Consistency Within and Between Methods of Health Status Valuation: A Within Subjecy Examination of the Willingness to Pay and Standard Gamble Methods

In a priority setting context, the attainment of a desirable allocation of health care resources is dependent upon the accurate estimation of health gains produced by different treatments and services. Therefore, the detection of disparities between methods of obtaining these estimates has serious implications for the allocative process. Using data from a Department of Transport study into the value of preventing non-fatal road injuries, health status values for a set of injuries obtained using the standard gamble (SG) and willingness to pay (WTP) methods are examined. The results show that significant differences arise even when responses from both methods are obtained from the same individuals. Moreover, internal consistency checks built into the SG and WTP sections of the questionnaires reveal inconsistencies within these methods. Systematic inconsistencies, both within and between methods, are identified at the individual and aggregate level, and a model predicting these results is presented. SG and WTP responses are then adjusted - isolating the different components of the model - and the relative injury severities are recalculated accordingly. The recalculated SG and WTP results are then compared to determine whether the disparities within and between methods have been reduced or removed. Finally, suggestions regarding the application of these methods to various ends is discussed in the light of the results.