Skip to main content

Currently Skimming:

3 Measures and Strategies for Obtaining Health Benefit Values for Regulatory Analysis
Pages 67-129

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.


From page 67...
... based on these metrics. We first introduce criteria for selecting among effectiveness measures for use in regulatory analysis, and then discuss various approaches in light of these criteria.
From page 68...
... In the following section we consider sources of health state values for regulatory analysis and review four commonly used generic HRQL survey instruments. The fifth section identifies data collection and research priorities as well as promising developments for improving the measurement of health effects for regulatory analysis.
From page 69...
... Because several generic indexes are well established and easy to use, the Committee expects that they will often be applied in regulatory analysis in the near term. As already discussed, regulatory analysts lack the time or resources to engage in the development of instruments for health status valuation in the context of individual regulatory analysis.
From page 70...
... A measure should be widely applicable to a range of health states and condi tions. It should be sensitive, that is, responsive to change, and not exhibit floor or ceiling effects in the range of anticipated effects.
From page 71...
... Generic HRQL instruments are designed for application to a wide range of health states that can result from a variety of health-related risks or interventions. Still, as described below, each generic instrument has distinctive features absent from the others.
From page 72...
... One indication of a measure's acceptability is the extent to which valuation survey respondents comprehend, and are willing to engage in, the preference elicitation exercise. In a broader sense, the ethical commitments and implications of the HRQL instrument and the health state values it generates must be viewed as legitimate by the ultimate users of the analytic results.
From page 73...
... In these cases, calculation of life years gained may capture the majority of the impact of the intervention on health. However, this will not be the case for programs or regulations that improve health and functioning but do not significantly change life expectancy, such as one might expect with mitigation of environmental exposures to lead or mercury.
From page 74...
... Describing Health States HRQL measurement relies on concepts such as "health status," "functional status," "well-being," and "quality of life." Although these terms, along with "health-related quality of life," are often applied interchangeably, in fact they encompass narrower or broader arrays of domains, with "health status" denoting a more restrictive concept and "quality of life" a more extensive one. Table 3-1 presents concepts and domains that fall within these broader rubrics.
From page 75...
... All generically described health states used in HRQL indexes depend on psychometric scaling and concepts to some degree. Such generic indexes thus share common features with health profiling instruments, such as the SF-36.
From page 76...
... 76 VALUING HEALTH TABLE 3-1 Concepts and Domains Used in Defining Self-Reported Health Status, Quality of Life, and Health-Related Quality of Life Concepts Domains Attributes Symptoms Reports of physical and psychological Frequency, severity, symptoms or sensations not directly bothersomeness observable, such as energy and fatigue, nausea, and irritability Functional status Frequency, difficulty, severity, ability, with help Physical Functional limitations and activity restrictions, such as self-care, walking, mobility, sleep, sexual Psychological Positive or negative affect and cognitive, such as anger, alertness, self-esteem, sense of well-being, distress Social Limitations in work or school, participation in community Health perceptions Frequency, severity/ intensity, satisfaction Global General ratings of health and quality of life, such as satisfaction or overall well-being Worries and About health, finances, the future concerns Spiritual Meaning and purpose of life or relationship to the universe Disadvantage/ Perceptions of stigma or reports of Frequency, impact opportunity discrimination because of health condition Resiliency Reports of ability to cope or withstand Frequency, satisfaction, stress and illness ability Environmental Evaluations of personal safety, Satisfaction, importance adequacy of housing, respect, freedom, and so on SOURCE: Reprinted from Patrick and Chiang (2000, Table 1)
From page 77...
... Unless new surveys are conducted to elicit values for specific health states, the elicitation technique is part and parcel of the choice of a generic, multiattribute HRQL index. Thus, although the following discussion addresses elicitation methods in isolation from other features of valuation surveys, in practice these methods are not readily mixed and matched with
From page 78...
... The SG is the only preference elicitation method directly linked to the axioms of expected utility theory. In order to establish the relative values of various health states on an interval scale, respondents must determine the conditions of indifference or equivalence between two outcomes.
From page 79...
