Anxiety and depressive symptoms in pregnancy predict low birth weight differentially in male and female infants-findings from an urban pregnancy cohort in India.

This study examined the contributions of antenatal anxiety, depression, and partner violence to low birth weight (LBW) in infants and to sex-specific birth weight outcomes among mothers from a cohort in urban India. Data from 700 mothers from the PRAMMS cohort (Prospective Assessment of Maternal Mental Health Study) were used. Pregnant women were assessed in each trimester—T1, T2 and T3, for symptoms of anxiety, and depression as well as partner violence. Multivariate analyses were performed for the whole sample and then for male and female infants separately. The final multivariable logistic regression models were each built using a backward selection procedure and controlling for confounders. To accommodate longitudinally measured data, change in scores (T2–T1 and T3–T2) of anxiety and depression were included in the model. Of the 583 women with a singleton live birth, birth weight was available for 514 infants  and LBW was recorded in 80 infants (15.6%). Of these, 23 infants were preterm. Overall, higher T1 Depression scores (OR: 1.11; 95% CI 1.040, 1.187) and an increase in both Depression scores (OR: 1.12; 95% CI 1.047, 1.195) from T1 to T2 and Anxiety scores (OR: 1.32; 95% CI 1.079, 1.603) between T2 and T3 were predictors of LBW. Female infants had a higher chance of LBW with increase in maternal anxiety between T1–T2 (OR: 1.69; 95% CI 1.053, 2.708) and T2–T3 (OR: 1.49; 95% CI 1.058, 2.086); partner violence during pregnancy just failed to reach conventional statistical significance (OR: 2.48; 95% CI 0.810, 7.581) in girls. Male infants had a higher chance of LBW with higher baseline depression scores at T1 (OR: 1.23; 95% CI 1.042, 1.452) and an increase in depression scores (OR: 1.25; 95% CI 1.060, 1.472) from T1 to T2. Increasing prenatal anxiety and depressive symptoms in different trimesters of pregnancy were associated with LBW with sex-specific patterns of association in this sample from a Low and Middle Income Country.