From clinically relevant outcome measures to quality of life in epilepsy

Objectives
Utilities can be easily derived using generic quality of life (QoL) instruments. However, problems in collecting utility scores may occur because clinical evaluations still favor disease specific instruments over generic quality of life instruments for reasons of sensitivity and reliability. Especially in case of the fluctuating nature of seizures in epilepsy, generic QoL-instruments are often find to be unsuitable for outcome research. A proposed method to bridge the gap between clinically relevant outcome measures and QoL is to derive utility scores for epilepsy health states. The aim of this study is to develop a scoring algorithm to transform epilepsy health states into utility scores.
Methods
The proposed scoring algorithm was based on valuations of health states generated by the Time Trade-Off (TTO) method. The TTO was based on clinically important attributes (seizure frequency, seizure severity and treatment related side-effects). A full factorial design was used which resulted in 78 scenario’s. Besides standard demographics, every participant was asked to value 10 or 11 different health states. A multilevel regression analysis was performed to account for the nested structure of the data. The TTO was conducted using online survey software.
Results
In total 531 subjects of the general population, with an average age of 42 years, have participated in the TTO study. Preliminary results show that the best health state (no seizures and no side-effects) is estimated at 0.89 utility and the worst state (seizures twice a day, many side-effects, type 5 seizure) is estimated at 0.16.
Conclusions
This study provides a scoring algorithm for transforming clinically relevant outcome measures of epilepsy into utility estimates which can be incorporated into economic evaluations. Although seizure frequency is the most commonly reported primary outcome measure in epilepsy research, this study suggest that the impact of seizure severity alone should not be underestimated.