Rasch Analysis of the Body Function Component of the ICF Rehabilitation Set in Functional Assessment of Orthopedic Patients

WANG Wenxuan,XU Guirong,HE Xinghui,LI Kun,WANG Yingmin
DOI: https://doi.org/10.3724/sp.j.1329.2022.05003
2022-01-01
Abstract:ObjectiveTo examine the feasibility of the Body Function component in the International Classification of Functioning, Disability and Health Rehabilitation Set (ICF-RS) as a functional assessment instrument for orthopedic patients by using Rasch analysis.MethodsA total of 119 orthopedic inpatients hospitalized in Shenzhen Longcheng Hospital were recruited from July 2018 to November 2019 through convenient sampling method. The component of Body Function (9 categories) of the ICF-RS was used to assess the patients. Rasch analysis was performed using the RUMM 2030 software. The psychometric properties were reported including overall goodness of fit, goodness of fit for each item, person separation index, differential item functioning, and person-item threshold distribution map.ResultsThe first-round Rasch analysis found that the overall goodness of fit for the Body Function component of the ICF-RS was poor (χ2=39.01, P=0.002 8, the significance level after Bonferroni correction was 0.05/9=0.005 6), the item-fit and individual-fit residuals were (0.36±1.34) and (-0.24±0.62), respectively. The person separation index was 0.55. The residuals of the category b130 Energy and Drive Function was 3.642, more than 2.5, and it fit the Rasch model poorly. After deleting the category b130 Energy and Drive Function, the second-round Rasch analysis found that the overall goodness of fit for the Body Function component of the ICF-RS was satisfied (χ2=25.60, P=0.059 9, the significance level after Bonferroni correction was 0.05/8=0.006 3), the item-fit and individual-fit residuals were (0.40±0.90) and (-0.27±0.59), respectively. The person separation index was 0.58. The residuals of the remaining items were all within ±2.5 with a nonsignificant P value. The results showed no differential item functioning in terms of age, gender, course of disease, and marital status. The person-item threshold distribution map showed that the categories basically covered the patient coordinates. The patients performed the worst in the category b280 Pain Sensation and the best in the category b640 Sexual Function.ConclusionAfter deleting the category b130 Energy and Drive Function, the remaining 8 categories of Body Function component in the ICF-RS fit well with the Rasch model, and could be used as a functional assessment instrument for orthopedic patients.
What problem does this paper attempt to address?