围生期抑郁和焦虑症状关联及方向的纵向研究

目的 利用重复测量资料探讨围生期抑郁和焦虑症状的纵向关联及方向。方法 2019年5—9月在马鞍山市妇幼保健院纳入孕早期妇女，分别在孕中、晚期及产后一年内对其进行抑郁、焦虑症状随访评估。对围生期抑郁、焦虑症状进行单因素分析后，在自回归交叉滞后模型中调整单因素分析有统计学意义的因素进行路径分析以阐明围生期抑郁和焦虑症状的关联及可能的预测方向。结果 1 094名妇女纳入分析，围生期抑郁、焦虑症状及其共患的检出率分别为27.3%、44.3%和23.5%。路径分析结果显示，同一时期的抑郁和焦虑症状均存在正相关（相关系数r分别为0.705、0.636、0.547、0.665，P＜0.001）；围生期抑郁、焦虑症状的自回归路径均有统计学意义（回归系数b抑郁分别为0.315、0.462、0.373，b焦虑分别为0.473、0.567、0.589，P＜0.001）；在整个围生期均观察到前一时期焦虑症状与后一时期抑郁症状的正向关联（b分别为0.155、0.234、0.271，P＜0.001），而只在孕中期抑郁症状对孕晚期焦虑症状中观察到正向关联（b = 0.120，P＜0.05）。敏感性分析显示路径分析结果具有稳健性。结论 围生期抑郁和焦虑症状存在共患，且焦虑症状对随后的抑郁症状有稳定正向预测作用。

A longitudinal study of the association and direction between perinatal depression and anxiety symptoms

Objective To explore the longitudinal association and direction between perinatal depression and anxiety symptoms by using repeated measurements. Methods Women were recruited in the first trimester from May to September 2019 in Ma’anshan Maternal and Child Health Care Center. Each participant was followed up in the second and third trimesters and within 1 year postpartum with assessed depression and anxiety symptoms. The autoregressive cross-lagged model (ARCLM) was used to conduct path analysis to clarify the association and possible direction of perinatal depression and anxiety symptoms adjusting for covariates having significant association with perinatal depression or anxiety symptom in univariate analysis. Results In total 1 094 women were included in this study. The detection rates of perinatal depression, anxiety symptoms and their comorbidity were 27.3%, 44.3%, and 23.5%, respectively. Results of path analysis showed that depression and anxiety symptoms were correlated positively in the same period (the correlation coefficient r were 0.705, 0.636, 0.547 and 0.665, respectively, P<0.001); the autoregressive paths of perinatal depression and anxiety symptoms were all statistically significant (the regression coefficient b for depression were 0.315, 0.462 and 0.373, b for anxiety were 0.473, 0.567 and 0.589, respectively, P<0.001); a positive correlation between anxiety symptom in the previous period and depression symptom in the later period was observed throughout the perinatal period (b were 0.155, 0.234 and 0.271, respectively, P<0.001). However, only depression symptom in the second trimester was positively correlated with anxiety symptom in the third trimester (b=0.120, P<0.05). Sensitivity analysis showed that the results of path analysis were robust. Conclusion Perinatal depression symptom has a significant comorbidity with perinatal anxiety symptom. Most importantly, anxiety symptom was a stable and positive predictor of subsequent depression symptom during the whole perinatal period.