OBJECTIVETo analyze the associations between pregnancy-related anxiety and the prevalence of subthreshold autism trait (SAT) in preschool children.

METHODSBaseline data came from the Ma'anshan Birth Cohort Study, a part of the China-Anhui Birth Cohort Study (C-ABCS). All the participants were enrolled among pregnant women who received prenatal health care in 4 municipal medical centers during Oct. 2008 to Oct. 2010. A total of 5 084 pregnant women were recruited at the beginning and 4 669 singleton live births were included until childbirth. The situation about pregnancy-specific anxiety during trimester and third trimester of women were evaluated by Pregnancy-specific Anxiety Questionnaire (PAQ). Between April 2014 and April 2015, the cohort was followed up again, and the Clancy Autism Behavior Scale (CABRS) filled out by parents was used for telling the SAT children from the healthy children among 3 663 preschool children. Univariate and binary regression model was used to estimate associations between the pregnancy-related anxiety during trimester and third trimester and the subthreshold autism trait in children.

RESULTSDuring the pregnancy, the detected rates of women with pregnancy-specific anxiety in trimester and the third trimester were 25.5%(935/3 663), 13.9%(501/3 592) respectively, and the detected rate of maternal pregnancy-specific anxiety in both periods was 7.7%(278/3 592). There were 290 positive children with SAT and the detection rate was 7.9%. After controlling possible confounding factors including children's genders, place of residence, supplement folic acid during pregnancy, preterm birth, exposure to second-hand smoke during pregnancy, the father (mother) cultural levels, the father (mother) nature of work and family income, the results of multinomial logistic regression analysis showed that maternal pregnancy-specific anxiety in trimester was the risk factor for SAT in preschool children (OR=1.51, 95% CI: 1.11-2.04), and there was no association between maternal pregnancy-specific anxiety in the third trimester and SAT in preschool children (OR=1.36, 95% CI: 0.82-2.22). Compared with the single function of maternal pregnancy-specific anxiety in trimester or the third trimester for SAT in preschool children, maternal pregnancy-specific anxiety in both periods presented a joint action that increasing the risk for SAT (OR=2.02, 95% CI: 1.36-2.98).

CONCLUSIONMaternal pregnancy-related anxiety was a risk factor for subthreshold autism trait in preschooler children. Pregnant women should try to keep a good mental state to create a good environment for fetal growth.