Lowering the 'floor' of the SF-6D scoring algorithm using a lottery equivalent method

A Spanish value set for the SF-6D based on the SF-12 v1

AIM

This paper reports the first estimation of an SF-6D value set based on the SF-12 for Spain.

METHODS

A representative sample (n = 1020) of the Spanish general population valued a selection of 56 hypothetical SF-6D health states by means of a probability lottery equivalent (PLE) method. The value set was derived using both random effects and mean models estimated by ordinary least squares (OLS). The best model was chosen on the basis of its predictive ability assessed in terms of mean absolute error (MAE).

RESULTS

The model yielding the lowest MAE (0.075) was that based on main effects using OLS. Pain was the most significant dimension in predicting health state severity. Comparison with the previous SF-6D (SF-36) model estimated for Spain revealed no significant differences, with a similar MAE (0.081). Nevertheless, the new SF-6D (SF-12) model predicted higher utilities than those generated by the SF-6D (SF-36) scoring algorithm (minimum value - 0.071 vs - 0.357).

CONCLUSION

A value set for the SF-6D (SF-12) based on Spanish general population preferences elicited by means of a PLE technique is successfully estimated. The new estimated SF-6D (SF-12) preference-based measure provides a valuable tool for researchers and policymakers to assess the cost-effectiveness of new health technologies in Spain.

The Short Form 6 Dimensions (SF-6D): Development and Evolution

This paper considers the development and evolution of the short-form 6 dimensions (SF-6D), a generic preference-weighted measure consisting of a health classification with accompanying value set that was developed from one of the widest used health related quality of life measures, the SF-36 health survey. This enabled health state utility values to be directly generated from SF-36 and SF-12 data for a range of purposes, including to produce quality adjusted life years for use in economic evaluation of healthcare interventions across a range of different conditions and treatments. This paper considers the rationale for the development of the measure, the development process, performance and how the SF-6D has evolved since its conception. This includes the development of an updated version, SF-6D version 2 (SF-6Dv2), which was generated to deal with some criticisms of the first version, and now includes a standalone version for inclusion in studies without relying on use of SF-36 or SF-12. Valuation methods have also evolved, from standard gamble in-person interviews to online discrete choice experiment surveys. International work related to the SF-6Dv1 and SF-6Dv2 is considered. We also consider recommendations for use, highlighting key psychometric evidence and reimbursement agency recommendations.

Comparison of four approaches in eliciting health state utilities with SF-6Dv2

OBJECTIVE

To empirically compare four preference elicitation approaches, the discrete choice experiment with time (DCETTO), the Best-Worst Scaling with time (BWSTTO), DCETTO with BWSTTO (DCEBWS), and the Standard Gamble (SG) method, in valuing health states using the SF-6Dv2.

METHODS

A representative sample of the general population in Quebec, Canada, completed 6 SG tasks or 13 DCEBWS (i.e., 10 DCETTO followed by 3 BWSTTO). Choice tasks were designed with the SF-6Dv2. Several models were used to estimate SG data, and the conditional logit model was used for the DCE or BWS data. The performance of SG models was assessed using prediction accuracy (mean absolute error [MAE]), goodness of fit using Bayesian information criterion (BIC), t-test, Jarque-Bera (JB) test,  Ljung-Box (LB) test, the logical consistency of the parameters, and significance levels. Comparison between approaches was conducted using acceptability (self-reported difficulty and quality levels in answering, and completion time), consistency (monotonicity of model coefficients), accuracy (standard errors), dimensions coefficient magnitude, correlation between the value sets estimated, and the range of estimated values. The variance scale factor was computed to assess individuals' consistency in their choices for DCE and BWS approaches.

RESULTS

Out of 828 people who completed SG and 1208 for DCEBWS tasks, a total of 724 participants for SG and 1153 for DCE tasks were included for analysis. Although no significant difference was observed in self-reported difficulties and qualities in answers among approaches, the SG had the longest completion time and excluded participants in SG were more prone to report difficulties in answering. The range of standard errors of the SG was the narrowest (0.012 to 0.015), followed by BWSTTO (0.023 to 0.035), DCEBWS (0.028 to 0.050), and DCETTO (0.028 to 0.052). The highest number of insignificant and illogical parameters was for BWSTTO. Pain dimension was the most important across dimensions in all approaches. The correlation between SG and DCEBWS utility values was the strongest (0.928), followed by the SG and BWSTTO values (0.889), and the SG and DCETTO (0.849). The range of utility values generated by SG tended to be shorter (-0.143 to 1) than those generated by the other three methods, whereas BWSTTO (-0.505 to 1) range values were shorter than DCETTO (-1.063 to 1) and DCEBWS (-0.637 to 1). The variance scale factor suggests that respondents had almost similar level of certainty or confidence in both DCE and BWS responses.

