover, there has been broader conceptual, methodologic, and developmental issues with measuring quality of life in the pediatric setting.
      • De Civita M
      • Regier D
      • Alamgir AH
      • Anis AH
      • FitzGerald MJ
      • Marra CA.
      Evaluating health-related quality-of-life studies in paediatric populations.
      Our study also found that caregivers are willing to accept a child's early death in exchange for the improved health of that child. Several previous studies that have used a TTO method to elicit child health utilities have asked caregivers to hypothetically trade time from the end of their own life to improve the health of a real or hypothetical child.
      • Lavelle TA
      • Weinstein MC
      • Newhouse JP
      • Munir K
      • Kuhlthau KA
      • Prosser LA.
      Parent Preferences for Health Outcomes Associated with Autism Spectrum Disorders.
      ,
      • Prosser LA
      • Ray GT
      • O'Brien M
      • Kleinman K
      • Santoli J
      • Lieu TA.
      Preferences and willingness to pay for health states prevented by pneumococcal conjugate vaccine.
      • Lee GM
      • Salomon JA
      • LeBaron CW
      • Lieu TA.
      Health-state valuations for pertussis: methods for valuing short-term health states.
      • Kuppermann M
      • Nease Jr., RF
      • Ackerson LM
      • Black SB
      • Shinefield HR
      • Lieu TA
      Parents' preferences for outcomes associated with childhood vaccinations.
      However, these questions may be challenging for parents to answer, and answers may be biased if respondents account for the real-life implications of their answers (eg, not being around to care for their child because of the time traded) or answer to extremes (eg, trade entire lifespan to improve health of child).
      The QALYs derived in this study could have relevance to both premarket and postmarket decisions makers. Several drugs are currently in development for PWS obesity and hyperphagia.
      • Tan Q
      • Orsso CE
      • Deehan EC
      • et al.
      Current and emerging therapies for managing hyperphagia and obesity in Prader-Willi syndrome: A narrative review.
      Clinical outcome measures include changes in weight and caregiver reported improvements in hyperphagia, such as time spent talking about food and trying to steal food, measured with the Hyperphagia Questionnaire.
      • Fehnel SE
      • Brown TM
      • Nelson L
      • et al.
      Development of the Hyperphagia Questionnaire for Use in Prader-Willi Syndrome Clinical Trials.
      However, these measures do not indicate how improvements in hyperphagia and obesity correspond to reductions in disease burden. By estimating QALYs for multiple PWS health states, we identified the burden of obesity and hyperphagia and the QALY increase that is associated with better PWS health states. By estimating these QALYs using the TTO method, where parents accept a child's early death in exchange for improved hyperphagia and/or obesity, we further found how meaningful these changes are to caregivers. After drug approval, the preference values derived from this study will be important for future economic evaluations of these new drugs, which is particularly relevant in countries where heath technology assessment agencies play pivotal roles in regulating drug reimbursement policies.
      • Denis A
      • Mergaert L
      • Fostier C
      • Cleemput I
      • Simoens S.
      A comparative study of European rare disease and orphan drug markets.
      It may also inform policy for rare disease and orphan drug coverage in other countries,
      • Crossnohere NL
      • Fischer R
      • Crossley E
      • Vroom E
      • Bridges JF.
      The evolution of patient-focused drug development and Duchenne muscular dystrophy.
      especially in many low-income countries where a study such as this may not be feasible because of resource constraints and insufficient sample of people with rare diseases.
      • Iskrov G
      • Stefanov R.
      Post-marketing access to orphan drugs: a critical analysis of health technology assessment and reimbursement decision-making considerations.
      ,
      • Schlander M
      • Garattini S
      • Holm S
      • et al.
      Incremental cost per quality-adjusted life year gained? The need for alternative methods to evaluate medical interventions for ultra-rare disorders.
      PWS is a rare disease, and the burden of the disorder is likely unfamiliar to most people. To contextualize QALYs from the present study, we can compare them to those previously reported by Carroll and Downs
      • Carroll AE
      • Downs SM.
