Electronic Quality of Life Assessment Using Computer-Adaptive Testing.

Area of Science:

• Psychometrics
• Health Outcomes Research
• Health Services Research

Background:

• Quality of Life (QoL) questionnaires are valuable in clinical practice but often lengthy, hindering widespread use.
• Traditional QoL assessments can be time-consuming to administer and interpret.

Purpose of the Study:

• To evaluate the performance of the World Health Organization Quality of Life (WHOQOL)-100 questionnaire as item banks for adaptive testing.
• To facilitate computerized adaptive tests (CATs) for physical, psychological, social, and environmental QoL domains.

Main Methods:

• Calibrated 40 WHOQOL-100 items into four domain-specific banks using item response theory with UK data (N=320).
• Conducted psychometric assessments including differential item functioning, local dependency, unidimensionality, and reliability.
• Simulated CATs to determine the number of items needed for reliable assessments.

Main Results:

• Item banks demonstrated good model fit, unidimensionality, reliability (Person Separation Index >.70), and were free from differential item functioning and local dependency.
• CAT-based assessments were 45%-75% shorter than paper-based WHOQOL measures for equivalent reliability.
• High reliability (alpha >.90) was achieved with a median of 9 items across the four domains.

Conclusions:

• Computerized adaptive testing (CAT) with WHOQOL-100 item banks provides reliable QoL assessments using fewer items.
• These item banks are suitable for computerized adaptive assessment and support international development.
• CAT offers a more efficient approach to measuring quality of life in clinical settings.