Comparative Study of Neonatal Hypothermia and its Associated Factors Among Neonates in Rural and, Urban of Shebadino Woreda, Sidama Region, South Ethiopia: A Community-Based Comparative Cross-Sectional Study

Introduction: Hypothermia is one of the major causes of new-born death, particularly in low-income nations. This was due to poor thermal care in most of the rural communities. There is no such previous community based comparative study. Therefore, this study aimed to assess the prevalence and factors associated with neonatal hypothermia among neonates in rural and urban of Shebadino woreda Sidama region, Ethiopia. 

Method: A comparative community-based cross-sectional study design was employed on 682 neonates in Shebadino Woreda, Sidama Region South Ethiopia 2023. A multistage sampling technique was employed, and the collected data were manually cleaned, coded, entered into EpiData version 4.6 before exported to SPSS version 26 software for analysis. Variables with a p-value <0.25 in bi-variable logistic regression were further analysed using multivariable logistic regression. The odds ratio (OR) with 95% CI was used as a measure of association, and variables that had a p-value less than 0.05 in the multivariable logistic regression were considered as significantly associated variables

Result: The overall prevalence of neonatal hypothermia in this study was 51.8% (95% CI: 47.2%-56.3%). The prevalence of neonatal hypothermia was higher among rural neonates, 55.1% when compared to those of urban neonates 48.6%. Bathing before 24 hr. (AOR = 3.64, 95% CI: 1.39, 7.16), Placing a cold object near babies’ head (AOR =2.97, 95% CI: 1.75, 5.03), Neonates who were given traditional medication (Amessa) (AOR=1.83 % CI; 1.04-3.20) and, not separated humans and animals house (AOR =1.75, 95%, 1.05-2.91)  were significantly associated with neonatal hypothermia in rural, while Night time delivery (AOR =1.81, CI: 1.01-5.62), Neonates who were given traditional medication (Amessa) (AOR=3.11 % CI; 1.85-5.21), and  Placing a cold object near babies’ head (AOR=2.40, 95% CI: 1.37, 3.29¬ were significantly associated with neonatal hypothermia among urban neonates

Conclusion: The prevalence of neonatal hypothermia in the study area was relatively high in rural then urban. Therefore, attention is needed for strict adherence to cost-effective thermal care such as warming the room, separating humans and animals house, teaching not to put cold objects near babies, giving special care for new-borns for those delivered from women with medical problems and giving priority for those delivered at night.

Neonatal hypothermia is defined by the World Health Organization (WHO) as a body temperature of 36.5°C (97.7°F) or below before the age of 28 days. If the axillary temperature is 36.0°C to 36.5°C, it is mild hypothermia; 32.0°C to 36.0°C is moderate hypothermia; and 32.0°C is severe hypothermia. New-borns are vulnerable to environmental interferences as they face to life outside the womb due to their systemic immaturity (1). Heat generation does not exceed heat loss as new-born body temperature drops from 2°C to 3°C in the first half-hour of life (1,2).The heat can be lost through evaporation (due to the evaporation of amniotic fluid from the skin surface), conduction by coming into contact with cold objects-cloth, etc ), convection (by air currents in which cold air replaces warm air around baby-open windows), and radiation [1].

Warm delivery room, immediate drying, skin-to-skin contact, breastfeeding, postponing weighing and bathing, appropriate clothing and bedding, keeping mother and baby together, warm transportation, warm resuscitation, training, and awareness-raising are the ten interlinked warm chain preventive mechanisms recommended by WHO [1]. On the contrary, hypothermia is still a major problem in poor countries, particularly in Sub-Saharan Africa [3]. Poverty, home delivery, low birth weight, early bathing of babies, delayed commencement of breastfeeding, and limited understanding among health personnel are all risk factors for neonatal hypothermia in the region [4].

Hypothermia plays a significant role in neonatal deaths, with global case fatality rates (CFR) ranging from 8.5% to 52% [2]. Mortality increased by approximately 80%, for every degree Celsius decrease in the first observed axillary temperature. The relative risk of death ranged from 2 to 30 times within the current WHO classification for moderate hypothermia, increasing with greater severity of hypothermia [3]. Around 20% of deaths are caused by prematurity, however,  better thermal care could avoid 10% of mortality in term newborns [4]. Its presence in most morbidity cases is more evident, as it is related to peri-intraventricular hemorrhage grade 3 and death [5]. According to recent studies, the death rate increases by five times when the temperature of newborns drops by one degree Celsius, and every one-degree drop in body temperature raises mortality by 80 percent [6]. It is  particularly common at home in low-income countries with weak health systems, where even health professionals have a poor understanding of hypothermia as a risk to the newborn [2].