      Improving decision analyses: parent preferences (utility values) for pediatric health outcomes.
      for more common childhood conditions. The values derived from this study for PWS without hyperphagia and obesity (mean QALY, 0.91) are consistent with QALY values previously reported for mild bilateral vision loss or mild to moderate asthma. Values for PWS with hyperphagia but no obesity (mean QALY, 0.79) are consistent with those reported for moderate cerebral palsy or moderate intellectual disability. In addition, values for untreated PWS (mean QALY, 0.69) are consistent with those reported for severe seizure disorder. One important difference between the study by Carroll and Downs
      • Carroll AE
      • Downs SM.
      Improving decision analyses: parent preferences (utility values) for pediatric health outcomes.
      and our study is that the previous study administered the TTO exercises to a community sample and our survey was administered to a sample of caregivers. We chose estimate preferences from parents instead of society/community members because the primary goal of our study was to inform the PFDD process during regulatory approval, which requires the input of patients or proxies (ie, parents).
      Patient-Focused Drug Development Guidance: Collecting Comprehensive and Representative Input
      US Food and Drug Administration.
      For most illnesses, patients who have experienced an illness report higher QALY values for that condition compared with respondents from community sample because of the patient's lived experience and adaptation to the health condition.
      • Ubel PA
      • Loewenstein G
      • Jepson C.
      Whose quality of life? A commentary exploring discrepancies between health state evaluations of patients and the general public.
      This same adaptation also likely affects QALY values derived from experienced caregivers, which may make the values derived from our study higher than would be expected if the same questions were posed to a community sample.
      Our study has several limitations. We asked caregivers to report the symptoms their child with PWS had experienced, but these answers may be subject to recall bias, particularly for symptoms that may have resolved over time (ie, feeding difficultly in infants). Our study also did not elicit QALY values associated with the health of caregivers themselves. Previous research indicates that caring for a child with a developmental disability is associated with lower caregiver quality of life and higher rates of anxiety and depression.
      • Masefield SC
      • Prady SL
      • Sheldon TA
      • Small N
      • Jarvis S
      • Pickett KE.
      The Caregiver Health Effects of Caring for Young Children with Developmental Disabilities: A Meta-analysis.
      Future research is needed to examine how improving hyperphagia and obesity in patients with PWS may affect the QALYs of caregivers and other family members to provide a more complete picture of the value of new therapies.
      • Lavelle TA
      • D'Cruz BN
      • Mohit B
      • et al.
      Family spillover effects in pediatric cost-utility analyses.
      In addition, respondents to this survey are representative of the registry from which they were sampled but may not reflect the entire population of caregivers/people with PWS. Finally, although our results indicate that addressing hyperphagia could affect quality of life, which may be beneficial in health technology assessment, this does not imply that the application of cost-effectiveness in rare diseases is necessarily easy or appropriate.
      • Hyry HI
      • Stern A
      • Cox T
      • Roos J.
      Limits on use of health economic assessments for rare diseases.
      The survey was administered online, and approximately half of respondents did not complete the TTO task or provided repetitive answers that we considered to be invalid. Previous research indicates that TTO surveys administered online produce lower response rates than surveys administered in-person,
      • Norman R
      • King MT
      • Clarke D
      • Viney R
      • Cronin P
      • Street D.
      Does mode of administration matter? Comparison of online and face-to-face administration of a time trade-off task.
      which is concerning if the participants who answer the survey are in any way different from those who do not. However, our sensitivity analyses indicate that those who completed the TTO questions had similar characteristics and treatment priorities to those who did not, supporting the generalizability of our findings. We also made the decision to set any TTO amounts >20 years back to 20 years maximum, which essentially did not allow any respondent to value a health state as worse than dead. This produced conservative estimates of the impact of hyperphagia and obesity in PWS; the impact may be greater if the original responses reflected true preferences.
      Finally, it is also possible that using a different elicitation approach, such as the standard gamble or a modification to the TTO method, would have resulted in different QALY estimates.
      • Oppe M
      • Rand-Hendriksen K
      • Shah K
      • Ramos-Goñi JM
      • Luo N.
      EuroQol Protocols for Time Trade-Off Valuation of Health Outcomes.
      • Bleichrodt H
      • Johannesson M.
      Standard gamble, time trade-off and rating scale: experimental results on the ranking properties of QALYs.
      • Arnold D
      • Girling A
      • Stevens A
      • Lilford R.
      Comparison of direct and indirect methods of estimating health state utilities for resource allocation: review and empirical analysis.