CONCLUSION

The SG had the narrowest value set, the lowest completion rates, the longest completion time, the best prediction accuracy, and produced an unexpected sign for one level. The BWSTTO had a narrower value set, lower completion time, higher parameter inconsistency, and higher insignificant levels compared to DCETTO and DCEBWS. The results of DCEBWS were more similar to SG in number of insignificant and illogical parameters, and correlation.

Comparison of Caregiver and General Population Preferences for Dependency-Related Health States

OBJECTIVE

We assess whether the preferences regarding dependency-related health states as stated by informal caregivers are aligned with those expressed by the general population.

METHODS

The preferences of a sample of 139 Spanish informal caregivers of dependent patients are compared with those obtained via a sample of 312 persons, also from the Spanish general population. We assess 24 dependency states extracted from the DEP-6D using the time trade-off method. Descriptive statistics and regression methods are used to explore differences between the two samples.

RESULTS

Mean difference tests establish that, for all but one of the 24 states, there are no significant differences between the samples. The estimated mean values ranged from - 0.64 to 0.60 for the caregiver sample and from - 0.60 to 0.65 for the general population sample, with a correlation of 0.96. On average, the classification of states as better or worse than dead matched in both samples (except for one state). Regression models also show that sample type does not have a significant average impact. After we introduce interaction effects, only the most severe level of two dimensions, cognitive problems and housework, result in significant differences-with the caregiver sample reporting higher values for the former, and lower values for the latter.

CONCLUSION

Caregivers and the general population exhibit quite similar preferences concerning dependency-related health states. This suggests that the results of cost-utility analyses, and the resource allocation decisions based on them, would likewise not be significantly affected by the preferences used to generate the weighting algorithm.

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

OBJECTIVES

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

METHODS

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

RESULTS

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

CONCLUSIONS

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

Mapping analysis to predict SF-6D utilities from health outcomes in people with focal epilepsy

BACKGROUND

Focal-onset seizures (FOS) are commonly experienced by people with epilepsy and have a significant impact on quality of life (QoL). This study aimed to develop a mapping algorithm to predict SF-6D values in adults with FOS for use in economic evaluations of a new treatment, cenobamate.

METHODS

An online survey, including questions on disease history, SF-36, and an epilepsy-specific measure (QOLIE-31-P) was administered to people with FOS in the UK, France, Italy, Germany, and Spain. A range of regression models were fitted to SF-6D scores including direct and response mapping approaches.

RESULTS

361 individuals were included in the analysis. In the previous 28 days, the mean number of FOS experienced was 3, (range 0-43) and the mean longest period of consecutive days without experiencing a seizure was 14 days (range 1-28 days or more). Mean responses on all SF-36 dimensions were lower than general population norms. Mean SF-6D and QOLIE-31-P scores were 0.584 and 45.72, respectively. The best performing model was the ordinary least squares (OLS), with root mean squared error and mean absolute error values of 0.0977 and 0.0742, respectively. Explanatory variables which best predicted SF-6D included seizure frequency, severity, freedom, and age.

CONCLUSION

People with uncontrolled FOS have poor QoL. The mapping algorithm enables the prediction of SF-6D values from clinical outcomes in people with FOS. It can be applied to outcome data from clinical trials to facilitate cost-utility analysis.

