Lower extremity prosthetic mobility: A comparison of 3 self-report scales

Objective: To assess and compare the reliability and validity of the Hought on Scale, the Prosthetic Profile of the Amputee Locomotor Capabilities Inde x (PPA-LCI), and the Prosthetic Evaluation Questionnaire (PEQ) mobility sub scale, 3 disease-specific self-report measures of functional mobility for l ower extremity prosthetic mobility. Design: Four-week test-retest: 1 sample for reliability analyses, 1 sample for validity analyses. Setting: University-affiliated outpatient amputee clinic, in Ontario, Canad a. Participants: Two outpatient amputee samples (sample 1 [n = 55], for reliab ility analysis; sample 2 [n = 329], for validity analysis). Interventions: Not applicable. Main Outcome Measures: Test-retest of reliability and convergent validity o f the 3 scales. Convergent validity and discriminative ability were also as sessed after setting a priori hypotheses for 2 scales of walking performanc e, balance confidence, and other indicators of ambulatory ability. Results: The reliability of the PPA-LCI (intraclass correlation coefficient [ICC] = .88) was slightly higher than the Houghton Scale (ICC = .85) and t he PEQ mobility subscale (ICC = .77). The PPA-LCI was prone to high ceiling effects (40%) that would limit its ability to detect improvement. Evidence for convergent validity, when compared with the 2-Minute Walk Test, Timed Up and Go, and the Activity-Specific Balance Confidence Scale, was supporte d as hypothesized in all the scales. Each of the scales was able to discrim inate between different groups for amputation cause, walking distance, mobi lity device use, and automatism, with each having varying strength related to relative precision. The Houghton Scale was the only scale able to distin guish between amputation levels. Conclusions: Reliability and validity of all the scales are acceptable for group level comparison. None of the scales had clearly superior psychometri c properties compared with the others. Further research is required to asse ss responsiveness.