The Patient‐Reported Outcomes Measurement Information System (PROMIS®) is a universally applicable set of item‐banks for the evaluation of patient‐reported health. Both the PROMIS Pain Interference (PROMIS‐PI) and the PROMIS Pain Behavior (PROMIS‐PB) item‐banks were found to have good psychometric properties for patients with rheumatoid arthritis and to be useful as computerized adaptive tests (CATs) in clinical practice and research, according to a study published in Arthritis Care & Research.
The objective of the current study was to evaluate the psychometric properties of the PROMIS-PI and the PROMIS-PB item banks (40 and 39 items, respectively) in Dutch and Flemish patients with rheumatoid arthritis. Properties examined in those assessments are unidimensionality, cross‐cultural validity (Differential Item Functioning [DIF] for language [Dutch vs Flemish]), other forms of measurement invariance, floor and ceiling effects, monotonicity, Graded Response Model (GRM) fit, local dependence, construct validity, and reliability.
Both the PROMIS‐PI and PROMIS‐PB item-banks were found to have sufficient unidimensionality (Omega‐H 0.99 and 0.95; ECV 0.95 and 0.78; respectively), to have negligible local dependence (0.3% and 1.4% of item‐pairs), good monotonicity (scalability coefficient of the scale, 0.75 and 0.46), and a good graded response model fit. Both item-banks also showed good cross-cultural validity (absence of differential item functioning for language), measurement invariance (absence of differential item functioning for age, sex, disease activity, and administration mode), good construct validity, high reliability (>0.90 in the range of patients with rheumatoid arthritis), and absence of floor and ceiling effects (0% maximum or minimum score for both). The PROMIS-PI correlated strongly with the Dutch-Flemish PROMIS Global Health Pain intensity (r=0.80), the Short-Form Health Survey Physical Functioning scale (r=-0.71) and the Health Assessment Questionnaire Disability Index (r=0.71). The PROMIS-PB also correlated strongly with the Dutch-Flemish PROMIS Global Health Pain intensity 266 (r=0.61). These findings add to the evidence that the PROMIS item-banks provide an adequate assessment of pain interference and behavior, respectively.
“Both the PROMIS-PI and PROMIS-PB banks showed good psychometric properties in patients with [rheumatoid arthritis]. Using the highly efficient PROMIS-PI and PROMIS-PB CATs in research and clinical practice is considered to be user-friendly and feasible with little administration time, and has the potential for valid and precise standardized and routine patient monitoring of pain interference and pain behavior,” concluded the study authors.
Crins MHP, Terwee CB, Westhovens R, et al. First validation of the full PROMIS pain interference and pain behavior item banks in patients with rheumatoid arthritis [published online September 28, 2019]. Arthritis Care Res (Hoboken). doi: 10.1002/acr.24077