      There are many ways to implement the TTO method, and the technique that we developed in collaboration with our community advisory board differs from other TTO protocols that have been developed, such as the TTO method used to value EQ-5D health states. The lack of standardization of the TTO method reduces the comparability of health state utilities derived in different studies, which can be problematic when these are used in cost-effectiveness analyses.
      • Arnesen T
      • Trommald M.
      Are QALYs based on time trade-off comparable? A systematic review of TTO methodologies.
      However, this is not a problem for the primary purposes of our study, which was to evaluate the within-study differences in QALYs among the PWS health states.

      Conclusions

      Changes in health states with and without hyperphagia and obesity in individuals with PWS can be measured using TTO methods and valued with the use of QALYs. PWS health states with controlled symptoms are associated with significantly higher QALYs compared with those for untreated PWS, and analyses indicate that these results are unbiased despite censoring in our data. The use of fixed vignettes of individuals with PWS were instrumental in our approach to evaluating these health states and allowed us to derive precise estimates of QALYs that differed significantly based on the presence or absence of hyperphagia and obesity. Although this study focused on premarket applications for the use of QALYs, findings might inform cost-effectiveness analysis for PWS and other rare diseases internationally.

      Acknowledgments

      We thank the study's advisory board: Theresa Strong, Nathalie Kayadjanian, Ann Scheimann, Shawn McCandless, Lauren Roth, Emma Roth, Sara Cotter, Maria Picone, Rob Lutz, Vonnie Sheadel, John Heybach, Conner Heybach, and Mark Greenberg. We also thank Jui-Hua (Sophie) Tsai for managing this study and the PWS Clinical Trial Consortium, Prader-Willi Syndrome Association, and Foundation for Prader-Willi Research for recruitment assistance. Dr Bridges designed the study. Drs Lavelle, Crossnohere, and Bridges participated in data analysis and manuscript drafting.

      References

        • Noyes J
        • Edwards RT.
        EQ-5D for the assessment of health-related quality of life and resource allocation in children: a systematic methodological review.
        Value Health. 2011; 14: 1117-1129
        • Longworth L
        • Yang Y
        • Young T
        • et al.
        Use of generic and condition-specific measures of health-related quality of life in NICE decision-making: a systematic review, statistical modelling and survey.
        Health Technology Assessment. 2014;
      1. Medical Device Innovation Consortium (MDIC). Patient centered benefit-risk project report: a framework for incorporating information on patient preferences regarding benefit and risk into regulatory assessments of new medical technology. Minneapolis, MN. 2015. https://mdic.org/wp-content/uploads/2015/05/MDIC_PCBR_Framework_Web.pdf. Accessed date: 14 May 2021.

        • de Bekker-Grob EW
        • Berlin C
        • Levitan B
        • et al.
        Giving patients’ preferences a voice in medical treatment life cycle: the PREFER public–private project.
        Springer, 2017
        • Patient-Focused Drug Development Guidance: Collecting Comprehensive and Representative Input
        US Food and Drug Administration.
        Silver Spring, MD2018
        • Brown L
        • Lee T-h
        • De Allegri M
        • Rao K
        • Bridges JF.
        Applying stated-preference methods to improve health systems in sub-Saharan Africa: a systematic review.
        Expert review of pharmacoeconomics & outcomes research. 2017; 17: 441-458
        • Beckham SW
        • Crossnohere NL
        • Gross M
        • Bridges JF.
        Eliciting Preferences for HIV Prevention Technologies: A Systematic Review.
        The Patient-Patient-Centered Outcomes Research. 2020: 1-24
        • Ozawa S
        • Wonodi C
        • Babalola O
        • Ismail T
        • Bridges J.
        Using best-worst scaling to rank factors affecting vaccination demand in northern Nigeria.
        Vaccine. 2017; 35: 6429-6437
        • Ozawa S
        • Grewal S
        • Bridges JF.
        Household size and the decision to purchase health insurance in Cambodia: results of a discrete-choice experiment with scale adjustment.
        Applied health economics and health policy. 2016; 14: 195-204
        • Hashim H
        • Beusterien K
        • Bridges JF
        • Amos K
        • Cardozo L.
        Patient preferences for treating refractory overactive bladder in the UK.
        International urology and nephrology. 2015; 47: 1619-1627
        • Bridges JF
        • Selck FW
        • Gray GE
        • McIntyre JA
        • Martinson NA.