A Program for the Comprehensive Cognitive Training of Excess Weight (TRAINEP): The Study Protocol for A Randomized, Controlled Trial

Background: The available treatments for people with excess weight have shown small effects. Cognitive training has shown promising results, but most of the research focused on normal-weight university students and reported immediate results after a single training session. This parallel group, randomized, controlled trial aims to study the efficacy of a program for the comprehensive cognitive treatment of excess weight. Methods and Analysis: Participants will be 150 people with excess weight recruited through social media, who will be randomized into three groups: cognitive intervention, sham cognitive intervention, and treatment as usual. All assessment and intervention sessions will be online in groups of 5–6 participants. The three groups will attend a motivational interviewing session, and they will receive individualized diet and physical exercise guidelines throughout the program. The cognitive training will consist of four weekly sessions of approximately 60–90 min, each based on approach–avoidance bias training, inhibitory control training, implementation of intentions, and episodic future thinking, respectively. The main outcome measure will be a change in Body Mass Index (kg/m2). Secondary outcomes include changes in cognitive measures, eating and physical exercise behaviors, and anthropometric measures. Assessments will be conducted up to 6 months after the end of the program. In addition, data on the use of the health system will be collected to analyze the cost-effectiveness and the cost-utility of training. Linear mixed models will be used for statistical analysis. Findings of this study will expand the available evidence on cognitive interventions to reduce excess weight.

Willingness to pay vs lottery equivalent to value the impact of alcohol misuse on quality of life

OBJECTIVES

To estimate the impact of alcohol disorders using lottery equivalent (LE) and willingness to pay (WTP) methods and compute the WTP for a QALY (WTP-Q) derived from these values.

METHODS

Two samples of 300 people valued nine states of alcohol misuse. LE and WTP were used in sample 1 and 2, respectively. The ability to discriminate between methods was tested. Regression models were performed to estimate the preference weights of dimensions. Several values of WTP-Q were obtained by combining the estimated values from both samples.

RESULTS

LE and WTP produce the same ranking of states but LE is more sensitive. The estimated impact of the nine states ranges between 0.91 and 0.22 QALYs, and the WTP for avoiding them ranges between €10,444 and €4132. WTP-Q varies between €11,473 and €19,092 when the mean values of the states are used. The WTP-Q tends to decrease with the severity.

CONCLUSIONS

Although LE and WTP provide values for cost-utility and cost-benefit analyses, respectively, LE seems to be preferable for measuring the impact of alcohol disorders. As the lower sensitivity of WTP seems to explain a WTP-Q decrease with severity, more research is needed before recommending the use of different WTP-Q values.

Properties of SF-6D when longitudinal data from 16,398 spine surgery procedures is applied to 9 national SF-6D value sets

Background and purpose - There are several national value sets for SF-6D. For studies conducted in countries without a country-specific value set the authors may use a value set from a neighboring or culturally similar county. We evaluated the consequences of using different national value sets in SF-6D index-based outcome analyses.Patients and methods - Patients surgically treated for lumbar spinal stenosis or lumbar disk herniation between 2007 and 2017 were recruited from the national Swedish spine register. 16,398 procedures were eligible for analysis. The SF-6D health states were coded to SF-6D preference indices using value sets for 9 countries. The SF-6D index distributions were then estimated with kernel density estimation. The change in SF-6D index before and after treatment was evaluated with the standardized response mean (SRM).Results - There was a marked variability in mean and shape for the resulting SF-6D index distributions. There were considerable differences in SF-6D index distribution shape before and after treatment using the same value set. The effect sizes of 2-year change (SRM) were in most cases similar when the 9 value sets were applied on pre- and post-treatment data.Interpretation - We found a marked variability in SF-6D index distributions when a single large data set was applied to 9 national SF-6D value sets. Consequently, we recommend that SF-6D index data from studies conducted in countries without country-specific SF-6D value sets is interpreted with caution.

Valuing SF-6Dv2 in Australia Using an International Protocol

BACKGROUND

The SF-6Dv2 is an updated version of the SF-6D, with improved consistency and dimension descriptors, and value sets are required. The aim of this study was to estimate an SF-6Dv2 value set for Australia using an international protocol, while secondary aims were to compare Australian and UK values and to understand heterogeneity.

METHODS

The study design was based on the SF-6Dv2 valuation protocol, which involved the administration of two different discrete choice experiment (DCE) tasks. The first task presented pairs of SF-6Dv2 health states with duration (Design 1), and the second task presented triplets that appended 'immediate death' to the health state pairs (Design 2). Respondents completed 12 choice sets online (10 from Design 1; 2 from Design 2). Value sets were estimated for Design 1 separately and combining Designs 1 and 2 using a variety of logit model specifications. Value set characteristics were compared descriptively with the UK model.