... Direct Rating: Category Rating and Visual Analogue Scales Direct rating approaches to preference elicitation ask respondents to assign a single number to a health state, usually on a scale of 0 to 100, with these anchors being the worst and best imaginable health states, or death and perfect health. Visual aids, such as the "feeling thermometer" in the EuroQol Group's generic HRQL survey instrument, the EuroQoL-5D (EQ-5D)
From page 80...
... . Health state values generated by VAS tend to correlate more closely with health status indicators such as pain, functioning, and clinical symptoms, and with health status profile scores, than do values generated by SG and TTO methods (Brazier et al., 1999a)
From page 81...
... . Comparisons Among Elicitation Methods This review reinforces the caveat stated at the beginning of the section: Each approach elicits relative health state values that incorporate different characteristics of the health states or aspects of the choices posed.
From page 82...
... Reliability Table 3-2 presents intrarater test­retest reliability results for the SG, TTO, and VAS methods from studies that resurveyed respondents at different time intervals, ranging from less than one week to a year. None TABLE 3-2 Intrarater Test­Retest Reliability of the Standard Gamble, Time Trade-Off, and Visual Analogue Scale Techniques Test­Retest Reliability Standard Gamble Time Trade-Off Visual Analogue Scale 0.80a 0.87a 0.77a 1 week or less 0.77­0.79b 0.70­0.95b 0 .
From page 83...
... . Empirical validity The SG and TTO methods have been compared in terms of producing logically consistent orderings of health states.
From page 84...
... Taken together, these measurement biases lead to higher SG values than TTO values for the same health states. Furthermore, many respondents are unwilling to accept any risk of death, or trade off any longevity, for a health improvement, leading to relatively high values for impaired health states (Reed et al., 1993)
From page 85...
... , and willingness-to-pay values for 13 generically described health states (taken from the EuroQol Group's EQ-5D classification system) did, however, find a consistent relationship between RS and TTO mean values, as shown in Figure 3-2.
From page 86...
... The choice among alternative preference elicitation techniques is embedded in the choice of generic index, because each index relies on a valuation survey that employed a particular elicitation method. If health state values are elicited directly in new surveys, however, the researcher must choose a preference elicitation method.
From page 87...
... . An additional assumption that is required when health states vary over the life span is that preferences for health in different time periods are additive, in accordance with individual preferences for health.
From page 88...
... Although the HYE has an advantage in that some of the restrictive assumptions associated with QALYs do not apply, preferences must be elicited for specific health profiles, or sequences of health states, rather than for individual health states as with QALYs. Although proponents of the HYE metric contend that the greater methodological demands of the approach are justified in terms of its closer adherence to the theoretical conditions of utility theory, critics counter that developing an empirical base of HYE values for widespread use is not practical.
From page 89...
... The DALY index scale is an inversion of the QALY scale: for DALYs, 0 corresponds to perfect health and 1 to death. DALY index values correspond to specific health conditions rather than to generically characterized health states.
From page 90...
... These measures do not adequately account for the value attached to saving lives relative to improving health or to the priority that may be given to improving outcomes for the most severely impaired, regardless of the size of the improvement. QALYs measure only the size of an improvement in health and disregard health state starting and endpoints.
From page 91...
... Furthermore, values for the wide range of health conditions considered in regulatory analysis are not likely to be developed in the near term using these approaches, given the complexities of establishing values (such as conditioning health state values on duration or transitions from prior health states) and the expense of related research.
From page 92...
... The following discussion reviews various ways of obtaining preference-based HRQL values, focusing on the information needs and constraints of those involved with risk regulation. This section reviews: · Primary elicitation of health state index values for specific conditions, · Four commonly used generic HRQL survey instruments or indexes, · Use of condition-specific indexes, · Use of experts to assign health states, · Use of data from routine population surveys, · Use of health state index values from prior studies and benefit transfer practices, and · Assessing uncertainty in the estimation of health-related effects from regulatory interventions.
From page 93...
... As a result, the sources of health state index values discussed in the remainder of this section are likely to be the more feasible options for regulatory CEA in the near term. Generic HRQL Indexes An alternative to directly eliciting preferences for specific conditions is to use a multiattribute health state classification system with predetermined index values for generically described health states.
From page 94...
... Index values for health states using multiattribute generic instruments
From page 95...
