Neonatal hypothermia prevalence
In our study; the prevalence of neonatal hypothermia at admission was found to be 70.5% (95%CI: 1.89, 3.02). This result is in line with reports from Gondar, Dessie and Harar,Ethiopia where the prevalence of neonatal hypothermia at admission was found to be 69.8%, 66.8% and 66.3%, respectively [9–11]. Studies from Rwanda and Addis Ababa, Ethiopia also reported almost similar prevalence of 65.3% and 64%, respectively [12, 13]. However, the prevalence of neonatal hypothermia in this study was higher than reports from Sidama, Ethiopia, Uganda and Nigeria where the prevalence of Neonatal hypothermia was 51.8%, 51% and 42.2%, respectively [14–16]. In comparison with our study result, a significantly lower prevalence (17.1%) was reported from a Californian study, USA [17]. A prospective study from Kenya reported a significantly higher (73.7%) prevalence than our finding [18].
These significant variations can be explained by the differences is socioeconomic and cultural related factors in which maternal behavioral factors like immediate care as proper wrapping, skin to skin contact, putting cap to the baby, and early initiation of breastfeeding within one hour of birth and weather conditions will have significant effect on the magnitude of neonatal hypothermia. Additionally; differences in study design, study setting and season of the study can have effects on neonatal hypothermia.
Factors associated with Neonatal hypothermia at admission
In this study, there was a statistically significant association between neonatal hypothermia and birth weight below 2500 gram. Neonates with birth weight below 2500gram were 8.3 times (AOR: 8.35; 95%CI: 2.34, 29.80) more likely to develop neonatal hypothermia than those with a birth weight 2500 gram or more. This finding is consistent with reports from Gondar, Rwanda, Arbaminch and California where a neonate being low birth weight is significantly associated with admission hypothermia [9, 12, 14, 18]. It is known that being low birth weight has reduced ability of thermogenesis due to ineffective shivering with low brown fat and also a low birth weight neonate is at risk of higher heat loss related to higher body surface area to weight and thin immature skin.
Head covering/cap wearing was significantly associated with neonatal hypothermia. Newborns who didn’t wear cap at admission were 3 times (AOR: 3.11; 95%CI: 1.16, 8.30) more likely to be hypothermic compared with their counterparts. The same findings are reported from Harar, Sidama, Kenya, West Shewa, and Addis Ababa showing that not having immediate head covering, no immediate skin to skin contact and inadequate clothing were risks of neonatal hypothermia [11, 13, 19–21].
Early bathing within 24hrs of delivery was also observed to be one of the significant contributors of neonatal hypothermia in our study where those neonates who were bathed within 24hrs of life were 4 times (AOR: 3.96; 95%CI: 1.06, 14.78) more likely to develop hypothermia when compared to those who had not been bathed within 24 hour of age. This finding was in line with other studies conducted in sub-Saharan Africa, Sidama and Tigray, [6, 14, 22]. This might occur because bathing newborns within 24 hour of age increases heat loss with evaporation, conduction, and convection due to the effect of the cold water and cold environment. Apart from total body exposure required during bathing, separation from the mother’s body contact can again increase the risk of hypothermia.
Regarding to the mode of delivery, newborns delivered with Cesarean Section (C/S) were 8.5 times (AOR: 8.54; 95%CI: 2.01, 36.39) more likely to develop neonatal hypothermia which is consistent with a study conducted at a tertiary hospital in South Africa where 41% of neonates born via Caesarean section had hypothermia at birth [23].This finding might be due to in most cases emergency Cesarean Section done in mothers came with possible obstetrics complications that may affect neonatal heat production. The other possible explanation might be heat loss during transportation since the Neonatal Intensive Care Unit relatively far from the delivery ward.
In this study, time of delivery was found to have a significant association with neonatal hypothermia. Neonates born during night time were 3 times (AOR: 2.92; 95%CI: 1.29, 6.61) more likely to develop neonatal hypothermia. Similar results were reported from studies conducted in Dessie, Bahirdar, Tigray and East Africa [10, 17, 23, 24]. This might happen due to temperature difference at night and daytime. The cold temperature during night time might increase heat loss and develop hypothermia. Additionally the work overload during night-time is not equal to the daytime that might affect application of warm chain.
Place of delivery was also having significant effect on neonatal hypothermia at admission in this study and neonates delivered out of the hospital were nearly 7 times (AOR: 6.84; 95%CI: 2.12, 22.13) more likely to develop neonatal hypothermia. This finding might be due to the fact that most of the neonates referred with possible neonatal health related problems and suboptimal thermal care practices. Studies from Uganda and Sub-Saharan Africa showed that home delivery was a significant risk of neonatal hypothermia [22, 24].