Impact of the COVID-19 epidemic on patterns of pregnant women's perception of threat and its relationship to mental state: A latent class analysis

The objective of this study was to define the threatened perception types of pregnant women during the COVID-19 pandemic and determine the correlations between the perception types and their demographic factors, their preventive knowledge of COVID-19 and their mental status in order to provide suggestions for pregnant women during pandemic. Latent class analysis were used to explore the optimal numbers of clusters. Multinomial logistic regression and multiple correspondence analysis were used to analyze the demographic variables of the latent categories. MANOVA was used to analyze the difference of knowledge of COVID-19 obtained among clusters and their psychological status, and chi-square test was used determine the relationship between the latent clusters and the participant's COVID-19 worry level. Five clusters were found: the first cluster (n = 120, 39%) was unthreatened and confident. Cluster 2(n = 84, 28%) was unthreatened but not confident. Cluster 3 (n = 49, 17%) was threatened but confident. Cluster 4 (n = 25, 9%) was threaten, not confident and knowledgeable, and Cluster 5 (n = 20, 7%) was threatened, not confident and lacking knowledge. Three demographic variables were shown an effect on the classification, they were support from work, family support and intrapartum and postpartum complications. This study can help assess the mental health risks of pregnant women during an epidemic. The results could be helpful for families, work units, communities and medical institutions to make targeted intervention decisions for pregnant women.