Carl Tilling

Protocols for Time Tradeoff Valuations of Health States Worse than Dead: A Literature Review

Background. The time tradeoff (TTO) method of preference elicitation allows respondents to value a state as worse than dead, generally either through the Torrance protocol or the Measurement and Valuation of Health (MVH) protocol. Both of these protocols have significant weaknesses: Valuations for states worse than dead (SWD) are elicited through procedures different from those for states better than dead (SBD), and negative values can be extremely negative. Purpose. To provide an account of the different TTO designs for SWD, to identify any alternatives to the MVH and Torrance approaches, and to consider the merits of the approaches identified. Methods. Medline was searched to identify all health state valuation studies employing TTO. The ways in which SWD were handled were recorded. Furthermore, to ensure that there are no unpublished but feasible TTO variants, the authors developed a theoretical framework for identifying all potential variants. Results. The search produced 593 hits, of which 218 were excluded. Of the remaining 375 articles, only 29 included protocols for SWD. Of these, 23 used the MVH protocol and 4 used the Torrance protocol. The other 2 used 1 protocol for SBD and SWD, one making use of lead time and the other using a 2-stage procedure with chaining. The systematic framework did not identify any alternatives to the Torrance and MVH protocols that were superior to the lead time approach. Conclusions. Few studies elicit values for SWD. The lead time approach is a potential alternative to the Torrance and MVH protocols. Key words: QALY; states worse than dead; health state valuation; preference elicitation.

In or out? Income losses in health state valuations: A review

BACKGROUND In 1996 the Washington Panel controversially recommended valuing productivity costs (PC) in terms of quality-adjusted life years. The Panels assumption that respondents in health state valuation (HSV) exercises take income losses into account could not be countered since there was no evidence regarding what people consider in HSV exercises. If they do consider income losses and if this changes HSVs, then all economic evaluations that have included PC in the numerator may have double-counted these costs. Alternatively, if respondents do not consider income losses then all past economic evaluations that have not included PC in the numerator have failed to account for sizeable societal costs. OBJECTIVES Through a review we aim to recapture the debate surrounding the appropriate method for including PC in health economic evaluations, to identify empirical evidence addressing the assumptions of the Panel, and recommend a future research agenda. METHODS Through a review we identify, outline, and critically appraise the existing empirical studies that attempt to address whether respondents include income effects in HSV exercises. RESULTS AND CONCLUSION Seven empirical studies were identified. Overall, it seems that not explicitly mentioning the inclusion of income will induce a minority of respondents to include these effects and this appears not to influence results. More empirical work is needed, using generic instruments, larger samples, and using the interview method of administration.

Does the EQ-5D Reflect Lost Earnings?

BackgroundAn important methodological issue in economic evaluations of healthcare is how to include productivity costs (the costs related to reduced productivity due to illness, disability and premature death). Traditionally, they were included in the numerator of a cost-effectiveness analysis, through either the human-capital or the friction-cost method. It has been argued, however, that productivity costs are already included in the denominator (i.e. in the QALY measure) because respondents consider the effect a given health state will have on their income when valuing health states. If that is the case, many previous economic evaluations might have double counted productivity costs by including them in both the numerator and the denominator.AimThe aim of this study was to determine whether respondents valuing EQ-5D health states using the time trade-off (TTO) method spontaneously consider income effects, whether this consideration influences subsequent valuations and whether explicit ex post instructions influence valuations.MethodsThrough an online survey, we asked 321 members of the Dutch general population to value four EQ-5D health states through three different TTO exercises. The first exercise was a standard TTO question. Respondents were then asked whether they had included income effects. Depending on their answer, the second TTO exercise instructed them to either include or exclude income effects. The third TTO exercise provided explicit information regarding the income loss associated with the health state.ResultsData were available from 321 members of the Dutch general public. Of these respondents, 49% stated they had spontaneously included income effects. Twenty-five percent of the sample did not trade any time in any of the TTO exercises and these respondents were excluded from the analysis. Results of t-tests showed there were only weakly significant differences in valuations for one health state between those who spontaneously included income effects and those who did not. Explicit instruction led to some significant differences at the aggregate level, but the effect was inconsistent at the individual level. When explicit information on the amount of income loss was provided, all states were valued lower when associated with a larger income loss.ConclusionsThis study offers further evidence indicating that income losses do not significantly affect health state valuations.

Exploring a new method for deriving the monetary value of a QALY

Several studies have sought to determine the monetary value of health gains expressed as quality adjusted life years (QALYs) gained, predominantly using willingness to pay approaches. However, willingness to pay has a number of recognized problems, most notably its insensitivity to scope. This paper presents an alternative approach to estimate the monetary value of a QALY, which is based on the time trade-off method. Moreover, it presents the results of an online study conducted in the Netherlands exploring the feasibility of this novel approach. The results seem promising, but also highlight a number of methodological problems with this approach, most notably nontrading and the elicitation of negative values. Additional research is necessary to try to overcome these problems and to determine the potential of this new approach.

PMC15 IN OR OUT? EMPIRICAL EVIDENCE ON INCOME LOSSES IN HEALTH STATE VALUATIONS AND IMPLICATIONS FOR ECONOMIC EVALUATIONS

PMC13 ESTIMATING COST-OF-ILLNESS USING GENERALIZED LINEAR MODELS: AN ALTERNATIVETOTHE SMEARING APPROACH Exuzides A, Colby C, Spalding JR ICON Clinical Research, San Francisco, CA, USA, Astellas Pharma US, Deerfield, IL, USA OBJECTIVES: Estimation of cost-of-illness typically involves the analysis of skewed medical costs that include large outliers. Log transformations are frequently used to overcome these problems. Linear regression models (OLS) are then applied to the transformed data. The estimated model coefficients are retransformed back to the linear scale using the smearing approach. Implementing this approach in statistical packages requires customized programming. We propose an alternative to using log transformations: Generalized Linear Models (GLM) with a log link function. We compare the performance of both models in estimating cost-of-illness. METHODS: We derived data from a large administrative database representing 143,593 discharges from 39 US hospitals from January 2004 to December 2005. We estimated total medical costs among hospitalized patients attributable to hyponatremia. Using a cross-validation approach, we compared the performance of two models: log transformed OLS with smearing and GLM with a log link function and a normal error distribution. We used the Root Mean Squared Error (RMSE) and the Mean Absolute Error (MAE) to assess model performance. Covariates in both models included patient age, gender, race, geographic region, Deyo-Charlson comorbidity index, primary diagnosis, teaching status of hospital, and admission source. All analyses were contacted using SAS. RESULTS: The GLM with log-link and a normal error distribution had both the smallest RMSE (23,688) and MAE (11,304) compared to the log transformed OLS with smearing (24,057 and 11,392, respectively). Furthermore, by using GLM, there was no need to compute a retransformation estimate, since the log link function relates the response mean to the original scale. CONCLUSIONS: In this cross-validation study, GLM outperformed OLS with smearing. GLM is easier to implement using SAS with no need for retransformation estimates. Because of its ease of use and statistical accuracy, GLM is a useful alternative to log-transformed OLS models with smearing, when estimating cost-of-illness.