In Sub-Saharan Africa, many births and deaths occur at home, which lacks vital records. On the contrary, there is little community-based research in the region. As a result, the actual incidence of newborn hypothermia in the region is unknown [7,8]. Due to this fact, the neonatal mortality caused by hypothermia may exceed the estimated based on complete data [3]. Additionally, the temperature is not taken, in most newborns, immediately after birth. This implies additional research should be done using a better design and methodological sound of community-based studies to understand the problem in Eastern African settings [9].Therefore the purpose of this study is to determine the prevalence and the factors associated with it in rural and urban of Shebadino Woreda Sidama region, southern Ethiopia.

Study settings and design

This is a community based comparative cross-sectional study conducted from June 24 to November 24, 2022, the research was carried out in Shebadino woreda, a territory of the Sidama National Regional State. Its elevation ranges from 1750 to 3000 meters above sea level. Shebadino woreda has an annual average rainfall of 1200 mm, with maximum and minimum values of 1600 mm and 800 mm, respectively. According to the Regional and Woreda metrology report, the average annual temperature is 28.5 °C, with hot days reaching a maximum of 31 °C in February and March and cold days reaching 21 °C and 22 °C in July and August, respectively. This region has 86 percent midland, 14 percent highland, and nearly no lowland climate zones (13). According to studies (14) pneumonia is the most common health condition in the woreda.

Population

Study population were neonates with their mothers in selected kebeles of Shebadino woreda, during the actual data collection period. However, neonate’s mothers who were stayed less than six months in the woreda were excluded from the study. Preterm neonates and newborns whose mother’s suffering from critical illness (postpartum hemorrhage) and Postpartum Psychosis were also excluded.

Sample size 

The sample size was estimated using double proportion formula; considering the proportion of neonatal hypothermia (rural 53% and Urban 40%) from previous study conducted in Arbaminch General Hospital [10]. The formula and calculation as follow:

N (in each group) = (p1q1 + p2q2) (f (α, β)) / ((p1—p2) 2

Where n = sample size for each group

P1 = proportion of neonatal hypothermia (rural 53%)

P2 = the proportion of neonatal hypothermia (40%)

F (α, β) = 7.84, when the power = 80% and the level of significance = 5%

q1 = (1-p1) = 1–0.53 = 0.47

q2 = (1-p2) = 1–0.4 = 0.6

n1 = n2 = (0.84+1.96)2 ((.47×.53) + (.4×.6)) / (.53 - .4)2 = 227 

Total sample size for both groups = 454

Multiplied by design effect of 1.5 i.e., 454*1.5 =681, by adding 5% none response rate, the final estimated total sample size for this study was 716 (358 and 358 study subjects for the rural and urban) 

 sampling procedure

There are 30 kebeles in the district, which means there are 25 rural kebeles and 5 urban kebeles [13]. A simple random selection procedure was used to select 40% (10 rural Kebeles) of all kebeles (the smallest administrative entities in an Ethiopian district). In six months, 2154 neonates were born in 10 rural kebeles, whereas 632 neonates were born in three urban kebeles. The determined sample size was proportional to the population size in each selected kebele. To identify study individuals, a multi-stage sampling procedure was used. Then participants were selected by using a systematic random sampling technique in order of birth registration from HEW, that is, every two birth reports until the required sample size was obtained for urban (K = 1.76 approximately every 2 neonate birth reports were taken, whereas the K value for rural is 6. Households of the women who gave birth were identified and reached with the help of the HEW and Health development army of Selected Kebeles.

Operational definitions

Non-hypothermic: an axillary neonatal temperature measurement of ≥36.5°C at the time the data collector is arriving home [11]. Hypothermic:  an axillary neonatal temperature measurement of less than 36.5 °C at the time the data collector is arriving home [11]. Hypothermic:  an axillary neonatal temperature measurement of less than 36.5°C at the time the data collector is arriving home.

Data collection tools and procedures

Six BSc midwives and two MSc supervisors were used for data collection and measurements in the mothers after home delivery. All data collectors and supervisors are chosen based on their prior data collection experience and their proficiency in Sidama affo and English in reading, writing, speaking, and understanding. The data collector went into every home and gathered data after each birth with the aid of HEW and the health development army. Data were collected for a mother who gave birth in a medical facility as soon as the mother was taken home.

