Wang Zhao-Ying, Hu Shi-Xiong, Lu Jian, Shang Wen, Chen Tao, Zhang Rui-Ting
Department of Psychology, Hunan Normal University, Changsha, China; Cognition and Human Behavior Key Laboratory of Hunan Province, Hunan Normal University, Changsha, China; Institute of Interdisciplinary Studies, Hunan Normal University, Changsha, China; Center for Mind & Brain Sciences, Hunan Normal University, Changsha, China.
Shuda College, Hunan Normal University, Changsha, China.
Child Abuse Negl. 2025 Feb;160:107201. doi: 10.1016/j.chiabu.2024.107201. Epub 2024 Dec 27.
Accumulating literature has found a close relation between early life adversity (ELA) and anxiety. However, previous studies did not rule out the high co-occurrence of different types of ELA when exploring the association of ELA and anxiety. In the present study, we carried out network analysis based on a cross-sectional sample and longitudinal sample to investigate the relationship between ELA and anxiety symptoms in non-clinical populations over time.
Online advertisement was carried out to recruit participants. The cross-sectional sample included 871 Chinese participants (M = 19.11, SD = 1.57), and the longitudinal sample involved 440 Chinese participants (M = 18.93, SD = 0.75). Three dimensions of ELA were assessed. The Threat/Harm dimension was assessed by subscales of physical abuse, emotional abuse, and sexual abuse of Childhood Trauma Questionnaire (CTQ). The Deprivation dimension of ELA was measured by subscales of physical neglect, and emotional neglect of CTQ. The Unpredictability dimension of ELA was evaluated by the Childhood Unpredictability scale. Anxiety symptoms were captured by the Generalized Anxiety Disorder-7 (GAD-7). Regularized partial correlation networks were constructed, and the expected influence (EI) as well as predictability of each node were calculated. Stability within the network was tested and the network comparison test was conducted to examine the difference between the cross-sectional network and the longitudinal network.
The cross-sectional network was relatively tight, and nodes within the dimension of ELA clustered together. Childhood unpredictability and emotional abuse revealed stronger associations with anxiety symptoms than other ELAs. Emotional abuse showed the highest EI in the network. These findings were replicated in the longitudinal network. The network comparison test indicated no significant difference between the cross-sectional network and the longitudinal network.
Childhood unpredictability and emotional abuse were strong predictors of anxiety symptoms, and the prediction was reliable and persistent. Reduce the experience of childhood emotional abuse can serve as a target to prevent anxiety.
越来越多的文献发现早年生活逆境（ELA）与焦虑之间存在密切关系。然而，以往研究在探讨ELA与焦虑的关联时，并未排除不同类型ELA的高共现情况。在本研究中，我们基于横断面样本和纵向样本进行了网络分析，以探讨非临床人群中ELA与焦虑症状随时间的关系。
通过在线广告招募参与者。横断面样本包括871名中国参与者（M = 19.11，SD = 1.57），纵向样本包括440名中国参与者（M = 18.93，SD = 0.75）。对ELA的三个维度进行了评估。威胁/伤害维度通过儿童创伤问卷（CTQ）中的身体虐待、情感虐待和性虐待分量表进行评估。ELA的剥夺维度通过CTQ中的身体忽视和情感忽视分量表进行测量。ELA的不可预测性维度通过儿童不可预测性量表进行评估。焦虑症状通过广泛性焦虑障碍-7（GAD-7）进行测量。构建了正则化偏相关网络，并计算了每个节点的预期影响（EI）以及可预测性。测试了网络内的稳定性，并进行了网络比较测试，以检验横断面网络与纵向网络之间的差异。
横断面网络相对紧密，ELA维度内的节点聚集在一起。童年不可预测性和情感虐待与焦虑症状的关联比其他ELA更强。情感虐待在网络中显示出最高的EI。这些发现在纵向网络中得到了重复。网络比较测试表明横断面网络与纵向网络之间没有显著差异。
童年不可预测性和情感虐待是焦虑症状的强预测因素，且该预测可靠且持续。减少童年情感虐待的经历可作为预防焦虑的目标。
