Outcome assessment in epilepsy: comparative responsiveness of quality of life and psychosocial instruments.

PURPOSE: Few data exist on the ability of instruments to detect within-patient change over time in epilepsy, a property referred to as responsiveness. Our aim was to compare the responsiveness of three instruments [i.e., Epilepsy Surgery Inventory-55 (ESI-55) and Washington Psychosocial Inventory (WPSI), both epilepsy specific, and Symptom Checklist-90-Revised (SCL-90-R), non-epilepsy specific].

METHODS

Instruments were administered at baseline and at 1 year in a prospective cohort of surgically (43) and medically (14) treated patients with temporal lobe epilepsy. Coefficient of Responsiveness and relative efficiency were computed for each scale and for the dimensions of mental health, physical health, and role function.

RESULTS

The ESI-55 contained the most responsive scales, whereas SCL-90-R contained the largest number of scales with moderate responsiveness. The largest number of scales with low responsiveness belonged to the WPSI. Sensitivity to between-treatment differences in change was highest for ESI-55 and SCL-90-R. The most efficient scales in detecting differences between treatment groups in the mental, physical, and role-function dimensions were ESI-55 emotional well-being, ESI-55 health perceptions, and SCL-90-R hostility, respectively.

CONCLUSIONS

Our results support the responsiveness of ESI-55 scales and suggest that SCL-90-R is a responsive tool for the assessment of psychologic function and distress in epilepsy. Comparatively, WPSI is relatively unresponsive to small or medium-size changes.