Neonatal hypothermia and associated factors within six hours of delivery in eastern part of Ethiopia: a cross-sectional study

This study was held in the capital city of Harari regional state, Harar, Eastern Ethiopia which is 525 km far from Addis Ababa and 48 km away from Dire Dawa city. The institutional delivery coverage of the region was 79%.There were 6 public health institutions that provide delivery service to the community [29]. The public health institutions in the city were first stratified into strata of specialized Hospital, General Hospital and Health center after which one specialized Hospital namely Hiwot Fana Specialized University Hospital, one General Hospital namely Jugal General Hospital and one Health Center namely Jenniela Health Center were selected randomly from each stratum respectively. The study was conducted from January 25/2018 until February 19/2018.

A descriptive institutional based quantitative cross-sectional study design was conducted. All newborns that were within six hours of their delivery at public health institutions of Harar city together with their mothers were included in the study. However, abandoned newborns were excluded of the study because without maternal involvement it wasn’t possible to address most of the variables in the study.

The desired sample was determined by using double population proportion formula with the assumptions of 95% CI, 5% margin of error, 80% power, exposed to unexposed ratio 1:1, 10% non-response rate and an important variable of the study (skin to skin contact(P1): 75% and Didn’t have skin to skin contact(P2): 87%) [22]. The final sample size used for the study was 403. By using systematic random sampling, every other eligible live newborn was included into the study.

Data collection was done by five diploma midwives after they were given one day training of data collection procedure. Data were collected carefully using a pretested and structured interviewer administered questionnaire that contained maternal socio-demographic characteristics and behavioral factors like skin to skin contact, breast feeding, immediate drying, newborn wrapping, cap wearing and warm intra-facility newborn transportation. Moreover, a checklist was used to obtain factors including maternal obstetric complication, neonatal age, birth weight, gestational age at birth, mode of delivery, time of delivery, APGAR score, resuscitation history and neonatal health problem. Axillary temperature of newborns was measured within six hours of delivery for 3 min of duration by using digital thermometer at either delivery or maternity ward.

The data collection tool was adopted and modified from studies conducted in Ethiopia, Nigeria, Tanzania and Uganda which passed through peer review for its validity and published [16‐18, 22, 24]. Moreover, a calibrated thermometer (model of MT-101 MT-111) that had measurement accuracy of ±0.1 °C for the temperature range of (35.5–42.9 °C) and ± 0.2 °C for the temperature range of (32.0–35.5 °C or above 42 °C) was used [24]. The data collection tool was pretested just two weeks prior to the actual data collection using 20 newborns at the study area based on which some modifications were made to the originally prepared tool. One day training and clear orientation was provided for data collectors and supervisors on the process of data collection. During data collection, data collectors were closely monitored and guided by two BSC nurse supervisors for complete and appropriate collection of the data and reporting to the principal investigator was done on a daily basis. The collected data were double entered into Epidata version 4.2 by two data clerks for validation purpose. The entered data were multivariate analyzed for statistical adjustment of possible confounders.

The collected data were cleaned manually, coded and entered into Epi data version 4.2 and exported to STATA Version 12 statistical software for data transformation and further analysis. Descriptive statistics like frequencies, proportion, and summary statistics (mean and standard deviation) were used to describe the study population in relation to relevant variables and presented in tables and graphs. Multi-collinearity between the study variables was diagnosed using standard error and correlation matrix. The assumptions for binary logistic regression model were first checked and then bivariate analysis was carried out to identify candidate variables (p < 0.25) for multivariate analysis. Using these candidate variables, multivariate analysis was performed to investigate statistically significant independent predictors of hypothermia by adjusting for possible confounders. Finally, variables whose p- value less than 0.05 (p < 0.05) from multivariate analysis were declared as statistically significant. Adjusted odds ratio with 95% CI was considered to identify the strength of association between neonatal hypothermia and its predictors.

Ethical approval was obtained from the Institutional Health Research Ethics Review Committee of College of Health and Medical Sciences of Haramaya University. Official letter was obtained from Haramaya University and submitted to each Hospital and health center to get permission from the respective directors of the institutions. Before conducting the interview, an informed and voluntarily signed written consent (thumb print for those unable to write) was obtained from all the eligible mothers of the newborns. The thermometer was disinfected by 70% ethyl alcohol disinfectant with a damp cloth after every measure of axillary temperature of the newborn to prevent infection transmission. Hypothermic premature newborns and those with associated problems were helped for referral to neonatal intensive care unit for better management whereas those mildly and moderately hypothermic ones without medical problem were helped by rehabilitative thermal care measures like proper wrapping, frequent breast feeding, skin to skin contact, closing doors and windows etc. Moreover, mothers were advised of thermal care measures on the way to their home and while at their home.

A total of 403 mothers with their neonates were included in the study making 100% response rate. The mean maternal age was 32 years (SD = ±5.5) and more than half of the mothers (51.9%) were in the age group between 20 and 34 years of age. Two hundred twenty nine (56.8%) were rural residents. One hundred thirty two respondents (32.8%) were unable to read and write and 119(29.5%) of the respondents were farmers. Furthermore, 258 respondents (64%) were multiparous (Table 1).

