Assessment of a Nine-Item Patient-Reported Outcomes Measurement Information System Upper Extremity Instrument Among Individuals With Upper Limb Amputation.

Abstract

The aim of the study was to assess the validity of a customized nine-item Patient-Reported Outcomes Measurement Information System Upper Extremity instrument being used with individuals with upper extremity amputation to inform potential modifications for clinical efficiency. A sample of 239 adults with upper extremity amputation (mean age = 48 ± 16 yrs; female = 69; prosthesis users = 150) were included. After clinical implementation of the nine-item Patient-Reported Outcomes Measurement Information System Upper Extremity, the following psychometric properties were examined: structural and known-groups validity, differential item functioning, and reliability. The nine-item Patient-Reported Outcomes Measurement Information System Upper Extremity short form demonstrated satisfactory psychometric properties. Convincing evidence of structural validity included no violation of unidimensionality, local independence (all local dependence χ2 < 10), monotonicity (Hij > 0, Hi > 0.3, and H = 0.57), and adequate model fit (P > 0.006). Known-groups analysis demonstrated that the nine-item Patient-Reported Outcomes Measurement Information System Upper Extremity instrument was able to discriminate between prosthesis users and nonprosthesis users and amputation level. The Cronbach's α and item response theory reliability at the selected range of T scores were greater than 0.9 indicating high reliability. No items were flagged for age in differential item functioning. The customized nine-item Patient-Reported Outcomes Measurement Information System Upper Extremity met the minimum criteria, according to the Patient-Reported Outcomes Measurement Information System plan, for psychometric validity supporting its use within the population of individuals with upper extremity amputation.