Dads matter too: the role of paternal factors on birth outcomes

Abstract

Background Vast evidence exists illustrating the role of maternal factors on key birth outcomes and adverse birth events in low to middle-income countries. Little attention is given to investigating the role of paternal factors, and their independent contribution after adjusting for maternal factors and other known covariates. More importantly, little is known of these associations in African populations undergoing significant health transitions. Aim The aim of the study was to investigate and explore the independent associations between paternal, maternal factors and neonatal outcomes namely; birth weight (BW), birth length (BL), head circumference (HC) and ponderal index (PI). Methods A cross-sectional study design including a secondary data analysis approach was used. The data were primarily collected by the Soweto first 1000 days (S1000) and Father of the Baby (FOB) studies conducted in Soweto, South Africa in 2013-2015. Paternal, maternal and other key data were used in the analysis. Descriptive statistics were conducted to summarise paternal, maternal and neonatal characteristics. Furthermore, to explore independent and adjusted associations of paternal and maternal factors with birth outcomes, multivariate linear regression models were computed. All statistical significance was defined set at p<0.05. Results In the current study, paternal factors, namely Body Mass Index (BMI) and waist circumference (WC) were significantly associated with BW (β: -41.00; 95%CI: -79.77; -2.30; p = 0.038), (β: 19.32; 95%Cl: 2.21; 36.42; p = 0.027), respectively. However, after adjusting for neonatal factors, self-reported marital status where fathers of the baby were cohabitating with their partners was significantly associated with BW (β: 194.00; 95%CI: 14.71; 373.26; p = 0.034). Once maternal factors were added to the model, all the paternal factors and neonatal BW associations became insignificant. Paternal BMI was significantly associated with HC (β: -0.15; 95%CI: -0.28;-0.01; p = 0.042), however the association diminished once neonatal and maternal factors were added to the model. Maternal BMI and age were independent predictors for BW, but after adjusting for paternal and neonatal factors only maternal BMI remained a significant predictor of BW (β: 21.43; 95%CI: 9.39; 33.48; p = 0.001). Paternal factors were not significantly associated with BW, HC and PI when maternal and neonatal factors were held constant. As expected, in the full models, gestational age (GA) was a consistent intermediary predictor for BW (β: 132.54; 95%CI: 95.08; 170.0; p = 0.001), BL (β: 0.91; 95%CI: 0.06; 1.23; p = 0.001) and HC (β: 0.42; 95%CI: 0.27; 0.56; p = 0.001). Whilst sex at birth consistently had a significant association with only BL (β: -1.52; 95%CI: -2.80; -0.23; p = 0.021). Conclusion Paternal measures of adiposity (BMI and WC) may play an important role in birth outcomes in addition to social support indexed by marital status and these factors may be mediated by maternal and intermediate neonatal factors like GA and sex. Findings affirm the important role of maternal, paternal and neonatal factors in relation to birth outcomes. The current evidence needs to be further researched and strengthened by using a larger sample size.