Lauzon S, Corbière M, Bonin J P, Bonsack C, Lesage A D, Ricard N
Faculté des sciences infirmières, Université de Montréal, Montréal, CP 6128, Succursale Centre-ville, Montréal, QC H3C 3J7.
Can J Psychiatry. 2001 Nov;46(9):841-6. doi: 10.1177/070674370104600908.
This study reports the validation of the French version of the Health of the Nation Outcome Scales (HoNOS-F), a questionnaire developed to measure health and social functioning of people with mental illness.
Once each statement was tested for readability, the scale was administered to 3 samples of people suffering from severe mental disorders to estimate its reliability and validity. More specifically, tests were run to establish the internal consistency, the stability, and the interrater reliability of the HoNOS-F. Confirmative factor analyses and mean differences according to age, sex, and diagnosis were also conducted to evaluate respectively construct- and criterion-related validity.
Coefficients obtained from the various tests show that the scale is reliable only when the total score is used. The confirmatory factor analyses indicate that the observed data do not fit the 2 proposed models, a unidimensional model and a 4-dimension model. However, the scale did show criterion-related validity.
Results of the present study converge with those obtained on the original widely used English version. Therefore, we suggest that clinicians use the questionnaire by referring to each item separately and by considering such patient characteristics as age, sex, and diagnosis. We also suggest that researchers wishing to evaluate health and social functioning of persons with serious mental disorders use the total score. Caution is, however, warranted when interpreting the total score for a French-speaking population, because the factorial solution 1-dimension model did not prove to be satisfactory.
本研究报告了《国民健康结果量表》法语版(HoNOS-F)的验证情况,该量表是为衡量精神疾病患者的健康状况和社会功能而编制的问卷。
在对每条陈述进行可读性测试后,将该量表施用于3组患有严重精神障碍的人群样本,以评估其信度和效度。更具体地说,进行了多项测试以确定HoNOS-F的内部一致性、稳定性和评分者间信度。还进行了验证性因素分析以及按年龄、性别和诊断分类的均值差异分析,以分别评估结构效度和效标关联效度。
各项测试所得系数表明,该量表仅在使用总分时才具有信度。验证性因素分析表明,观测数据不符合所提出的两个模型,即单维模型和四维模型。然而,该量表确实显示出效标关联效度。
本研究结果与在广泛使用的原始英文版上获得的结果一致。因此,我们建议临床医生在使用问卷时分别参考每个项目,并考虑患者的年龄、性别和诊断等特征。我们还建议,希望评估严重精神障碍患者健康状况和社会功能的研究人员使用总分。不过,在为说法语的人群解释总分时应谨慎,因为单维模型的因子解并不令人满意。