首页|基于Hammersmith功能运动量表扩展版的脊髓性肌萎缩症功能评估项目反应理论模型研究

基于Hammersmith功能运动量表扩展版的脊髓性肌萎缩症功能评估项目反应理论模型研究

扫码查看
目的 基于Hammersmith功能运动量表扩展版(HFMSE)探索建立用于评估脊髓性肌萎缩症(SMA)的项目反应理论(IRT)模型.方法 选择2021年2月—2023年6月在南京医科大学第一附属医院老年神经内科及儿科接受修正治疗的SMA患者23例,采用堆栈设计对重复测量数据进行建模;基于似然比计算组成最优模型的HFMSE项目区分度、猜测度、困难度和患者能力水平(θ)参数;采用测试功能差异(DTF)分析量表是否受到康复治疗或性别影响而存在偏倚;采用蒙特卡洛模拟证明模型项目难度分布未受采样过程影响,根据难度预期值中位数绘制人项图,并探讨HFMSE量表中各项目的难度、患者能力水平在首次评估和后续评估时的差异.结果 本研究共纳入32个项目构建4个备选模型,其中3参数逻辑斯蒂模型(3PLM)具有最佳拟合优度(P>0.05,M2统计量=479.75).3PLM中所有项目谨慎度均为1,可计算出项目的区分度、猜测度和困难度,且该模型在性别和康复治疗这2个因素上无明显评估偏倚.3PLM有较高信度(Cronbach's α=0.955,LCRC=0.981)和效度[θ值与原始总分存在正相关关系(r=0.99,P<0.001)],可通过二项式换算患者能力水平θ值与量表总分.通过可展示项目参数和个人参数在患者能力水平(θ)标尺上位置分布的人项图,观察θ值与项目难度的关系.结论 基于HFMSE建立的IRT模型可为不同场景SMA患者功能评估和制订康复治疗决策提供建议,但应注意其项目猜测度.
Item Response Theory Model for Functional Assessment of Spinal Muscular Atrophy Based on the Hammersmith Functional Motor Scale Expanded
Objective To establish an item response theory(IRT)model for functional assessment of spinal muscular atrophy based on the Hammersmith functional motor scale expanded(HFMSE).Methods A total of 23 patients with SMA treated with cor-rective treatment in the Department of Geriatric Neurology and Pediatrics of the First Affiliated Hospital of Nanjing Medical Univer-sity from February 2021 to June 2023 were included.Stack design was used to construct item response models(IRMs)for repeated measurement data.The likelihood ratio was used to calculate the parameters of the discrimination degree,guessing degree,difficulty and patient ability level(θ)of the HFMSE items for the optimal model.Differential test functioning(DTF)was used to analyze whether the scale was biased by rehabilitation treatment or gender.Monte Carlo simulation was used to prove that the difficulty dis-tribution of the model items was not affected by the sampling process,a person-item plot was drawn according to the median expect-ed difficulty,and the differences of the difficulty of each item in the HFMSE scale and the patient ability levels between the first as-sessment and the follow-up assessment were also discussed.Results A total of 32 items were included in this study to construct four alternative models,of which the 3 parameter logistic model(3PLM)had the best goodness of fit(P>0.05,M2 statistic=479.75).The caution degree of all items was 1 in 3PLM,and the discrimination degree,guessing degree and difficulty of the items were cal-culated.The model had no significant evaluation bias on the factors of gender and rehabilitation treatment.The results showed high reliability(Cronbach's α=0.955,LCRC=0.981)and validity[θ value was positively correlated with the original total score(r=0.99,P<0.001)]of 3PLM,allowing conversion of the θ value of the patient's ability to total score of the scale by binomial transformation.A person-item plot showing the distribution of item parameters and individual parameters on the patient ability level(θ)scale was used to observe the relationship between θ values and item difficulties.Conclusion IRT model based on HFMSE can provide sugges-tions on functional assessment and decision-making in rehabilitation treatment for patients with SMA in different scenarios,but at-tention should be paid to its item guessing degree.

spinal muscular atrophyHammersmith functional motor scale expandeditem response theoryfunctional analysis

范亚蓓、陈瑾瑾、葛盼丽、程曦、张旖宸、林枫

展开 >

南京医科大学第一附属医院,江苏 南京 210029

南京医科大学康复医学院,江苏 南京 210003

南京医科大学附属逸夫医院,江苏 南京 211100

脊髓性肌萎缩症 Hammersmith功能运动量表扩展版 项目反应理论 功能分析

2024

康复学报
福建中医药大学

康复学报

CSTPCD
影响因子:0.545
ISSN:2096-0328
年,卷(期):2024.34(6)