RESULTS

The sample included 3001 Australians. A range of ordered and consistent models were estimated. The models only including Design 1 data resulted in a relatively wider utility range. The value range was reduced by an interaction added when the health state included a dimension with a severe level. The model matched with the UK value set resulted in a larger utility range and consistent ordering of the most important dimensions.

CONCLUSION

The model including the Design 1 data and moderating interaction was chosen for the Australian value set. This allows the SF-6Dv2 to be used in health technology assessment decision making in Australia.

Analysis of SF-6D Health State Utility Scores: Is Beta Regression Appropriate?

Background: Typically, modeling of health-related quality of life data is often troublesome since its distribution is positively or negatively skewed, spikes at zero or one, bounded and heteroscedasticity. Objectives: In the present paper, we aim to investigate whether Bayesian beta regression is appropriate for analyzing the SF-6D health state utility scores and respondent characteristics. Methods: A sample of 126 Lebanese members from the American University of Beirut valued 49 health states defined by the SF-6D using the standard gamble technique. Three different models were fitted for SF-6D via Bayesian Markov chain Monte Carlo (MCMC) simulation methods. These comprised a beta regression, random effects and random effects with covariates. Results from applying the three Bayesian beta regression models were reported and compared based on their predictive ability to previously used linear regression models, using mean prediction error (MPE), root mean squared error (RMSE) and deviance information criterion (DIC). Results: For the three different approaches, the beta regression model was found to perform better than the normal regression model under all criteria used. The beta regression with random effects model performs best, with MPE (0.084), RMSE (0.058) and DIC (−1621). Compared to the traditionally linear regression model, the beta regression provided better predictions of observed values in the entire learning sample and in an out-of-sample validation. Conclusions: Beta regression provides a flexible approach to modeling health state values. It also accounted for the boundedness and heteroscedasticity of the SF-6D index scores. Further research is encouraged.

Measuring intangible cost-of-morbidity due to substance dependence: implications of using alternative preference-based instruments

OBJECTIVE

Drug and/or alcohol dependence (DAD) generates substantial costs to society. One of the main consequences of DAD is its negative impact on health-related quality of life (HRQoL). The main objective of this study is to analyse the impact of using EQ-5D-5L, SF-6D_SG (SF-6D using standard-gamble as the preference-eliciting method) and SF-6D_PG (SF-6D using a paired-gamble method), to estimate the HRQoL burden, attributable to DAD, within the cost-of-illness framework.

METHODS

A convenience sample of 109 patients with a diagnosis of substance use disorder was recruited. SF-6D and EQ-5D-5L were administered and then the utility scores were computed. The impact of employing different instruments to estimate the HRQoL burden was assessed by comparing the utility scores of patients and general population after controlling for sex and age through regression analysis. The analysis was reproduced for two subgroups of severity.

RESULTS

All instruments detect that DAD significantly affects the HRQoL. However, the estimated impact changes, according to the instrument used, whose pattern varies by severity group. Nonetheless, regardless of severity, SF-6D_PG always estimates a higher or equal DAD burden than the other instruments considered. These results are compatible with the presence of the floor effect in SF-6D_SG, the ceiling effect in EQ-ED-5L, and a smaller presence of both biases in SF-6D_PG.

CONCLUSIONS

The SF-6D_PG instrument emerges as a good candidate to avoid under-estimating intangible costs within the cost-of-illness framework. However, further research is needed to assess the validity of our results in the context of other health problems.

Measuring health-related quality of life in patients treated for substance dependence: differences among instruments and methods of eliciting preferences

BACKGROUND

Measuring the health-related quality of life is an essential estimation in cost-utility studies. In this research, we provide new evidence about comparing utility scores - in the field of substance dependence. Although the main objective is to compare the EQ-5D-5L and SF-6D with paired gamble, evidence about the SF-6D with standard gamble is also provided.

METHODS

Ninety-four patients with substance dependence were recruited; the SF-6D and the EQ-5D-5L were administered at the beginning of treatment and 6 months thereafter. Differences in treatment effect were estimated by comparing utility gains. All analyses were reproduced for two subgroups of severity.

RESULTS

Both the baseline scores and the treatment effect are sensitive to the instrument used. For severe states, the SF-6D with paired gamble (SF-6D with standard gamble) estimates the lowest (highest) utility. With regard to the impact of treatment, the EQ-5D-5L and SF-6D with paired gamble estimate strongly similar effects for severe states (and both estimate greater effects than does the SF-6D with standard gamble).