        Condom avoidance and determinants of demand for male circumcision in Johannesburg, South Africa.
        Health policy and planning. 2011; 26: 298-306
        • Khadka J
        • Kwon J
        • Petrou S
        • Lancsar E
        • Ratcliffe J.
        Mind the (inter-rater) gap. An investigation of self-reported versus proxy-reported assessments in the derivation of childhood utility values for economic evaluation: A systematic review.
        Soc Sci Med. 2019; 240112543
        • Brower A.
        Is It Time To Take a Harder Look at the QALY?.
        Biotechnol Healthc. 2008; 5: 47-48
        • Crossnohere NL
        • Fischer R
        • Lloyd A
        • Prosser LA
        • Bridges JFP.
        Assessing the Appropriateness of the EQ-5D for Duchenne Muscular Dystrophy: A Patient-Centered Study.
        Medical decision making: an international journal of the Society for Medical Decision Making. 2021; 41: 209-221
        • Efthymiadou O
        • Mossman J
        • Kanavos P.
        Health related quality of life aspects not captured by EQ-5D-5L: Results from an international survey of patients.
        Health Policy. 2019; 123: 159-165
        • Patient-Focused Drug Development Draft Guidance: Methods to Identify What is Important to Patients
        US Food and Drug Administration.
        Silver Springs, MD2019
        • Whitehead SJ
        • Ali S.
        Health outcomes in economic evaluation: the QALY and utilities.
        British Medical Bulletin. 2010; 96: 5-21
        • Hauber AB
        • Johnson FR
        • Muhlbacher A
        • Marshall D.
        Maximum acceptale risk: Alternative measures for quantifying patients’ tolerance for theraputic benefit. International Society for Pharmacoeconomics and Outcomes Research.
        Prauge, Czech Republic. 2010;
        • Tsai JH
        • Scheimann AO
        • McCandless SE
        • Strong TV
        • Bridges JFP.
        Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best-worst scaling.
        Journal of medical economics. 2018; 21: 1230-1237
        • Tan Q
        • Orsso CE
        • Deehan EC
        • et al.
        Current and emerging therapies for managing hyperphagia and obesity in Prader-Willi syndrome: A narrative review.
        Obes Rev. 2020; 21: e12992
        • Fehnel SE
        • Brown TM
        • Nelson L
        • et al.
        Development of the Hyperphagia Questionnaire for Use in Prader-Willi Syndrome Clinical Trials.
        Value Health. 2015; 18: A25
        • Dykens EM
        • Maxwell MA
        • Pantino E
        • Kossler R
        • Roof E.
        Assessment of hyperphagia in Prader-Willi syndrome.
        Obesity (Silver Spring). 2007; 15: 1816-1826
      2. Kayadjanian N. PWS Clinical Trials Consortium Meets with FDA for a Critical Path Innovation Meeting: Foundation for Prader-Willi Research; 2019. https://www.fpwr.org/blog/pws-clinical-trials-consortium-meets-with-fda-for-a-critical-path-innovation-meeting Accessed 14 May 2021.

        • Albrecht GL
        • Devlieger PJ.
        The disability paradox: high quality of life against all odds.
        Social Sci Med. 1999; 48: 977-988
        • Simon NJ
        • Richardson J
        • Ahmad A
        • et al.
        Health utilities and parental quality of life effects for three rare conditions tested in newborns.
        J Patient-Reported Outcomes. 2019; 3: 4
        • Pyne JM
        • Fortney JC
        • Tripathi S
        • Feeny D
        • Ubel P
        • Brazier J.
        How bad is depression? Preference score estimates from depressed patients and the general population.
        Health Serv Res. 2009; 44: 1406-1423
        • Bridges JF
        • Kinter ET
        • Schmeding A
        • Rudolph I
        • Mühlbacher A.
        Can patients diagnosed with schizophrenia complete choice-based conjoint analysis tasks?.
        The Patient: Patient-Centered Outcomes Research. 2011; 4: 267-275
        • The EuroQol Group
        EuroQol–a new facility for the measurement of health-related quality of life.
        Health Policy. 1990; 16: 199-208
        • Perfetto EM
        • Burke L
        • Oehrlein EM
        • Epstein RS.
        Patient-focused drug development: a new direction for collaboration.