Three hundred four (75.4%) of the newborns were born to mothers who had antenatal care. Almost all of the newborns, 390(96.8%), were immediately dried. More than half of the newborns, 207 (51.4%), weren’t wrapped properly. Nearly three-fourth of the newborns, 301 (74.7%) weren’t made in contact with their mother’s skin. One hundred eighty seven newborns (46.4%) didn’t wear cap. Breastfeeding was initiated within one hour of birth for 287 newborns (71.2%) and 251(62.3%) newborns weren’t warmly transported from delivery ward to post natal or NICU (Table 2).

It was during the day time that more than half of the newborns, 213(52.9%), were delivered. Furthermore, 257(63.8%) of the newborns were delivered through spontaneous vaginal delivery. Most of the newborns were delivered at the specialized University Hospital (Table 3).

The average neonatal age was 3.37 h (SD = ±1.72) while the average birth weight was 2.94Kg (SD = ±0.63). Moreover, the average gestational age at birth was almost 38 weeks (SD = ±2.52) with an average fifth minute APGAR score of 7 (SD = ±1.60). Male newborns, 202 (50.1%), were almost equal to females, 201(49.9%). Majority of the newborns, 347(86.1%), were born single. One hundred four (25.8%) of the total newborns were born with health problems. From the overall newborns, one hundred nineteen (29.5%) were born to mothers who had obstetric complication during pregnancy or labor and/delivery. From these complications, preeclampsia/eclampsia disorders comprises the highest percentage, 63(52.9%), followed by PROM/Sepsis 23(19.3%), APH, 19(16%), Obstructed/prolonged labor, 12(10.1%) and PPH, 2(1.7%) respectively. Forty (9.9%) of the newborns were given Cardio-Pulmonary Resuscitation after their birth (Table 4).

The prevalence of neonatal hypothermia among newborns who were delivered at public health institutions of Harar city was 267 (66.3%) [95% CI: 61.1, 70.5%)] with mean axillary temperature of 36.24 degree Celsius (SD = ±0.69) (Fig. 1). Majority of the hypothermic newborns, 150(56.2%), were mildly hypothermic whereas the rest 117(43.8%) were moderately hypothermic. The prevalence of hypothermia was highest, 94(90.4%), among newborns that had health problem(s) (Fig. 2). These problems were respiratory distress syndrome 39(37.5%), hypoglycemia 33(31.7%), perinatal asphyxia 25(24.0%), meconium aspiration syndrome 16(15.4%), congenital malformation 14(13.5%), early onset neonatal sepsis 12(11.5%) and jaundice 8(7.7%). Moreover, hypothermia was found to be more prevalent among preterm newborns (86.9%) than term ones (59.5%) as shown from the cross tabulation (Table 5). Regarding newborns with perinatal asphyxia, it was found that all the perinatally asphyxiated newborns were hypothermic which may be due to the fact that these newborns can’t generate heat as they are in scarce of oxygen (hypoxia) [Fig. 3]. From the scatter plot of their exact body temperature, it can also be easily understood that nearly all the perinatally asphyxiated newborns were moderately hypothermic with a median body temperature of 34.1 °C and IQR of 2.35 °C (Table 6).

Newborns who weren’t made in skin to skin contact with their mothers after delivery were 2.9 times more likely to be hypothermic as compared to those who had skin to skin contact (AOR = 2.87,95% CI: 1.48, 5.57). Newborns who didn’t wear cap were 2 times more likely to be hypothermic when compared to those who were dressed with cap (AOR = 2.10, 95% CI: 1.17, 3.76). Newborns who weren’t warmly transported from one unit (delivery) to the other (postnatal unit or NICU) were 3.2 times more likely to be hypothermic when compared to those warmly transported ones (AOR = 3.18, 95% CI: 1.84, 5.48). Those newborns who were born to mothers having obstetric complication were 2.4 times more likely to be hypothermic as compared to those born to mothers without any obstetric complication (AOR = 2.42, 95% CI: 1.28, 4.57). Preterm newborns were 3.4 times more likely to be hypothermic as compared to term ones (AOR = 3.37, 95% CI: 1.53, 7.44). Moreover, newborns with health problems were about 4 times more likely to be hypothermic in relative to those who didn’t have any problem (AOR = 4.24, 95% CI: 1.92, 9.34) (Table 7).

The study lacks support of qualitative data. As this study was conducted only in the summer season, it couldn’t show the significance of seasonal variation for neonatal hypothermia. Moreover, the results may not be representative of the entire newborns in Ethiopia due to a small sample size consideration in this study. Besides, the impact of air temperature of the delivery and postnatal units wasn’t considered due to absence of inbuilt thermostat and humidity control in the studied public health institutions. The study also shares drawbacks of a cross-sectional design.