CONCLUSIONS

These findings have implications for cost-utility analyses. The incremental cost-utility ratio of treatments intended for severe states is barely sensitive to the choice of EQ-5D-5L or SF-6D with paired gamble.

Health utility estimation in children and adolescents: a review of health technology assessments

Objective: Health utility estimates for children and adolescents are critical for cost-utility analyses informing health technology assessment (HTA) authorities' decisions governing access to pediatric treatments. However, in a recent review, only 29% of published pediatric cost-utility models used a utility measure validated for children. We examined utility estimates used in pediatric HTAs.Methods: A targeted review of pediatric HTAs was performed, focusing on agencies reporting utility estimate sources and methods.Results: Searches identified 11 HTAs in pediatric indications and five in mixed populations with separate analyses for adults and children. Among 13 appraisals reporting methodological detail, five used pediatric utility estimates (based on the Health Utilities Index [HUI], n = 3; Atopic Dermatitis Quality of Life [ADQoL], n = 1; or mapping, n = 1). Issues were identified with mapping, use of adult data for some health states, and assumptions about ADQoL responses. In the remaining eight appraisals, adult utility estimates were applied. Caregiver utility was included in two of 16 appraisals.Conclusions: Only 38% of pediatric HTAs reviewed used pediatric utility estimates, and HTA authorities raised concerns about these data in many cases; only 12% of HTAs included caregiver utility. Although several preference-based utility measures are available for pediatric populations, limited data and guidance on selection of measures are available. When estimating pediatric utility weights, alternative measures should be reviewed for suitability given the model population and health condition. Pediatric and adult utility estimates should be applied appropriately as patients age over time, and caregiver and/or family member utility should be included, where relevant. Gaps exist in utility measures for children aged <4 years and caregivers.

Interchangeability of the EQ-5D and the SF-6D, and comparison of their psychometric properties in a spinal postoperative Spanish population

OBJECTIVE

EuroQol-5D (EQ-5D) and Short-Form Six-Dimensions (SF-6D) are widely used to calculate quality-adjusted life-years in cost-utility analysis. The choice of the instrument could influence the results of cost-utility analysis. Our objective was to compare the psychometric properties of the EQ-5D and SF-6D in a postoperative Spanish population, as well as assess their interchangeability in a cost-utility analysis.

DESIGN

Ambispective study.

SETTING

Tertiary public hospital.

PARTICIPANTS

275 Spanish patients who had undergone surgery for lumbar disc herniation.

INTERVENTION(S)

Patients completed EQ-5D-3L and Short-Form 36 (SF-36v2) questionnaires. Internal consistency, floor and ceiling effects, agreement, and construct validity (convergent validity, including dimension-to-dimension correlations, and "known groups" validity) were assessed. The Spanish tariffs were applied.

MAIN OUTCOME MEASURE(S)

Cronbach's α coefficient, Spearman's rank correlation coefficient, Lin's concordance correlation coefficient, intraclass correlation coefficient and Bland-Altman plot.

RESULTS

Main findings were: (a) lack of agreement between EQ-5D and SF-6D utilities (Lin's concordance correlation coefficient: 0.664 [95% CI: 0.600-0.720]; the Bland-Altman plot showed a mean difference of 0.0835 and wide limits of agreement [- 0.2602-0.4272]). (b) Lack of correlation between domains that theoretically measure similar aspects of quality of life, with the exception of "pain" domain.

CONCLUSIONS

The preference-based EQ-5D and SF-6D instruments showed valid psychometric properties to assess generic outcome in a Spanish population who had undergone surgery for lumbar disc herniation; however, utility scores derived from the measures were different. Thus, these two instruments cannot be used interchangeably to perform a cost-utility analysis, and they should both be included in sensitivity analyses.

A hybrid modelling approach for eliciting health state preferences: the Portuguese EQ-5D-5L value set

BACKGROUND

The EQ-5D is a generic preference-based quality of life measure considered useful for supporting clinical and policy decisions by providing utility values that can easily be converted into quality-adjusted life years to be integrated in cost-utility economic evaluations. Although the three-level classification system of the EuroQol questionnaire (EQ-5D-3L) is still the most popular preference-based instrument used worldwide, several studies reported a ceiling effect on this version, especially in healthy and/or young individuals. In 2009, the EuroQol Group introduced a five-level EQ-5D, which expands the descriptive system from three to five levels within the same five dimensions. For this version to be used in health economic evaluation, societal values need to be assigned to the 3125 health states generated by this instrument.

