Damiano Rodolfo Furlan, Hoffmann Maurício Scopel, Gosmann Natan Pereira, Pan Pedro Mario, Miguel Eurípedes Constantino, Salum Giovanni Abrahão
Departamento e Instituto de Psiquiatria, Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil.
Departamento de Neuropsiquiatria, Universidade Federal de Santa Maria, Santa Maria, Rio Grande do Sul, Brazil. Care Policy and Evaluation Centre, London School of Economics and Political Science, London, UK. Postgraduate Program in Psychiatry and Behavioral Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
Braz J Psychiatry. 2023 Mar 19;45(4):310-7. doi: 10.47626/1516-4446-2022-2945.
To provide practical norms for measuring depressive symptoms using the Patient Health Questionnaire 9 (PHQ-9) in Brazil using a state-of-art psychometrics analysis.
We used a large and representative Brazilian dataset from the 'Pesquisa Nacional de Saúde - 2019'(PNS-2019), which includes 90,846 Brazilian citizens. First, to assess the scale structure, we assessed the unidimensional model using Confirmatory Factor Analysis (CFA). Second, we used Item Response Theory (IRT) to characterize depressive symptoms´ distribution. Then, we linked summed- and meanbased PHQ-9 scores with the IRT-based score by using generalized additive models. Finally, we generated percentiles, T scores, and a newly developed score, called D scores (decimal scores), to describe the PHQ-9 norms for Brazilian population.
CFA revealed a good fit to the unidimensional model, showing to be invariant to age and sex. IRT captured item-level information of the latent trait (reliable from 1 to 3 standard deviations above the mean). Brazilian norms were presented using summed-, T-scores, and D-scores.
This is the first study to define Brazilian´s norms for the PHQ-9 among a large representative sample, using robust psychometric tools. More precise PHQ-9 scores are now available and may be widely used in primary and specialized clinical care settings.
运用先进的心理测量学分析方法，为在巴西使用患者健康问卷9（PHQ - 9）测量抑郁症状提供实用规范。
我们使用了来自“2019年全国健康调查”（PNS - 2019）的一个大型且具有代表性的巴西数据集，其中包括90,846名巴西公民。首先，为评估量表结构，我们使用验证性因子分析（CFA）评估单维模型。其次，我们使用项目反应理论（IRT）来描述抑郁症状的分布特征。然后，我们通过广义相加模型将基于总分和平均分的PHQ - 9得分与基于IRT的得分联系起来。最后，我们生成百分位数、T分数以及一个新开发的分数，称为D分数（十进制分数），以描述巴西人群的PHQ - 9规范。
CFA显示该模型与单维模型拟合良好，且对年龄和性别具有不变性。IRT获取了潜在特质的项目层面信息（在均值以上1至3个标准差范围内可靠）。使用总分、T分数和D分数呈现了巴西的规范。
这是第一项在大型代表性样本中使用强大的心理测量工具为巴西定义PHQ - 9规范的研究。现在可以获得更精确的PHQ - 9分数，并且可能在初级和专科临床护理环境中广泛使用。
