A major advantage of using a rating scale in health-utility measurement is its practical applicability: the method is relatively easy to understand, and various health states can be assessed simultaneously. However, a theoretical foundation for rating-scale valuations has not been established. The primary aim of this paper is to present a theoretical foundation for rating-scale valuations based on the theory of measurable value functions and to provide a consistency test to see whether rating-scale valuations do indeed elicit a measurable value function. If rating-scale valuations elicit a measurable value function, then Dyer and Sarin have shown how they are related to von Neumann-Morgensterm (vNM) utilities. The appropriate technique to measure vNM utilities is the standard gamble. Torrance has suggested that rating-scale valuations and standard-gamble valuations are related by a power function. A secondary aim of this paper is to examine the relationship between rating-scale valuations and standard-gamble valuations hypothesized by Torrance. An experiment was designed to test consistency of rating-scale valuations and the relationship between rating-scale valuations and standard-gamble valuations. The experiment tested whether rating-scale valuations are independent of the context in which they are elicited, as they should be if they elicit points on a measurable value function. 80 Swedish and 92 Dutch respondents participated in the experiment. The results showed that rating-scale valuations depend on the number of preferred alternatives in the task and thus violate a basic property of measurable value functions. The estimation of the power function did not result in stable results: parameter estimates varied, in some cases there was indication of misspecification, and in most cases there was indication of heteroskedastic errors. The implications of these findings for the common use of rating-scale valuations in cost-utility analysis are serious: the dependency of the rating-scale valuations on the other health states included in the task casts serious doubts on the validity of the rating-scale method.
Comment In: Med Decis Making. 1998 Apr-Jun;18(2):2369566457
This paper gives a detailed presentation of a computer model for evaluating the cost-effectiveness (CE) of hormone replacement therapy (HRT), describing the model's design, structure, and data requirements. The model needs data specified for costs, quality of life, risks, and mortality rates. As an illustration, the CE of HRT in Sweden is calculated. Two treatment strategies are evaluated for asymptomatic women: estrogen-only therapy and estrogen combined with a progestin. The model produces similar results compared with earlier studies. The CE ratios improve with the size of the risk reduction and generally with age. Further, estrogen-only therapy is associated with a lower cost per gained effectiveness unit compared with combined therapy. Uncertainty surrounding the long-term effects of HRT means that the CE estimates should be interpreted carefully. The model permits the inclusion of indirect costs and costs in added life-years, allowing the analysis to be made from a societal perspective, which is an improvement relative to previous studies.
In recent years, there has been a growing interest in the contingent valuation method for measurement of monetary values of various commodities. However, the validity and reliability of the method need to be examined thoroughly. This paper reports results of a test of scope and question order effects in a contingent valuation experiment in the health care field. Using three binary valuation questions, data were collected on willingness to pay for superior treatment of reflux oesophagitis. To test for scope effects, different probabilities of successful short- and long-term treatments were evaluated using a split sample approach. The presence of question order effects was tested by assigning respondents to different question orders. The contingent valuation method proved sensitive to changes in scope in that the willingness to pay increased with the probability of being free from symptoms and with a reduced risk of having a relapse once recovered. Also, regression analysis indicate that people who suffer from severe reflux oesophagitis are more willing to pay for more effective treatment. No question order effects were detected in the data.
An analysis of the cost-effectiveness of simvastatin was conducted, based on the Scandinavian Simvastatin Survival Study (4S). The total cost of hospitalization in the placebo group was 52.8 million Swedish kronor (SEK) (5.15 million pounds), compared with SEK 36.0 million (3.51 million pounds) in the simvastatin group. This amounts to a 32% reduction, or a saving of SEK 16.8 million (1.6 million pounds) or SEK 7560 (738 pounds) per patient. The net cost per patient for the duration of the study (5.4 years) was SEK 13,540 (1324 pounds). Simvastatin treatment saved an estimated 0.377 undiscounted life years (0.240 life years discounted at 5% per annum). The cost of simvastatin therapy per discounted life-year saved was therefore SEK 56,400 (5502 pounds). Sensitivity analysis, examining the effect of different life expectancies, costs of initiation and monitoring of simvastatin therapy, and discount rates, showed the results to be stable. Conclusion. The cost per life-year saved of simvastatin in the treatment of post-myocardial infarction and angina patients, as determined from 4S data, is well within the range normally considered cost-effective.
Comment In: Eur Heart J. 1996 Jul;17(7):974-58809507
It would be a major advance if quality-of-life instruments could be translated into health-state utilities. The aim with this study was to investigate the relationship between the SF-12 and health-state utilities, based on responses to a postal questionnaire sent to a random sample of 8,000 inhabitants, aged 20-84 years, in the general population. The questionnaire included the SF-12, a rating-scale (RS) question, and a time-tradeoff (TTO) question; the response rate was 68%. Age, gender, and the 12 items of the SF-12 were used as explanatory variables in a linear regression analysis of the health-state utilities. The regression models explained about 50% of the variance in the RS answers and about 25% of the variance in the TTO answers. Most of the SF-12 items were related to the health-state utilities in the expected ways, with especially strong results for the RS method. The results suggest that the SF-12 can be converted to health-state utilities, but that further work is needed to reliably estimate the conversion function.
To compare the costs of health care programs, with the benefits, the values of changes in health status must be expressed in monetary terms. The development of methods to estimate willingness to pay for changes in health status is therefore of interest. This paper reports the results of a contingent valuation study measuring willingness to pay for reductions in angina pectoris attacks. An innovative study design allowed analysis on the data on willingness to pay using two approaches, a binary question and a bidding-game technique. Percentage reductions in anginal attacks were varied randomly in different subsamples, and data were collected about angina pectoris status, attack rate, and income to test the internal validity of the contingent valuation method. Willingness to pay for a 50% reduction in the attack rate for three months was estimated to be about SEK 2,500 ($345) with the binary approach, and about SEK 2,100 ($290) using the bidding-game technique. Regression analyses showed that income, angina pectoris status, attack rate, and percentage reduction in attack rate were all related to willingness to pay, in agreement with the authors' hypothesis.