        Med Care. 2015; 53: 9-17
        • Hsu JC
        • Wu H-C
        • Feng W-C
        • Chou C-H
        • Lai EC-C
        • Lu CY.
        Disease and economic burden for rare diseases in Taiwan: A longitudinal study using Taiwan's National Health Insurance Research Database.
        PloS One. 2018; 13e0204206
        • Wästfelt M
        • Fadeel B
        • Henter JI.
        A journey of hope: lessons learned from studies on rare diseases and orphan drugs.
        J Intern Med. 2006; 260: 1-10
        • Hollin IL
        • Young C
        • Hanson C
        • Bridges JF
        • Peay H.
        Developing a patient-centered benefit-risk survey: a community-engaged process.
        Value Health. 2016; 19: 751-757
        • Miller JL
        • Lynn CH
        • Driscoll DC
        • et al.
        Nutritional phases in Prader-Willi syndrome.
        Am J Med Genet. Part A. 2011; 155a: 1040-1049
        • Tsai JH
        • Janssen E
        • Bridges JF.
        Research as an event: a novel approach to promote patient-focused drug development.
        Patient Preference Adherence. 2018; 12: 673-679
        • Rodrıguez G.
        Models for count data with overdispersion.
        Princeton. November 2013;
        • Hanley JA
        • Negassa A
        • Edwardes MDd
        • Forrester JE.
        Statistical analysis of correlated data using generalized estimating equations: an orientation.
        Am J Epidemiol. 2003; 157: 364-375
        • Bridges JF
        • Tsai J-H
        • Janssen E
        • Crossnohere NL
        • Fischer R
        • Peay H.
        How do members of the Duchenne and Becker muscular dystrophy community perceive a discrete-choice experiment incorporating uncertain treatment benefit? An application of research as an event.
        Patient-Centered Outcomes Res. 2019; 12: 247-257
        • Patient Preference Information–Voluntary Submission, Review in Premarket Approval Applications, Humanitarian Device Exemption Applications, and De Novo Requests, and Inclusion in Decision Summaries and Device Labeling
        US Food and Drug Administration.
        Silver Springs, MD2016
        • Fries JF.
        Aging, natural death, and the compression of morbidity.
        N Engl J Med. 1980; 303: 130-135
      3. The MaxDiff System Technical Paper. Sawtooth Software Technical Paper Series. 2020. https://sawtoothsoftware.com/uploads/sawtoothsoftware/originals/0114c81d-43ca-4339-bd30-fbf45146e8b2.pdf Accessed 14 May 2021.

        • Matza LS
        • Boye KS
        • Feeny DH
        • Johnston JA
        • Bowman L
        • Jordan JB.
        Impact of caregiver and parenting status on time trade-off and standard gamble utility scores for health state descriptions.
        Health Qual Life Outcomes. 2014; 12: 48
        • Chevreul K
        • Berg Brigham K
        • Clement MC
        • Poitou C
        • Tauber M
        Economic burden and health-related quality of life associated with Prader-Willi syndrome in France.
        J Intellectual Disabil Res. 2016; 60: 879-890
        • Lopez-Bastida J
        • Linertova R
        • Oliva-Moreno J
        • et al.
        Social/economic costs and health-related quality of life in patients with Prader-Willi syndrome in Europe.
        Eur J Health Econ. 2016; 17: 99-108
        • Sampson C
        • Garau M.
        How Should We Measure Quality of Life Impact in Rare Disease? Recent Learnings in Spinal Muscular Atrophy: Office of Health Economics.
        2019
        • The EuroQol Group
        EuroQol–a new facility for the measurement of health-related quality of life.
        Health Policy. 1990; 16: 199-208
        • Carroll AE
        • Downs SM.
        Improving decision analyses: parent preferences (utility values) for pediatric health outcomes.
        J Pediatr. 2009; 155 (e21–25): 21-25
        • Lloyd A
        • Hodgkins P
        • Sasane R
        • et al.
        Estimation of utilities in attention-deficit hyperactivity disorder for economic evaluations.
        Patient-Centered Outcomes Res. 2011; 4: 247-257
        • Prosser LA
        • Hammitt JK
        • Keren R.
        Measuring health preferences for use in cost-utility and cost-benefit analyses of interventions in children.
