Comparison between the QOLIE-31 and derived QOLIE-10 in a clinical trial of levetiracetam

Purpose: to determine whether the QOLIE-10, an abbreviated quality of life questionnaire, provides results similar to the more detailed QOLIE-31 instrument when the ten items are derived from the QOLIE-31. Methods: the QOLIE-31 was completed by 246 patients participating in UCB protocol N132 at baseline and after 18 weeks of treatment with levetiracetam (LEV 1000 or 3000 mg) or placebo added to standard therapy. QOLIE-10 components and total scores were calculated from the QOLIE-31 data. Results: baseline QOLIE-10 components and total score correlated highly with corresponding QOLIE-31 scores, both at baseline and follow-up (range 0.70-0.95). Changes from baseline to follow-up were significantly different (ANCOVA) among treatment groups for both the QOLIE-10 and QOLIE-31 for the total score (P = 0.02, P = 0.009, respectively), seizure worry (P = 0.005, P = 0.0003) and cognitive functioning (P = 0.01, P = 0.01). One subscale (overall QOL) showed significant change with the QOLIE-31 (P = 0.04), but not with the QOLIE-10 (P = 0.07). Differences in QOLIE-10 scores were found between responders (greater than or equal to 50% partial onset seizure reduction) and non-responders for the total score (P = 0.0001) and two components (overall QOL P = 0.002, social function P = 0.0003). In the QOLIE-31, the total score and six subscale scores tall except medication effects) were significantly different. Both instruments were able to detect change over time. Responsiveness assessed by effect sizes (- 0.1 for non-responders, 0.4 for responders, 0.8 for seizure-free patients) and the Guyatt statistic (- 0.1, 0.6 and 1.0, respectively) was similar for both instruments. Conclusions: although the QOLIE-10 was designed as a screening tool, it can be scored and used in research. The total score did discern differences among treatments in a clinical trial. Nonetheless, questionnaires with multiple, multi-item subscales provide more detailed information than abbreviated forms. The QOLIE-31 is preferred where time and resources are available. (C) 2000 Elsevier Science B.V. All rights reserved.