OBJECTIVES

The aims of this study were to elicit the EQ-5D-5L health state preferences from the general Portuguese population and to derive the Portuguese value set for the EQ-5D-5L.

METHODS

A representative sample of the Portuguese general population aged above 18 years was stratified by age and gender (n = 1451). Between October 2015 and July 2016, 28 interviewers carried out a series of 1-h-long computer-assisted personal interviews following the EuroQol Valuation Technology protocol. Each interview included the valuation of ten health states using the composite time trade-off (cTTO) and seven pairs of discrete choice experiments (DCEs). A standardized tool for quality control was used to assess the quality of the data as well as direct supervision and cross-examination of 10% of the global sample size. Data from both cTTO and DCE valuation tasks were modelled using a censored heteroskedastic hybrid model.

RESULTS

Interviewers complied with the quality control protocol in providing high-quality valuation data. The hybrid econometric model had consistent and significant parameters. The derived societal values for the Portuguese population ranged from - 0.603 to 1.

CONCLUSION

This study provided the Portuguese value set for the EQ-5D-5L on the basis of a hybrid econometric model using cTTO and DCE data. These results represent the preferences of the Portuguese population and are recommended to inform health decision-making in Portugal.

An Updated Systematic Review of Studies Mapping (or Cross-Walking) Measures of Health-Related Quality of Life to Generic Preference-Based Measures to Generate Utility Values

BACKGROUND

Mapping is an increasingly common method used to predict instrument-specific preference-based health-state utility values (HSUVs) from data obtained from another health-related quality of life (HRQoL) measure. There have been several methodological developments in this area since a previous review up to 2007.

OBJECTIVE

To provide an updated review of all mapping studies that map from HRQoL measures to target generic preference-based measures (EQ-5D measures, SF-6D, HUI measures, QWB, AQoL measures, 15D/16D/17D, CHU-9D) published from January 2007 to October 2018.

DATA SOURCES

A systematic review of English language articles using a variety of approaches: searching electronic and utilities databases, citation searching, targeted journal and website searches.

STUDY SELECTION

Full papers of studies that mapped from one health measure to a target preference-based measure using formal statistical regression techniques.

DATA EXTRACTION

Undertaken by four authors using predefined data fields including measures, data used, econometric models and assessment of predictive ability.

RESULTS

There were 180 papers with 233 mapping functions in total. Mapping functions were generated to obtain EQ-5D-3L/EQ-5D-5L-EQ-5D-Y (n = 147), SF-6D (n = 45), AQoL-4D/AQoL-8D (n = 12), HUI2/HUI3 (n = 13), 15D (n = 8) CHU-9D (n = 4) and QWB-SA (n = 4) HSUVs. A large number of different regression methods were used with ordinary least squares (OLS) still being the most common approach (used ≥ 75% times within each preference-based measure). The majority of studies assessed the predictive ability of the mapping functions using mean absolute or root mean squared errors (n = 192, 82%), but this was lower when considering errors across different categories of severity (n = 92, 39%) and plots of predictions (n = 120, 52%).

CONCLUSIONS

The last 10 years has seen a substantial increase in the number of mapping studies and some evidence of advancement in methods with consideration of models beyond OLS and greater reporting of predictive ability of mapping functions.

Eliciting Health State Utilities Using Paired-Gamble Methods: The Role of the Starting Point

BACKGROUND

Paired-gamble methods have been proposed to avoid the "certainty effect" associated with standard gamble methods.

OBJECTIVE

This study examines the role of starting-point effects in paired-gamble methods. In particular, it examines how the utilities so derived vary as a function of the probabilities of the stimulus lottery.

METHODS

A sample of 455 members of the Spanish general population valued 9 health states via face-to-face interviews. Subjects were randomly placed into 3 subgroups, which differed in terms of the stimulus gamble's probability. Nonparametric tests and an interval regression model were used to test if utilities change when the probability distribution is modified.