        Pharmacoeconomics. 2007; 25: 713-726
        • Lavelle TA
        • Weinstein MC
        • Newhouse JP
        • Munir K
        • Kuhlthau KA
        • Prosser LA.
        Parent Preferences for Health Outcomes Associated with Autism Spectrum Disorders.
        PharmacoEconomics. 2019; 37: 541-551
        • De Civita M
        • Regier D
        • Alamgir AH
        • Anis AH
        • FitzGerald MJ
        • Marra CA.
        Evaluating health-related quality-of-life studies in paediatric populations.
        Pharmacoeconomics. 2005; 23: 659-685
        • Prosser LA
        • Ray GT
        • O'Brien M
        • Kleinman K
        • Santoli J
        • Lieu TA.
        Preferences and willingness to pay for health states prevented by pneumococcal conjugate vaccine.
        Pediatrics. 2004; 113: 283-290
        • Lee GM
        • Salomon JA
        • LeBaron CW
        • Lieu TA.
        Health-state valuations for pertussis: methods for valuing short-term health states.
        Health Qual Life Outcomes. 2005; 3: 17
        • Kuppermann M
        • Nease Jr., RF
        • Ackerson LM
        • Black SB
        • Shinefield HR
        • Lieu TA
        Parents' preferences for outcomes associated with childhood vaccinations.
        Pediatr Infect Dis J. 2000; 19: 129-133
        • Denis A
        • Mergaert L
        • Fostier C
        • Cleemput I
        • Simoens S.
        A comparative study of European rare disease and orphan drug markets.
        Health Policy. 2010; 97: 173-179
        • Crossnohere NL
        • Fischer R
        • Crossley E
        • Vroom E
        • Bridges JF.
        The evolution of patient-focused drug development and Duchenne muscular dystrophy.
        Expert Rev Pharmacoecon Outcomes Res. 2020; 20: 57-68
        • Iskrov G
        • Stefanov R.
        Post-marketing access to orphan drugs: a critical analysis of health technology assessment and reimbursement decision-making considerations.
        Orphan Drugs Res Rev. 2014; 4: 1-9
        • Schlander M
        • Garattini S
        • Holm S
        • et al.
        Incremental cost per quality-adjusted life year gained? The need for alternative methods to evaluate medical interventions for ultra-rare disorders.
        J Comparative Effectiveness Res. 2014; 3: 399-422
        • Ubel PA
        • Loewenstein G
        • Jepson C.
        Whose quality of life? A commentary exploring discrepancies between health state evaluations of patients and the general public.
        Qual Life Res. 2003; 12: 599-607
        • Masefield SC
        • Prady SL
        • Sheldon TA
        • Small N
        • Jarvis S
        • Pickett KE.
        The Caregiver Health Effects of Caring for Young Children with Developmental Disabilities: A Meta-analysis.
        Matern Child Health J. 2020; 24: 561-574
        • Lavelle TA
        • D'Cruz BN
        • Mohit B
        • et al.
        Family spillover effects in pediatric cost-utility analyses.
        Appl Health Econ Health Policy. 2019; 17: 163-174
        • Hyry HI
        • Stern A
        • Cox T
        • Roos J.
        Limits on use of health economic assessments for rare diseases.
        Q J Med. 2014; 107: 241-245
        • Norman R
        • King MT
        • Clarke D
        • Viney R
        • Cronin P
        • Street D.
        Does mode of administration matter? Comparison of online and face-to-face administration of a time trade-off task.
        Qual Life Res. 2010; 19: 499-508
        • Oppe M
        • Rand-Hendriksen K
        • Shah K
        • Ramos-Goñi JM
        • Luo N.
        EuroQol Protocols for Time Trade-Off Valuation of Health Outcomes.
        Pharmacoeconomics. 2016; 34: 993-1004
        • Bleichrodt H
        • Johannesson M.
        Standard gamble, time trade-off and rating scale: experimental results on the ranking properties of QALYs.
        J Health Econ. 1997; 16: 155-175
        • Arnold D
        • Girling A
        • Stevens A
        • Lilford R.
        Comparison of direct and indirect methods of estimating health state utilities for resource allocation: review and empirical analysis.
        BMJ. 2009; 339: b2688
        • Arnesen T
        • Trommald M.
        Are QALYs based on time trade-off comparable? A systematic review of TTO methodologies.
        Health Econ. 2005; 14: 39-53