Franco Franchignoni1, Andrej Bavec2, Urška Zupanc3, Andrea Giordano4, Caterina Albensi5, Helena Burger6. 1. Maugeri Scientific Clinical Institutes, Institute of Care and Scientific Research, Physical Medicine and Rehabilitation Unit, Tradate Institute, Tradate, Italy. 2. University Rehabilitation Institute, Ljubljana, Slovenia; Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia. Electronic address: andrej.bavec@ir-rs.si. 3. University Rehabilitation Institute, Ljubljana, Slovenia. 4. Istituti Clinici Scientifici Maugeri IRCCS, Bioengineering Unit, Veruno Institute, Veruno, Italy. 5. 'Policlinico Tor Vergata' Foundation - U.O.S.D. Stroke Unit, Rome, Italy. 6. University Rehabilitation Institute, Ljubljana, Slovenia; Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia.
Abstract
OBJECTIVE: To examine the psychometric properties of the Activities-specific Balance Confidence (ABC) scale administered in the Slovene version with a simplified 5-option response format (ABC-5/SLO) using Rasch analysis. DESIGN: Methodological research on data gathered in a cross-sectional study. SETTING: Outpatient university rehabilitation clinic. PARTICIPANTS: A convenience sample of adults with unilateral lower-limb amputation (N=138; 75% men) longer than 6 months who regularly wear a prosthesis. INTERVENTION: Not applicable. MAIN OUTCOME MEASURES: We evaluated functioning of rating scale categories, internal construct validity, reliability indices, and dimensionality using the ABC-5/SLO (0=no confidence to 4=complete confidence). RESULTS: The ABC-5/SLO rating scale fulfilled the category functioning criteria. All items fit the underlying scale construct (balance confidence) except item 8 ("walk outside the house to a car parked in the driveway"), which was overfitting. The person abilities-item difficulty matching (targeting) was good. The person separation reliability was .92, and the item separation reliability was .99. Analysis of the standardized Rasch residuals showed the scale's unidimensionality and absence of high item dependency (residual correlations, <.30). The correlation between the ABC-5/SLO and the Prosthetic Mobility Questionnaire (Rasch measures) was high (ρ=.84), as expected. Minor signs of item redundancy were found. CONCLUSIONS: The simplified ABC-5/SLO scale is a valid and reliable measure of balance confidence for individuals with lower-limb amputation. It is possible to transform the ordinal summed raw scores of the ABC-5/SLO into interval-level measurements using a nomogram.
OBJECTIVE: To examine the psychometric properties of the Activities-specific Balance Confidence (ABC) scale administered in the Slovene version with a simplified 5-option response format (ABC-5/SLO) using Rasch analysis. DESIGN: Methodological research on data gathered in a cross-sectional study. SETTING:Outpatient university rehabilitation clinic. PARTICIPANTS: A convenience sample of adults with unilateral lower-limb amputation (N=138; 75% men) longer than 6 months who regularly wear a prosthesis. INTERVENTION: Not applicable. MAIN OUTCOME MEASURES: We evaluated functioning of rating scale categories, internal construct validity, reliability indices, and dimensionality using the ABC-5/SLO (0=no confidence to 4=complete confidence). RESULTS: The ABC-5/SLO rating scale fulfilled the category functioning criteria. All items fit the underlying scale construct (balance confidence) except item 8 ("walk outside the house to a car parked in the driveway"), which was overfitting. The person abilities-item difficulty matching (targeting) was good. The person separation reliability was .92, and the item separation reliability was .99. Analysis of the standardized Rasch residuals showed the scale's unidimensionality and absence of high item dependency (residual correlations, <.30). The correlation between the ABC-5/SLO and the Prosthetic Mobility Questionnaire (Rasch measures) was high (ρ=.84), as expected. Minor signs of item redundancy were found. CONCLUSIONS: The simplified ABC-5/SLO scale is a valid and reliable measure of balance confidence for individuals with lower-limb amputation. It is possible to transform the ordinal summed raw scores of the ABC-5/SLO into interval-level measurements using a nomogram.