RESULTS

Nonparametric tests showed that the probability of a health state being considered worse than death did not differ among subgroups. Nevertheless, changes in the stimulus gamble did produce significant differences in the distribution of utilities: the higher the probability of full health in the stimulus, the higher the utility elicited. Regression estimates support the existence of starting-point effects when the utilities are obtained under expected utility. According to the prospect theory, the conclusions depend on the reference point considered. When the reference points used are death or the health state evaluated, we observe differences among these groups. Nevertheless, when full health is used, these differences disappear.

CONCLUSION

This research suggests that paired-gamble methods may also be susceptible to starting-point effects. Yet the differences are small, and they disappear when the data are analyzed using prospect theory with full health as the reference point.

Health utilities in cancer patients: A study protocol for a prospective, longitudinal cohort using online survey

BACKGROUND

Cost-utility analysis (CUA) is becoming more commonly used in healthcare decision-making. CUA uses the quality-adjusted life-years (QALY) metric, which combines the length of life with the health-related quality of life (HRQoL). Most QALY-measuring instruments were validated for general populations. For patients with cancer, the perception of their health state is different and may vary by the type of cancer considered. In Quebec, no preference weights for QALY have been developed, neither for the general population nor particular subpopulations.

METHODS/DESIGN

This survey is a prospective, longitudinal cohort study. The study objectives are: to assess the extent of difference in health utilities between the general population and patients with breast or colorectal cancer; to develop a QALY preference weights dataset for patients with cancer; and to perform "mapping" with different HRQoL questionnaires by correlating the SF-6Dv2 with the EQ-5D-5L, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire C30, and functional assessment of cancer therapy - general questionnaires. Data will be collected via a self-administered online survey. Patients' health utilities will be measured within 2 days before the beginning of a chemotherapy treatment cycle and about 8 days after the start of the chemotherapy. Health utilities will be measured by a hybrid method using the time-trade-off and discrete choice experiment methods.

ETHICS AND DISSEMINATION

The proposed research was reviewed and approved by the Institutional Research Ethics Review Boards of the CHUS. We will disseminate our study findings through peer-reviewed publications and conference presentations.

Measuring the impact of alcohol-related disorders on quality of life through general population preferences

OBJECTIVE

To estimate the intangible effects of alcohol misuse on the drinker's quality of life, based on general population preferences METHODS: The most important effects (dimensions) were identified by means of two focus groups conducted with patients and specialists. The levels of these dimensions were combined to yield different scenarios. A sample of 300 people taken from the general Spanish population evaluated a subset of these scenarios, selected by using a fractional factorial design. We used the probability lottery equivalent method to derive the utility score for the evaluated scenarios, and the random-effects regression model to estimate the relative importance of each dimension and to derive the utility score for the rest of scenarios not directly evaluated.

RESULTS

Four main dimensions were identified (family, physical health, psychological health and social) and divided into three levels of intensity. We found a wide variation in the utilities associated with the scenarios directly evaluated (ranging from 0.09 to 0.78). The dimensions with the greatest relative importance were physical health (36.4%) and family consequences (31.3%), followed by psychological (20.5%) and social consequences (11.8%).

CONCLUSIONS

Our findings confirm the benefits of adopting a heterogeneous approach to measure the effects of alcohol misuse. The estimated utilities could have both clinical and economic applications.

The DEP-6D, a new preference-based measure to assess health states of dependency

In medical literature there are numerous multidimensional scales to measure health states for dependence in activities of daily living. However, these scales are not preference-based and are not able to yield QALYs. On the contrary, the generic preference-based measures are not sensitive enough to measure changes in dependence states. The objective of this paper is to propose a new dependency health state classification system, called DEP-6D, and to estimate its value set in such a way that it can be used in QALY calculations. DEP-6D states are described as a combination of 6 attributes (eat, incontinence, personal care, mobility, housework and cognition problems), with 3-4 levels each. A sample of 312 Spanish citizens was surveyed in 2011 to estimate the DEP-6D preference-scoring algorithm. Each respondent valued six out of the 24 states using time trade-off questions. After excluding those respondents who made two or more inconsistencies (6% out of the sample), each state was valued between 66 and 77 times. The responses present a high internal and external consistency. A random effect model accounting for main effects was the preferred model to estimate the scoring algorithm. The DEP-6D describes, in general, more severe problems than those usually described by means of generic preference-based measures. The minimum score predicted by the DEP-6D algorithm is -0.84, which is considerably lower than the minimum value predicted by the EQ-5D and SF-6D algorithms. The DEP-6D value set is based on community preferences. Therefore it is consistent with the so-called 'societal perspective'. Moreover, DEP-6D preference weights can be used in QALY calculations and cost-utility analysis.

