METHOD A criterion based approach was employed using a series of case studies in the course of a feasibility study on using an internet panel to obtain preference data from members of the general public. The criterion was the results of RCTs of health technologies which showed statistically and clinically significant results on a disease specific quality of life measure. Examples were computerized CBT for depression, hip resurfacing, CBT for insomnia, pulmonary rehabilitation, infliximab for Crohn's disease. Health state descriptions corresponding to key summary values (before and after treatment and between groups) were developed for each example, and valued using the standard gamble via the internet by a panel of members of the public. The incremental utilities between each health technology and comparator were compared using parametric and non-parametric approaches. A value of 0.1 was taken as “clinically significant” difference in utility, following the basis for the sample size calculation undertaken to estimate preferences for the EQ5D in the UK.

RESULTS The mean incremental utility gains associated with each of the technologies ranged from 0.0005 to 0.23 across the five examples. In four cases, statistically significant differences in outcomes demonstrated in the original trial were preserved in the mean values obtained from the panel. In all cases, median utility values were significantly different between groups, although in one case this difference was not clinically significant.

CONCLUSION Precision is an important aspect of the measurement techniques which has not hitherto been considered in the context of utility measurement. Using a simple approach, this study demonstrates that the process of health state description and valuation did not obscure clinically important effects of health technologies in most, but not all cases.