神经精神科问卷知情者版中文译本的信效度

Reliability and validity of Chinese version of Neuropsychiatric Inventory-Questionnaire in patients with Alzheimer' s disease

目的:评价神经精神科问卷知情者版(Neuropsychiatric Inventory-Questionnaire,NPI-Q)中文译本的信度与效度。方法:从北京大学精神卫生研究所记忆中心的病例库中,选取51例符合NINCDS-ADR-DA诊断标准的阿尔茨海默病(probable AD)患者,接受简明智能精神状况检查(Mini-mental Status Exami-nation,MMSE)、日常生活活动量表(Activities of Daily Living,ADL)、阿尔茨海默病评定量表(Alzheimer'sdisease Assessment Scale-Cognitiveportion,ADAS-cog)、神经精神科问卷(Neuropsychiatric Inventory,NPI)评定,并由其知情者进行NPI-Q评分。其中16例在临床干预3个月后进行重测。结果:(1)NPI-Q克朗巴赫系数α=0.851,Guttman分半系数为0.825;NPI-Q总分重测信度为ICC=0.860(P<0.01),NPI-Q苦恼程度总分重测信度ICC=0.480(P>0.05)。(2)NPI-Q各条目评分之间的相关系数为0.01～0.97,均与总分呈正相关(r=0.33～0.87,P<0.05)。NPI-Q各条目苦恼程度评分之间的相关系数为0.05～0.72,与苦恼程度总分正相关(r=0.47～0.90,P<0.01)。NPI-Q严重程度总分与苦恼程度总分相关(r=0.78,P<0.01)。NPI-Q严重程度评分与MMSE总分呈负相关(r=-0.39,P<0.01),与ADL、ADAS-cog、NPI总分、NPI苦恼程度评分均相关(r=0.47～0.82,均P<0.01)。NPI-Q苦恼程度评分与MMSE总分负相关(r=-0.32,P<0.05),与ADL、ADAS-cog、NPI总分、NPI苦恼程度相关(r=0.29～0.68,P<0.05)。因子分析提示NPI-Q含有精神病性、情感损害和额叶功能3个因子,总贡献率达62.23%。结论:神经精神科问卷知情者版中文译本具有较好的信度和效度,值得推广应用。

Objective： To evaluate the reliability and validity of Chinese version of Neuropsychiatric Inventory-Questionnaire （ NPI-Q） . Methods： Fifty-one patients meeting criteria of probable Alzheimer＇s disease （ AD ） of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer＇ s disease and Related Disorders Association （NINCDS-ADRDA） were studied. All patients were drawn from database of Dementia Care and Research Center at Peking University Institute of Mental Health. Each subject was administered with the Mini-mental Status Examination （MMSE）, Activities of Daily Living （ ADL）, Alzheimer＇s disease Assessment Scale-Cognitive portion （ ADAS-Cog）, and Neuropsychiatric Inventory （ NPI ） . The informants completed rating with NPI-Q. Sixteen subjects were retested at 3 months after initial assessment. Results ： （ 1 ） Reliability. The coefficient of inter-item consistency （Cronbach＇s α） was 0. 851 , and the Guttman split-half reliability was 0. 825. The test-retest reliability of NPI-Q total score inter-rater reliability （ICC） was 0. 860 （ P 〈 0. 01 ）, and test-retest reliability of NPI-Q total distress score ICC was 0. 480 （ P 〉 0. 05）. （ 2 ） Validity. The item scores of severity of NPI- Q were correlated with each other （ r = 0. 01 - 0. 97 ）, and with total score of NPI-Q significantly （ r = 0. 33 - 0. 87, P 〈 0. 05 ） . The item scores of distress of NPI-Q were correlated with each other （ r = 0.05 - 0. 72 ）, and with total score of distress significantly （ r = 0. 47 - 0.90, P 〈 0. 01 ） . For all items, the scores of distress were correlated with scores of severity significantly （ r =0. 60 -0. 86, P 〈0. 01 ） . The total score of NPI-Q severity was correlated with total score of distress （ r = 0. 78, P 〈 0. 01 ） . The total score of NPI-Q severity was negatively associated with total score of MMSE （ r = - 0. 39, P 〈 0. 01 ） , positively with total score of ADL, ADAS-cog, NPI and NPI-distress （ r = 0.47 - 0.82, P 〈 0. 01 1 - The total score of NPI-Q distress was negatively associated with that of MMSE （ r = - 0. 32, P 〈 0. 05 ）, and positively correlated with ADL, ADAS-Cog, NPI, and NPI-distress （r = 0. 29 -0. 68, P 〈 0. 05 ） . NPI-Q consisted of three factors counting for 62. 23 % of variance, including psychosis, impaired affect and frontal function. Contusion： The Chinese version of NPI-Q has good reliability and validity, worthy of extensive application.