中国儿童保健杂志 ›› 2020, Vol. 28 ›› Issue (3): 273-276.DOI: 10.11852/zgetbjzz2019-0856

• 科研论著 • 上一篇    下一篇

中文版Hammersmith婴儿神经学检查量表的信度和效度

刘振寰, 金炳旭, 刘陈, 曾贞, 钱旭光, 赵勇, 谭媛, 周园, 庞碧徽   

  1. 广州中医药大学附属南海妇产儿童医院儿童康复科,广东 佛山 528200
  • 收稿日期:2019-06-18 发布日期:2020-03-10 出版日期:2020-03-10
  • 作者简介:刘振寰(1958-),男,教授,主任医师,主要研究方向为小儿神经康复。

Reliability and validity of the Chinese Version of Hammersmith Infant Neurological Examination

LIU Zhen-huan, JIN Bing-xu, LIU Chen, ZENG Zhen, QIAN Xu-guang, ZHAO Yong, TAN Yuan, ZHOU Yuan, PANG Bi-hui   

  1. Department of Children Nerve Rehabilitation of Nanhai Maternity and Kid Hospital, Guangzhou University of Traditional Chinese Medicine, Foshan, Guangdong 528200, China
  • Received:2019-06-18 Online:2020-03-10 Published:2020-03-10

摘要: 目的 检验中文版Hammersmith 婴儿神经学检查量表(HINE)婴儿神经学检查量表的信度和效度。方法 将符合《中国脑性瘫痪康复指南(2015)》脑瘫高危儿诊断2月~1岁的59例患儿(胎龄32~36周和胎龄37~42周)和年龄相匹配的30例正常儿童进行中文版HINE评价。以三组间HINE总分和各因子评分比较确定区分效度;以脑瘫高危儿临床诊断为金标准,以ROC方法确定界限分及灵敏度、特异度;每组随机抽取10例由同一医师在首次检查后1周再次进行检查评估,以检测重测信度;每组随机抽取10例由两组评估医师同时进行HINE的检查和计分,以检测评分者信度。结果 早产脑瘫高危儿组、足月脑瘫高危儿组的量表总分和5个分项目评分均低于正常对照组,差异具有统计学意义(P<0.01);早产脑瘫高危儿组ROC曲线的线下面积为0.95,界限分为21分时灵敏度0.97,特异度0.93;足月脑瘫高危儿组ROC曲线的线下面积为0.93,界限分为26分时灵敏度0.97,特异度0.95,两组的灵敏度、特异度均≥0.9;量表总分评分者信度组内相关系数(ICC)>0.8,5个分项目评分者信度ICC>0.7;该量表总分和5个分项目重测信度r>0.8。结论 中文版HINE的效度、信度较好,灵敏度、特异度较高,可用于2月~1岁脑瘫高危儿的临床筛查。

关键词: 脑瘫高危儿, 中文版Hammersmith婴儿神经学检查量表, 信度, 效度

Abstract: Objective To test the validity and reliability of the Chinese version of Hammersmith Infant Neurological Examination (HINE). Methods Totally 59 children aged 2 months to 1 year old in diagnosis of high-risk infants with cerebral palsy were selected according to Chinese guidelines for rehabilitation of cerebral palsy (2015),devided into two groups by gestational-age,premarure group (32 to 36 weeks gestational-age) and term-infant group (37 to 42 weeks gestational-age).30 normal children matched with group in age were selected.All children were evaluated with the Chinese version of HINE.The total score of HINE and each factor score among the three groups were used to discriminate validity.As the gold standard the diagnosis of high-risk children with cerebral palsy,ROC was used to determine the cutoff point,tensitivity and specificity.10 cases were randomly selected from each group and re-examined and evaluated by the same physician 1 week after the first examination to test the test-retest reliability.10 cases were randomly selected from each group.The two groups of evaluators simultaneously conducted the examination and scoring of the HINE to test the inter-rater reliability. Results The total score of scale and the scores of 5 items in the preterm and term-infant groups were lower than those in the normal control group, and the differences were statistically significant (P<0.01).The area under the ROC curve of the preterm group was 0.95, when the limit was 21 points,the sensitivity was 0.97,and the specificity was 0.93.While the area under the ROC curve of the term-infant group was 0.93, when the limit was 26 points,the sensitivity was 0.97,and the specificity was 0.95. The sensitivity and specificity of both groups were ≥0.9.For the inter-rater reliability and the test-retest reliability,the intraclass correlation coefficient (ICC) of total score were both >0.80,the intraclass correlation coefficient (ICC) of the scores of 5 items was >0.70 and >0.80 respectively. Conclusions The Chinese version of HINE has good validity and reliability,high sensitivity and specificity.It could be used as a clinical screening instrument for high-risk cerebral palsy children aged from 2 months to 1 year.

Key words: high-risk cerebral palsy children, the Chinese version of Hammersmith Infant Neurological Examination, reliability, validity

中图分类号: