In the present study, an association was observed between antenatal depression symptoms and low birth weight. Women with significantly higher EPDS score were more likely to have low birth weight infants. This association persisted after adjusting for confounders, including pre-pregnancy BMI, neonate gender, maternal age, degree of education, high-risk women and parity. Findings in previous studies regarding the association between antenatal depression and low birth weight remain inconclusive. This disparity may result in different sample size, confounders adjusted, and depression symptoms measurements.
In order to facilitate large-scale epidemiological investigation, EPDS was used as the depression symptom measurement. Although clinician administered diagnostic instruments are the gold standard, they are time-consuming and staff intensive to administer[18]. Conversely, patient-rated screening instruments are easier to use. Because adverse obstetrical outcomes can have significant, long-term, negative health impacts, it would be useful to screen depression at the initial obstetric visit predicts the risk of adverse pregnancy outcomes. EPDS is the only instrument designed to exclude somatic depressive symptoms that are also common symptoms of pregnancy. The effectiveness in antenatal women has been verified by previous studies with excellent reliability[12, 14]. Additionally, a Chinese translation of the EPDS has been demonstrated to have good reliability and validity[15]. We focused on depression symptoms rather than clinically diagnosed depression. The results showed that depression symptoms during pregnancy still had a positive effect on LBW after adjusting several confounders.
Our findings are in keeping with those studies in America black women with large sample size. Some studies[19-21]agreed that depression symptoms are one of the causes of LBW, especially in Neggers’s study[19], depression symptoms rather than depression was concerned. Consistent with our results, Negger et al found that women who are depressed during pregnancy are at increased risk of giving birth to low birth weight(AOR =1.4, 95% CI : 1.1-1.7), in which the measure of depression was conducted by the Center for Epidemiologic Studies Depression Scale (CES-D). Although the depression measure was not the same, we both concerned about the depression symptoms rather than the diagnosis of depressive disorder. However, our attention is paid on the second trimester, suggesting that the depressive stage should be concerned and controlled in earlier pregnancy. Additionally, EPDS is more convenient to use in large population survey. Evans's study used EPDS to screen depression symptoms at 18 and 32 weeks respectively in 13,194 British women, showed that depression symptoms at 18 weeks increased the risk of LBW, but the risk was disappeared after adjusting smoking [22]. It was possible that smoking lies on the causal pathway between depression and intrauterine growth restriction. Adjustment for smoking removed any evidence of a detect association between birth weight and mood during pregnancy. This study mainly focused on white women. We have noticed that the results in different races are conflicting. In a prospective cohort study of 819 African-American women, El-Mohands et al[23]. Found that women with depression, assessed at <29 weeks gestational age, had 71% increased odds of LBW in adjusted analysis (OR=1.71 [95%CI=1.12,2.62]). Another study among 583 women in Bangladesh[24], showed that depression in the 3rd trimester, assessed via the EPDS, doubled the odds of LBW(OR=2.24 [95%CI-1.37, 3.68]) in adjusted analysis. Whether the association between depression and LBW are affected by race, socioeconomic level remains to be further explored.
Relevant studies from China and with large sample size are limited. Yang's case-control study in Wuhan found that antenatal depression or anxiety was not significantly associated with LBW, but individuals with antenatal depression combined with anxiety were at a higher risk of LBW[25]. However, the self-designed questionnaires, rather than a typical screening scale were used in the study for measuring depressive states. Existing reviews tend to suggest that depression, especially in early pregnancy, may be a risk factor for LBW[26, 27]. Our study supported this view, but more evidence is needed.
We found that young mothers (<25 years of age) seem to be more susceptible to antenatal depression. This might be related to depressive younger pregnant women lacking acknowledgement on pregnant health care. Some studies showed that depression has the tendency to occur in younger pregnant women[28]. We also found that antenatal depression symptoms were more common among temporary residents than permanent residents. This may be related to immigration factors, such as dialect, social isolation, life adaption, unfair opportunities, economic problems or the realization of health care[29].
There are several mechanisms that have been proposed to underlie the association between antenatal depression and neonatal outcomes. A hypothesis has been most extensively investigated which proposed that the hypothalamic-pituitary- adrenocortical axis, which stimulates the release of stress hormones such as cortisol and catecholamines, has a mediation on the association [30]. Study also suggested that the association between self-reported depression and gestational age at birth and fetal growth rate was mediated by antenatal maternal cortisol [31]. Additionally, adverse health behaviors were associated with depression, such as depressed women are more likely to smoke and drink during pregnancy than non-depressed women.
There are limitations in this study. Firstly, depression symptoms were assessed at the second-trimester only, so we probably failed to observe the trend of mood changes throughout the pregnancy. Some women whose depression symptoms occurred at the third-trimester might be misclassified into the non-deprssion group, then the adverse effect of antenatal depression on LBW may be underestimated. Furthermore, data on the history of low birth weight delivery were undocumented. Thus there might be some residual confouders were not adjusted.