Survival and Risk Factors for Late Preterm and Early Term Neonates Compared to Full-Term Infants at Garissa County Referral Hospital, Kenya

Abstract

Preterm birth constitutes a significant health burden in Kenya, yet there is hardly any targeted research looking at the survival and specific risk factors associated with birth of late preterm neonates (LPNs) and early term neonates (ETNs), particularly in Garissa County. While studies on preterm birth determinants have been done in Kenya, factors specifically influencing LPNs and ETNs births and their survival relative to their full-term counterparts remain understudied. Studies done on preterm in general have identified maternal age, history of preterm birth, pregnancy-induced hypertension, and prolonged premature rapture of the membrane among others as significant risk factors, but their relevance to LPNs and ETNs specifically has not been fully investigated. This study aimed to address these gaps by determining maternal as well as fetal risk factors associated with LPN and ETN births compared to full-term neonates (FTNs) born at Garissa County Referral Hospital. Further, it sought to identify predictors of survival across these neonatal groups. The study conducted at Garissa County Referral Hospital employed a prospective cohort design.  Mother-neonate dyads were enrolled using convenience sampling method until the required sample was achieved. Data was collected using pretested and validated questionnaires.  STATA version 17 was used for analysis. Multinomial logistic regression analysis was performed to determine Relative Risk Ratio. The P-value was set α=0.05 and P<0.05 was considered statistically significant. Survival rates were estimated using Kaplan-Meier survival analysis. Bivariate Cox regression analysis was employed to identify independent predictors at p<0.20. However, since all variables had p-values >0.02, multivariate Cox regression was not conducted. The study revealed that maternal age (P=0.042), occupation (P=0.024), ethnicity (P=0.021), religion (P=0.016) and absence of previous abortion/still birth/premature deliveries (P=0.015) were maternal related factors associated with birth of LPN, ETN and FTN. Birth weight was associated with LPN (P<0.001), while FTN had higher likelihood of delayed initiation of breastfeeding (P=0.038) but were less likely to have feeding difficulties compared to LPN and ETN (P=0.012). A comparison of fit model with the complete set of predictors with an intercept-only, or null model revealed that P-values for maternal (P=0.0175) and fetal (P<0.001) related risk factors were less than Fisher’s value of 0.05, hence the null hypothesis was rejected. The Kaplan-Meier survival analysis revealed high survival rates across all gestational age categories, with 100% survival among LPNs. In conclusion, the study showed that maternal- and fetal-related risk factors associated with LPNs and ETNs are distinct from those of FTNs.