Department of Epidemiology, Center of Clinical Epidemiology and Evidence Based Medicine, School of Public Health, Shanxi Medical University, 56 Xinjian South Road, Taiyuan, Shanxi 030001, China; Department of Health Education, Shanxi Women and Children Health Hospital, Taiyuan 030013, China.
Department of Child and Adolescent, School of Public health, Shanxi Medical University, Taiyuan 030001, China; Department of Mental Health, Shanxi Women and Children Health Hospital, Taiyuan 030013, China.
Midwifery. 2024 May;132:103963. doi: 10.1016/j.midw.2024.103963. Epub 2024 Feb 29.
There are inconsistent results on the Edinburgh Postnatal Depression Scale's (EPDS) factor structure and longitudinal invariance among different cultures. Furthermore, limited relevant studies in Chinese pregnant women exist.
To test the factor structure of the Chinese Mainland EPDS during pregnancy and conduct longitudinal invariance analyses.
A national multi-centre cohort study was conducted among 1207 pregnant women selected consecutively by convenience sampling from five hospitals in Zhuhai, Taiyuan, Haidian, Changchun, and Shenzhen in China between August 2015 and October 2016. Depression was measured by the EPDS during gestational weeks 10-13, 15-18, 23-25, 30-32 and 36-37, respectively.s RESULTS: Three factors with eigenvalues nearly larger than 1.0 were optimal for the Chinese Mainland EPDS, labelled "anxiety," "anhedonia," and "depression," and contained items 3-5, 1-2, and 6-10, respectively. The confirmatory factor analysis results of standardized root mean square residual (SRMR) = 0.034, root mean square error of approximation (RMSEA) = 0.049, comparative fit index (CFI) = 0.968, Tucker-Lewis index (TLI) = 0.954, and χ, p < 0.05 indicated good fit. For the longitudinal invariance tests, the configural invariance was met, with the CFI and TLI both higher than 0.90 and the RMSEA lower than 0.08 (CFI = 0.919, TLI = 0.908, and RMSEA = 0.034). The metric-, scalar-, and strict invariances were met.
The three-factor model of the Chinese Mainland EPDS is invariant in pregnancy, suggesting stability and comparability in identifying the women screened positive at different points during pregnancy and making the scale feasible to screen prenatal depression and anxiety simultaneously.
爱丁堡产后抑郁量表（EPDS）在不同文化中的因子结构和纵向不变性存在不一致的结果。此外，针对中国孕妇的相关研究有限。
检验中国大陆 EPDS 在孕期的因子结构，并进行纵向不变性分析。
2015 年 8 月至 2016 年 10 月，采用便利抽样法，连续选取中国珠海、太原、海淀、长春和深圳的 5 家医院的 1207 名孕妇进行全国多中心队列研究。分别在妊娠 10-13、15-18、23-25、30-32 和 36-37 周时，采用 EPDS 评估抑郁情况。
对于中国大陆 EPDS，特征值均大于 1.0 的三个因子为“焦虑”、“快感缺失”和“抑郁”，包含项目 3-5、1-2 和 6-10。标准化根均方残差（SRMR）=0.034、近似误差均方根（RMSEA）=0.049、比较拟合指数（CFI）=0.968、塔克-刘易斯指数（TLI）=0.954、χ2，p＜0.05 提示拟合良好。进行纵向不变性检验时，符合模型配置不变性，CFI 和 TLI 均高于 0.90，RMSEA 低于 0.08（CFI=0.919，TLI=0.908，RMSEA=0.034）。达到度量不变性、标度不变性和严格不变性。
中国大陆 EPDS 的三因子模型在孕期具有不变性，提示在不同孕期筛查阳性的女性中具有稳定性和可比性，使该量表能够同时筛查产前抑郁和焦虑。
