Demir Pervin, Küçükdeveci Ayşe Adile, Kutlay Şehim, Elhan Atilla Halil
Department of Biostatistics and Medical Informatics, Ankara Yıldırım Beyazıt University Faculty of Medicine, Ankara, Türkiye.
Department of Physical Medicine and Rehabilitation, Ankara University Faculty of Medicine, Ankara, Türkiye.
Turk J Phys Med Rehabil. 2023 Jul 7;69(3):350-365. doi: 10.5606/tftrd.2023.12387. eCollection 2023 Sep.
Various scales exist to assess different domains of functioning in knee osteoarthritis (OA). This study aimed to explore whether it is possible to develop a common metric (CM) from the frequently used scales to assess functioning in knee OA.
The methodological study evaluated 411 patients (81 males, 330 females; mean age: 61.8±10.5 years; range, 41 to 88 years) with knee OA. Data from the Health Assessment Questionnaire, Oxford Knee Score, Medical Outcomes Study Short Form 36, Knee Injury and Osteoarthritis Outcome Score, Western Ontario and McMaster Universities Arthritis Index, and the Nottingham Health Profile were used, and the items focusing on self-care, mobility, and domestic activity domains based on the activities and participation component of the International Classification of Functioning, Disability, and Health were included. Concurrent calibration was performed to combine the items of the scales. The CM parameters were estimated using the Rasch measurement model. Reliability was assessed using the person separation index. The CM was utilized to generate a transformation table to convert the scale scores to each other based on the reference metric score.
Each scale fitted the Rasch model. Item invariance was achieved for the CM (p=0.775). The CM had a person separation index of 0.827. Age, sex, and disease duration did not cause difference in item functions. The CM satisfied the assumptions of unidimensionality and local independence.
A reliable CM was created from the commonly used scales to measure functioning in individuals with knee OA. Thus, clinicians and researchers can refer to the transformation table to directly compare scores of those scales and use them interchangeably.
存在多种量表用于评估膝骨关节炎（OA）不同方面的功能。本研究旨在探讨是否有可能从常用量表中开发一种通用指标（CM）来评估膝OA的功能。
该方法学研究评估了411例膝OA患者（81例男性，330例女性；平均年龄：61.8±10.5岁；范围41至88岁）。使用了健康评估问卷、牛津膝关节评分、医学结局研究简表36、膝关节损伤和骨关节炎结局评分、西安大略和麦克马斯特大学骨关节炎指数以及诺丁汉健康概况的数据，并纳入了基于国际功能、残疾和健康分类中活动与参与部分，聚焦自我护理、活动能力和家务活动领域的条目。进行了同时校准以合并各量表的条目。使用拉施测量模型估计CM参数。使用人员分离指数评估信度。利用CM生成转换表，以便根据参考指标分数将各量表分数相互转换。
每个量表均符合拉施模型。CM实现了条目不变性（p = 0.775）。CM的人员分离指数为0.827。年龄、性别和病程未导致条目功能出现差异。CM满足单维度性和局部独立性的假设。
从常用量表创建了一种可靠的CM，用于测量膝OA患者的功能。因此，临床医生和研究人员可以参考转换表直接比较这些量表的分数并可互换使用。