The study done in Addis Abeba and the Arbminch General Hospital was refined and utilized as the basis for a structured, interviewer-administered questionnaire [10,12]. The interview questionnaire was written in English, translated into the regional Sidamo affo language, and then, in order to ensure uniformity, translated back into English. The personnel responsible for supervising the data collection was trained to measure neonatal temperature, baby weight, room temperature, and calculate gestational age. During the study visit, the temperature was taken using a digital thermometer at the axilla until an audible beep was heard automatically. These are the ones we picked since they are accessible locally, affordable, and simple for community workers to utilize. Additionally, we had utilized axillary measurements rather than rectal measurements because they were simpler, safer, and more socially acceptable [13].

The axillary temperature of the newborn was measured, as soon as arrival, by a digital thermometer (model- MT-101) that has a measurement accuracy of ±0.1°C for the temperature range from 35.5°C–42.0°C, and ±0.2°C for the temperature range of 32.0°C–35.5°C or above 42.0°C (12,14). The room temperature was also measured by Gera Mercury thermometer with a measurement range of (− 37°C to 356°C) and measurement accuracy of ±3 °C.

 As soon as we arrived home, we positioned the mercury thermometer more than 2 feet above the ground in the middle of the room and waited at least 15 minutes for the temperature reading. With an emphasis on reducing the amount of time the babies may be exposed to the cold, the temperature of the baby was measured before the baby's weight was taken by a professional data collector. To ensure high reliability, two measurements in degrees Celsius were made repeatedly at the same time, and the average of these two values was used. The gestational age was calculated based on the woman's account of her most recent period and an early ultrasound. The weight was measured using a weighing scale model called the RGZ 20 that was accurate to within 50 grammes [23] of the measurement. The oxygen saturation and pulse rates were determined using a pulse ox meter.

Data quality assurance

In the Arbegona woredas, which had a setting similar to the study site, the pretest of the data collection was conducted on 36 neonates, or 5% of the sample size. This pretest was created to determine how long it would take, whether the answers were accurate, the language was clear, and the data collection tool was adequate. In-depth training covering the research objective, the list of eligible study participants, the tools and procedures for collecting data, and interviewing techniques was provided to data collectors over the course of two days. The principal investigator, data collectors, and supervisors verified the accuracy of the data each day before entering it. Non-respondents were also included in the total for any surveys that were. 

Data processing and analysis

We discussed maintaining participant confidentiality throughout the entire data collection procedure during the training. To assure the consistency and thoroughness of the surveys, the entire process was carefully observed. Data was originally entered into Epi Data version 4.6 before being exported to SPSS (Statistical Package for Social Sciences, version 26) to be analyzed. Using descriptive statistics like frequencies, proportions, and summary statistics (mean and standard deviation), the study population was described in relation to relevant factors and then presented in tables and graphs. 

Bi-variable analysis was used to discover candidate variables (p less than 0.25) for multivariate analysis after assumptions including dichotomous, multi-co linearity issue, Chi-square test, and mutual exclusivity were first validated. In order to control the confounder, multivariable logistic regression was used to further analyses variables that had a p-value of less than 0.25 in the bi-variable analysis. As a measure of association, the odds ratio (OR) with 95% confidence interval (CI) was utilized, and factors with a p-value in the multivariable logistic regression of less than 0.05 were regarded as significantly associated variables. The goodness of fit, was tested by Hosmer and Lemes how which was a sig = 0.95. Variance inflation factors (VIF), which should be less than 10, were used to assess the multi-co linearity. A histogram and Q-Q plot test were used to determine the normality of the data.

Socio‑demographic characteristics of respondents

A total of 682 mothers with neonates (49.3% from rural and 50.7% from urban) participated in the study with 95.25% response rate. The majority of the participants (47.3 % from rural and 48.8 % from urban) were in the age group of 20-30 years with a mean age of 26. 75, and 25.43 respectively. 91% of rural, and 86% of urban participants were married, whereas 65% of rural and, and 50% of urban were Protestant religious followers. 50.4% of women in Rural and, 32.7 from urban were house wife regarding there occupations (Table 1).

Table 1: Socio-demographic characteristics of neonate’s parents in Shebadino wereda, Sidama, Ethiopia 20222.