[Quality of life of patients with asthma on beclomethasone/formoterol. Cost-utility analysis]

AIM

To perform a cost-utility analysis on asthmatic patients on beclomethasone/formoterol fixed combination in Primary Health Care. Material and methods Non-probability sampling was used to select a group of asthmatic patients with moderate/severe persistent severity (GEMA 2009), treated with beclomethasone/formoterol fixed combination, over 18 years, had given their informed consent. The study observation period was 6 months. The variables studied were: age, sex, duration of disease, health resources used, analysis of health related quality of life by EQ-5D and SF-36, and the specific Asthma Quality of Life Questionnaire. For the qualitative variables, the frequency and percentages were calculated, and for the quantitative variables, the mean, SD and 95% CI. Chi-square, Student t-test and ANOVA were used for statistical inference. Comparisons were made with a statistical significance of 0.05.

RESULTS

Of the 64 patients that completed the study, 59.4% were female. The mean age was 49 years, and mean disease duration was 93 months. For asthma control, 53% of patients had a prescription pattern of one/12h. All health related quality of life scales were modified with respect to the baseline and the differences were statistically significant. Our patients had a better health related quality of life than Spanish asthma cohort. The incremental cost utility beclomethasone/formoterol versus usual treatment option was € 6,256/QALY.

Using the SF-6D to measure the impact of alcohol dependence on health-related quality of life

Alcohol dependence not only reduces life expectancy, but also causes considerable loss of quality of life of the dependents of and persons around those with alcohol dependence. This article presents new evidence on the impact of alcohol dependence on health-related quality of life in Spain. Three samples were recruited: 150 alcoholics and 64 family members of alcoholics, with both samples taken from an alcoholism treatment unit, and 600 persons from the general population. We used the short form 6D, a preference-based generic instrument, applying the utility scores estimated for Spain. It was found that the annual mean loss of quality-adjusted life years associated with alcohol dependence was 0.144 and 0.083 for the alcoholics and their close family members, respectively. This impact becomes more notable after controlling for socio economic variables and was higher than that estimated in similar studies. Possible explanations for these differences are discussed. The results from this work can be applied to economic evaluation studies measuring benefits from policies targeted at reducing the prevalence of alcohol dependence.

SF-6D Portuguese population norms

BACKGROUND

Population-based norms are important because they provide benchmarks that allow the interpretation of subgroups of population health, comparisons between studies, information on health inequalities and support for health-care interventions. The SF-6D is a preference-based measure of health, the use of which has been increasing throughout the world.

OBJECTIVE

Other countries have provided population norms for the SF-6D. To date, SF-6D population norms for Portugal do not exist. This study therefore aimed to derive SF-6D Portuguese population norms.

METHODS

A representative sample of the Portuguese general public (n = 1,500) was used in a telephone-based interview. The SF-36v2 was applied and the Portuguese SF-6D value set was used to obtain the SF-6D index. Descriptive and inferential data analyses were applied to sociodemographic variables, the dimensions of the SF-6D and the SF-6D index. Health index scores of the target population and selected subpopulations were estimated using domain estimation techniques when necessary.

RESULTS

SF-6D utility scores decline with increasing age. Women, widowed individuals and individuals with low educational attainment report lower levels of SF-6D utility scores than men, those with a high educational attainment and those who are single.

CONCLUSIONS

This study reports Portuguese population norms measured using the SF-6D based on a national representative sample. These results can be used as reference data to inform policy.

Inverse probability weighted estimation of social tariffs: an illustration using the SF-6D value sets

This paper presents a novel approach to model health state valuations using inverse probability weighting techniques. Our approach makes no assumption on the distribution of health state values, accommodates covariates in a flexible way, eschews parametric assumptions on the relationship between the outcome and the covariates, allows for an undetermined amount of heterogeneity in the estimates and it formally tests and corrects for sample selection biases. The proposed model is semi-parametrically estimated and it is illustrated with health state valuation data collected for Spain using the SF-6D descriptive system. Estimation results indicate that the standard regression model underestimates the utility loss that the Spanish general population assigns to departures from full health, particularly so for severe departures.