... . See Appendix B for complete descriptions and sources for these generic indexes.
From page 96...
... S . community 3,773 with complete TTO; VAS for own 45/15 243 residents age 18+ data/2002/59% health state only SF-6D U.K.
From page 97...
... Weighting formulas can be additive or multiplicative under either approach. Each of the generic indexes used in the case studies and described below has at least one set of values for all possible health states that is based on a general population or community valuation survey, presented in Table 3-4.
From page 98...
... . The survey asked respondents to consider the relative value of being in the health state in question for a single day.
From page 99...
... Each of the 8 HUI-3 domains has 5 or 6 levels, resulting in 972,000 possible health states and making it in this sense the most detailed of the four instruments with respect to the measurement of generic health-related characteristics. Availability Since 1990, the HUI-3 has been included in every major Canadian population health survey, and more recently in three major U.S.
From page 100...
... In a side-by-side comparison to the HUI-3, little difference was found between the EQ-5D and the HUI-3 with respect to their ability to discriminate between respondents with and without a variety of self-reported health conditions. Those who are assigned to the best EQ-5D health state are, however, somewhat more differentiated by the HUI-3 (Houle and Berthelot, 2000)
From page 101...
... In the late 1990s, a British research group developed a simplified six-dimension health state classification system derived from the data collected in the SF-36. The SF-6D instruments use 11 items from the SF-36 (Brazier et al., 2002)
From page 102...
... For SF-6D states taken from the SF-36 version 2, each domain has from 4 to 6 levels, defining a possible total of 18,000 health states. Valuation A representative sample of 836 residents of the United Kingdom participated in interviews and ranked and then valued a total of 249 SF-6D health states from the SF-36 version 2 (each participant rated 6 health states)
From page 103...
... The wider compass of the domains and attribute levels of a generic HRQL instrument, which make it less attuned to any particular health condition and its impacts on symptoms and function, also ensures that it can be applied broadly and provide comparability of results across health conditions. Although the PCEHM recommended that analysts use generic indexes, it concluded that, if disease-specific classification systems are used, health states still should be framed in terms of overall health.
From page 104...
... Children's HRQL measurement has been handled in several ways. First, parents and clinicians have served as proxy respondents, both in characterizing children's HRQL and in valuing children's health states and outcomes.
From page 105...
... . Griebsch and colleagues argue that the evidence base for developing best practices, both in the characterization and description of health states and in valuing them for children, has yet to be established.
From page 106...
... Generic HRQL survey instruments have been developed or modified for ad ministration to children, and even more have been developed to assess HRQL in children with a specific disease. In a survey of the field of pediatric HRQL instru Assignment of Health States by Experts or Other Proxies Proxies are used in HRQL assessments for a variety of reasons.
From page 107...
... . One study that evaluated the ability of children with asthma, ages 7 through 17, to comprehend and provide reliable re sponses to questions eliciting their preferences for different health states conclud ed that at least sixth-grade reading skills were necessary for SG exercises and that at least second-grade reading skills were necessary for using a VAS technique (Juniper et al., 1997)
From page 108...
... In these cases, index values for health states are obtained separately from community or general population valuation surveys. The Committee explored this approach in the three case studies conducted as part of its investigations.
From page 109...
... Despite the difficulties and uncertainties engendered by an expert elicitation approach to applying generic indexes, as well as its cost, several features of regulatory analysis make such approaches potentially necessary. First, the health states of interest may differ from those measured in clinical outcomes studies (as discussed in a following section on use of index values from prior studies)
From page 110...
... In addi Approaches Based on Population Survey Data The PCEHM urged the development of a standard catalogue of index values for "well-described health states" that would facilitate valid comparisons of CEA across conditions and illnesses and eliminate the need for collecting primary data for every analysis (Gold et al., 1996b)
From page 111...
... used the MEPS to develop EQ-5D index values for a number of chronic conditions, based on pooled MEPS data for the years 2000, 2001, and 2002 for respondents ages 18 or older. MEPS includes data on sociodemographic characteristics as well as responses to the EQ-5D health status questionnaire; valid responses were received for about 38,000 unique respondents.
From page 112...
