Cross-cultural adaptation, reliability, internal consistency and validation of the Trinity Amputation and Prosthetic Experience Scales-Revised (TAPES-R) for French speaking patients with lower limb amputation

AimThe aim of this study was the French cross-cultural adaptation and validation of The Trinity Amputation and Prosthetic Experience Scales-Revised (TAPES-R-F), in a lower limb amputation population. This self-reported multidimensional amputee-specific questionnaire [1] evaluates with 33 items psychosocial adjustment (3 subscales), activity restriction (1 subscale) and satisfaction with prosthesis (2 subscales).Patients and methodsOne hundred and twenty-nine patients with a mean age of 62 years and with lower limb amputations for at least 1 year were recruited in 5 clinical centers. Translation and cross-cultural adaptation were made according to international guidelines. Internal consistency of each subscale was measured with Cronbach's alpha. Test-retest reliability was assessed by intraclass correlation in a sub- group of 24 subjects who completed the TAPES-R twice with an interval of 7 days. Construct validity was estimated through correlations with 2 main components of SF-36 (MCS, PCS). Correlations were also calculated with 3 scales of the Brief Pain Inventory (BPI).ResultsCronbach's alpha was high, ranging from 0.85 to 0.95. Reliability was acceptable to high (ICC=0.72 to 0.91) for all subscales with the exception of the Social adjustment (ICC=0.67) and Adjustment to limitation (ICC=0.42) subscales. The 2 component of SF-36 was significantly correlated with all subscales (PCS: r=0.24 to 0.66); MCS: r=0.30 to 0.58), except with aesthetic satisfaction and adjustment to limitation. Regarding the BPI, significant correlations were found for all subscales (r=–0.20 to –0.68) with the exception of adjustment to limitation. Ceiling or floor effects (>15%) were found for all but Activity Restriction and Functional Satisfaction Scales.DiscussionThe TAPES-R-F has acceptable psychometric properties for most of its subscales. Our results may suggest that the French version is more useful in a population research perspective than in an individual perspective. Other studies are necessary to clarify the role and the psychometric properties of this measurement