Related Experiment Video
Updated: Jan 17, 2026
Computerized Adaptive Testing System of Functional Assessment of Stroke
Published on: January 7, 2019
Psychometric Properties and Feasibility of PROMIS Computerized Adaptive Tests Compared with Disease-Specific Measures
Olivier Dhollander1,2,3, Leo D Roorda4, Seydou Diarra2,3
1Department of Orthopaedic Surgery and Traumatology, Universitair Ziekenhuis Brussel (UZ Brussel), Brussels, Belgium.
Background:
The efficient assessment of health outcomes in knee arthroplasty may benefit from universally applicable Patient-Reported Outcomes Measurement Information System computerized adaptive tests (PROMIS CATs), rather than disease-specific measures. This study aimed to evaluate and compare some psychometric properties and the feasibility of various PROMIS CATs (Pain Interference [PROMIS-PI-CAT, v1.1], Physical Function [PROMIS-PF-CAT, v2.0], Mobility [PROMIS-Mob-CAT, v2.0], Ability to Participate in Social Roles and Activities [PROMIS-AS-CAT, v2.0], and Satisfaction with Social Roles and Activities [PROMIS-SS-CAT, v2.0]), with the Knee Injury and Osteoarthritis Outcome Score (KOOS) scales, including the KOOS Physical Function Shortform [KOOS-PS] and KOOS for Joint Replacement [KOOS-JR], and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scales.
Methods:
Patients (n = 193; mean age [and standard deviation], 64.4 ± 10.1 years; 56% female; mean body mass index, 29.6 ± 5.2 kg/m 2 ) undergoing unilateral or bilateral primary or revision knee arthroplasty at AZ Alma (Eeklo, Belgium) completed the measures 6 weeks before and 6 weeks and 3, 6, and 12 months after surgery. The study evaluated precision (standard error as a percentage of scale range [SE%]), responsiveness (hypothesized correlations and standardized response mean [SRM]), floor and ceiling effects (percentage with the worst and the best scores), and feasibility (completion time and number of items).
Results:
The PROMIS-PI-CAT and PROMIS-PF-CAT showed better precision at baseline compared with corresponding KOOS/WOMAC scales (SE%, 4.6 versus 7.1/9.3 and 3.6 versus 4.4/4.4), but less precision at 12 months of follow-up (SE%, 6.8 versus 4.8/5.5 and 3.6 versus 3.0/3.0). All PROMIS CATs had good responsiveness (75% to 100% of hypotheses not rejected; SRMs at 12 months: PROMIS-PI-CAT = -1.35 versus KOOS Pain = 1.78 and WOMAC Pain = -1.59; PROMIS-PF-CAT = 1.14 versus KOOS-ADL/WOMAC-PF = 1.43/-1.44; PROMIS-AS-CAT = 0.93 and PROMIS-SS-CAT = 0.93). The PROMIS-PF-CAT did not show ceiling effects at 12 months, unlike the KOOS-ADL/WOMAC-PF (17.5%). PROMIS CATs were more feasible at baseline and follow-ups compared with KOOS and WOMAC scales.
Conclusions:
PROMIS-CATs effectively assess health outcomes in knee arthroplasty patients, showing strong psychometric properties and favorable feasibility, supporting their role in value-based health care.
Level Of Evidence:
Prognostic Level II . See Instructions for Authors for a complete description of levels of evidence.