Obstetric characteristics of the mothers

Two hundred forty (71.4%) of the mothers from rural and three hundred thirty-three (96.2%) of the mothers from urban  had visited health facilities for antenatal care (ANC) during the recent pregnancy at least one time. 175 (52.1%) of the mothers from rural and, 53 (15.3%) of the mothers from urban were given birth at home. 166 (77%) from rural and, 185 (75%) from urban were delivered at >=37 completed weeks. About 268 (79.4%) of the neonates from rural and 272 (78.6%) from urban were delivered single (Table 2).

Table 2: Obstetric characteristics of the Neonates mothers in Shebadino woreda, Sidama region, Ethiopia 2022 (n=682).

Behavioural and Neonatal Factors

One hundred seventy-one (50.9%) from rural and, one hundred eleven (32.7%) of the Neonates from urban were bathed within 24hr after delivery. A majority, 142 (42.3%) of the Neonates from rural and, 234 (67.6%) from urban were placed on the mother’s abdomen (Skin to skin contact) after delivery. Traditional medication (Amessa) was applied to 195 (58%) of the neonate from rural and, 204 (60.7%) of the neonate from urban. Of the total participants, 197 (57.7%) rural and, 217(62.7%) of the neonate were greater than 7 days of age with a mean age of 11.21 and 9.35 days respectively (Table 3). 

Table 3: Behavioral and Neonatal Factors among neonates in Shebadino woreda, Sidama, South Ethiopia 2022 (n =682).

Environmental factors

The majority, one hundred ninety-five (58%) of rural and two hundred fifteen (62.1%) of urban neonates were delivered at night respectively. About 176 (52.4%) of rural neonates' and 137(40.2%) of urban neonate’s families placed cold objects or metal nearby the bed of the baby whereas, 131(39%) and 217(62.7%) of rural and urban neonates' mothers are non-hypothermic (Table 4).

Table 4: Environmental conditions assessed during the time of data collection from neonates in shebadino woreda, Sidama region, South Ethiopia 2022 n= (682).

Prevalence of neonatal hypothermia

The overall prevalence of neonatal hypothermia in this study was 51.8% (95% CI: 47.2%-56.3%). The prevalence of neonatal hypothermia was higher among rural neonates, 55.1% when compared to those of urban neonates 48.6%. The mean axillary temperature was 36.25°C (SD ±1.24) and, 36.54°C (SD ±1.51) for rural and urban respectively. 

Region, south Ethiopia 2022 (n=583

Factors associated with neonatal hypothermia

Factors associated with neonatal hypothermia among neonates in rural Kebeles 

Factors that were found to be significantly associated with neonatal hypothermia among neonates from rural in the bivariable analysis were, Not warming room, no history of ANC follow up, Age of neonate, giving baby food by mouth, applying traditional medicine on neonate, Bathing baby before 24hr, putting cold object near babies’ bed, Not separating animals and humans house and mother's body temperature. In multivariable logistic regression, four of them were found to be statistically significant at a p-value of less than 0.05. Those neonates, who taken traditional medication by mouth were 1.83 times more likely to be hypothermic when compared to those who were on exclusive breast feeding (AOR = 1.83, 95% CI: 1.45–3.24). Additionally, neonates who were slept near a metal or other cold Object were 2.64 more likely to be hypothermic when compared to those who were not (AOR = 2.97,95% CI: 1.75–5.03). Those neonates, who were slept in room where humans and animals house did not separate were 2.3 times more likely to be develop hypothermia related to those, whose house were separated (AOR = 2.3, 95% CI:1.67–3.52). Neonates who were bathed within 24hr were 3.64 times more likely to develop hypothermia when compared to their counterparts (AOR = 3.64, 95% CI:1.39–7.16) Table 5.

Table 5: Bi-variable and multivariable logistic regression model with Cross tabulation for factors associated with Neonatal Hypothermia Among rural neonates in Shebadino woreda, Sidama Region, South Ethiopia 2022 (n=336)

Factors associated with neonatal hypothermia among neonates in Urban kebeles

Factors such as, number of delivered neonate, time of delivery, placing cold object near baby’s, Covering Baby with dry cloth, Previous NICU admission history, and applying traditional medicine to neonate were found to be significantly associated with neonatal hypothermia among neonates from urban in the bivariable analysis at a P value less than 0.25. In multivariable logistic regression, three variables were found to be statistically significant at a p-value less than 0.25. In multivariable logistic regression, three variables were found to be statistically significant at a p-value of less than 0.05. Among them, neonates who were slept near a cold object (metal), were 2.01 times more likely to develop hypothermia, when compared to those who were not (AOR = 2.01, 95% CI:1.59–4.38). Additionally, Neonates who were started taking amessa for traditional medication were 3.11 times more likely to develop hypothermia then, their counter parts (AOR = 3.11, 95% CI:1.85–5.21). Neonates who were delivered at night time were 1.81 times more likely to develop Hypothermia than those delivered during the daytime (AOR =1.81, 95% CI: 1.01-5.02) Table 6.

