Translation and validation of a Norwegian version of the prosthetic limb users survey of mobility and assessment of self-reported mobility of lower limb prosthetic users in Norway.

Reed-Schwanborg, Linn; Starholm, Inger Marie; Solberg, Mari Bergelien; Langseth, Ingrid Iversen; Gjøvaag, Terje · Prosthet Orthot Int · 2025

cross_sectional · Level IV

Where this comes from

Abstract

The aim of the current study is to translate the original prosthetic limb users survey of mobility (PLUS-M) instrument to Norwegian, investigate its psychometric properties, and conduct a survey on mobility in Norwegian lower limb prosthetic users (LLPU). The Functional Assessment of Chronic Illness Therapy methodology was followed for translation and cross-cultural validation. After translation, invitations to participate in a survey was sent from prosthetic and orthotic clinics in Norway to registered LLPU. Of 1279 invitations, 454 people with unilateral lower limb amputation (age, 62.6 ± 14.4 years) were included in the study. Known-groups construct validity was investigated by comparing the T-scores of men vs. women, transtibial vs. transfemoral amputation, vascular vs. nonvascular etiology, and younger vs. older persons. The overall PLUS-M T-score (mean ± SD) was 53.2 ± 11.1. Men (n = 318) had better mobility than women (n = 137), with T-scores of 54.7 ± 10.4 and 49.6 ± 12.4, respectively (P < 0.0005). All hypotheses about assumed differences in T-scores between known-groups were confirmed (all comparisons; P < 0.0005). Internal consistency (Cronbach α, 0.962) and test-retest reliability (intraclass correlation coefficient 0.936, 95% confidence interval, 0.871-0.968) were excellent. Standard error of measurement was 2.02, and minimal detectable change (95% CI) was 5.59. Furthermore, floor and ceiling effect was 1.8% and 10.9%, respectively. The Norwegian version of the PLUS-M 12-item short form is valid and has excellent reproducibility and psychometric properties. The overall T-score for the Norwegian LLPU is marginally higher compared to the mean ± SD T-score (50 ± 10) of the original development sample (N = 1091).