... NOTES: EVGGFP = Five-item global health status measure: excellent, very good, good, fair, poor MEPS = Medical Expenditure Panel Survey NHANES = National Health and Nutrition Examination Survey NHIS = National Health Interview Survey SEER = Surveillance, Epidemiology, and End Results Program attribute scores using a model derived from a valuation survey of a representative sample of the U.S. adult population (Shaw et al., 2005; see Table 34 for a summary of the survey)
From page 113...
... and conditions activity limitations EQ-5D TTO 68 conditions NHIS self-rated health Statistically inferred 21 conditions status; chronic conditions and 2 co-morbid from NHIS + SEER health states See Fryback et al. Statistically inferred, based on 33 chronic Fryback et al.
From page 114...
... Condition-specific values could be particularly useful in the context of regulatory analysis because the risk assessments underlying these analyses frequently report health-related impacts in terms of cases of particular diseases. Incorporation of Health Profiles and HRQL Questions and Instruments in Routine Population Surveys Routine and periodic national health surveys in the United States have included various health profiles, HRQL questions, and generic HRQL instruments over the past few decades.
From page 115...
... ; range: Risk Factor 18+/telephone/ Days" measure 32­66%; conducted Surveillance annual, continuous within each state by System health department National ~5,000 adults and "Health Days" Each survey focused Health and children/personal questions administered on particular health Nutrition interview, physical to all participants problem in addition Examination exam, lab tests/annual 12+ years to core data Survey Medicare ~16,000 Medicare SRHS, ADL, IADL, Current beneficiaries/personal chronic conditions Beneficiary interview/annual Survey Medicare ~200,000 initially, SRHS, "Healthy Survey of Medicare Days," SF-36, ADL, beneficiaries in Health 60,000 follow-up Outcomes (longitudinal) /mail chronic conditions managed care plans; Survey with phone follow- 1,000 respondents/ up/annual plan Medicare ~200,000/mail with SRHS, SF-12, ADL 600 Medicare Fee-for-Service phone follow-up/ beneficiaries in each CAHPS annual geographic area Medicare+ ~200,000/mail with SRHS 600 managed care Choice phone follow-up/ enrollees per plan CAHPS annual area NOTES: ADL = activities of daily living; IADL = immediate activities of daily living; LE = life expectancy; SRHS = self-reported health status; Healthy Days measure: core includes four questions encompassing SRHS, number of physically and/or mentally unhealthy days within the past month, and restricted activity days within the past month.
From page 116...
... Increased reliance on HRQL measurement for regulatory analysis will require more regular and coordinated surveys for valuation and establishment of population baselines. Health State Index Values from Prior Studies and Benefits Transfer When it is not practicable for analysts to conduct primary research on HRQL values for the specific health states and affected populations addressed by regulatory CEAs, another alternative is to use estimates from published research, commonly referred to as "off-the-shelf" values or preference weights.8 This strategy, known as "benefits transfer" by welfare economists, refers specifically to using values estimated in one context (the "study scenario")
From page 117...
... Nevertheless, this registry is a convenient source of health state index values for regulatory CEA and has been used by agencies such as FDA and EPA for this purpose. Box 3-6 describes the registry in greater detail.
From page 118...
... The format for reviewing potentially applicable index values is also useful for deriving possible ranges of values for uncertainty analysis. The first step in applying index values from the research literature is to define the health endpoints in the regulatory analysis as precisely and accurately as possible.
From page 119...
... The Committee's review of published studies for applicable health state values for the air quality case study revealed both the advantages and drawbacks of using index values from prior studies for regulatory analysis. On the positive side, it confirmed that the published literature can be a fruitful source of health state values for at least some regulatory health endpoints, and that using index values from the published literature is a relatively simple and inexpensive approach.
From page 120...
... after each acute exacerbation as well as once every 3 months. During the study period, the health state index values for these patients averaged 0.79 or 0.76 (depending on the treatment)
From page 121...
... report index values for chronic bronchitis that range from 0.37 to 0.75, depending on the study approach, the disease sever ity, and the age of the patient. Estimates for post-MI health states also varied, in part because of the different populations studied, the different approaches to HRQL measurement used, and the different severities of illness considered.
From page 122...
... Even taking the quantified estimates of cases averted as givens, however, uncertainty remains in the characterization and measurement of HRQL effects of those conditions. At least four aspects of HRQL measurement contribute to the uncertainty of the ultimate values assigned to the estimated health-related impact of a regulation: · Variability in preferences across individuals, which contributes to uncertainty in estimating population means; · Variability in the estimation of preferences for health states depending on the elicitation technique; · Differences in the specificity and scope of attributes included by the generic HRQL instruments; and · The statistical models that assign relative health state values for each of the generic instruments.
From page 123...
... Last, some experts expressed skepticism about the ability of clinicians to characterize the impact of a condition on patients' functioning and experience, despite having professional familiarity with the condition. RESEARCH AND DEVELOPMENT OF METRICS AND VALUATION METHODOLOGIES From the many fruitful avenues of research in the measurement and valuation of health-related quality of life, we focus on three issues with particular relevance to regulatory CEA: · correlating and estimating conversion factors among generic indexes so that values based on different instruments can be compared; · using information about ordinal rankings of health states to develop HRQL value scales with interval properties; and · applying insights and best practices from willingness-to-pay survey research to HRQL valuation.
From page 124...
... Using Ordinal Data for HRQL Valuation Ranking of health states is often used as a preliminary step in preference elicitation exercises involving TTOs or SGs. Recent studies have explored using aggregated ranking data to predict health state valuations that closely match interval-level values produced by TTO methods (Salomon, 2003; Salomon and Murray, 2004)
From page 125...
... In particular, the Committee has formulated criteria for the selection of HRQL instruments and characterized alternative strategies for obtaining health state values for use in QALYbased CEA of regulatory interventions. In great measure, our recommendations conform to the guidelines and underlying rationales of the PCEHM, whose 1996 report constitutes the reference standard of best practices in CEA for clinical and public health interventions.
From page 126...
... Australian Community: N = 396 Inpatients: N = 266 (2001) community Outpatients: N = 334/NA population and hospital inpatients and outpatients age 16+ NOTES: ADL = activities of daily living; AQoL = Assessment of Quality of Life instrument; EVGGFP = five-item global health status measure: excellent, very good, good, fair, poor; NHIS = National Health Interview Survey; NMES = National Medical Expenditure Survey; WHOQOL-Bref = World Health Organization Quality of Life abbreviated assessment instrument
From page 127...
... R2 = 0.86 for conditions Linking NMES 7 conditions: diabetes, Yes: EQ-5D and HUI-1 imputations responses to HUI-1 atherosclerosis, cancer, had correlations ranging between and EQ-5D myocardial infarct, 67% and 74% questions heart disease, hypertension, stroke SF-36; HUI-2; No SF-36 and HUI-2: 50% of variation chronic disease in HUI-2 predicted by SF-36 scores score SF-12; EQ-5D; No HUI-3 and EQ-5D: 0.69; predicted HUI-3 HUI/HUI: 0.71; predicted EQ w/EQ: 0.77 SF-12; EQ-5D No Regression of EQ-5D scores onto mental and physical component summary scores of SF-12; physical component R2 = 0.67; mental component R2 = 0.47 SF-12; EQ-5D EQ-5D values reported Mean EQ-5D scores predicted from for: asthma, diabetes, mean physical and mental emphysema, high blood component summary scores pressure, heart attack, R2 = 0.61 stroke AQoL; SF-6D (36) ; No Spearman correlations of AQoL WHOQOL-Bref; with EQ-5D: 0.73; HUI-3: 0.74; EQ-5D; HUI-3; 15D: 0.80; SF-6D: 0.74 Finnish 15D
From page 128...
... In practice, we recognize that such original research will often not be possible to support regulatory analysis. The use of generic indexes, possibly with expert characterization of the health states of interest, and the transfer of health state values from existing research databases are the more likely, and also acceptable, approaches.
From page 129...
... In particular, establishing the relationships among and conversion factors for estimates derived from the most commonly used generic HRQL instruments would make integration and synthesis of the results from different studies possible and thus expand the tools and data available for regulatory analysis. In addition, it would improve the reliability of costeffectiveness comparisons among different analyses and regulations.


This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More information on Chapter Skim is available.