Table 6: Bi-variable and multivariable logistic regression model with Cross tabulation for factors associated with Neonatal Hypothermia among urban neonates in Shebadino woreda, Sidama Region, South Ethiopia 2022 (n=346)

Overall Factors Associated with neonatal hypothermia among neonates in Shebedino Woreda Sidama region South Ethiopia. Babys who were previously admitted to NICU were 3.88 times more likely to develop hypothermia when compared to those who were not (AOR =3.88, 95% CI:1.52 -8.37)). Neonates, whose mother had obstetrical complication(s) during pregnancy/ labor were 2.38 times more likely to develop hypothermia, when compared to their counterparts (AOR = 2.38 95% CI: 1.05-5.14). Putting a cold object (metal) near a baby’s bed were 4.20 times more likely to develop hypothermia when compared to its counter parts (AOR =4.20, 95% CI: 2.24-7.71) Table 7.

Table 7: Bi-variable and multivariable logistic regression model with Cross tabulation for factors associated with Neonatal Hypothermia among neonates in Shebadino woreda, Sidama Region, South Ethiopia 2022 (n=682)

The overall prevalence of neonatal hypothermia in this study was 51.8% (95% CI: 47.2%-56.3%). Depending on residence, the prevalence neonatal hypothermia was 55.1% among rural neonates and 48.6 % among urban neonates which shows there was a significant difference in prevalence of neonatal hypothermia among neonates in shebedino woreda (p=0.003). This finding revealed that, there was a significant higher neonatal hypothermia in rural neonates then urban neonates. This difference may be due to the lack of thermal care among rural neonates, when compared to those of urban. The  overall prevalence was comparable with the study findings in Islamic Republic of Iran 53% [15], and Northern Uganda 51% [16]. Another study conducted in Pakistan in revealed that, the prevalence of neonatal hypothermia was 49.5% [17] which is in line with our study. Additionally, this overall finding is in line with study conducted in Arbaminch General hospital 50.3% [10] This similarity may be due to the study sites and the large sample size used. However, the overall prevalence finding of our study was higher than in another community-based study conducted in India 45% [3]. This variation might be due to seasonal conditions, data collection tools, differences in temperature measurement sites, and economic and cultural differences in those communities. Additionally, the overall prevalence was higher than another study conducted among home-delivered neonates in North India 11% [18]. This was due to Hypothermia definition variation as the author of that research defined neonatal hypothermia as a temperature less than 35.6°C whereas we defined it based on WHO definition recommendation definition [11].

This study's finding is lower than other studies conducted at the hospital with 77% on admission to neonatal intensive care Unit in tertiary Hospital in Malawi [19], 69.8% in Gonder teaching and referral Hospital [20], and 64% in Addis Ababa Hospital [12]. The possible justification for this difference is due to, neonates who were admitted to NICU were for different indications, which could decrease their ability to adapt to the external environment out of the womb and easily develop hypothermia. On the other hand, unlike those studies, late neonates were included in our study in which neonates can resist heat loss as their age increases and easily defend against hypothermia [12,21]. From rural, neonates who were bathed within 24hr after delivery, were 3.64 times more likely to develop hypothermia when compared to their counterparts. This finding was supported by study conducted in South Ethiopia [10] and, Central zone of Tigray [22], Uganda [23]. The possible justification is, due to the effect of the cold water and, exposing the neonates to the cold environment, neonates can easily expose and their thermal temperature may decrease. Additionally, at the time of bathing, a neonate must be separated from the mother’s body contact which can again increase the risk of hypothermia.

Another factor which has been associated with neonatal hypothermia, among neonates in both rural and urban were giving amessa for traditional medication. Those neonates who were started drinking amessa for traditional medication were 1.83 and 3.11 times more likely to develop hypothermia for rural and urban respectively, when compared to those who were on exclusive breast feeding. This medication was given as people of this woreda belief this medication protects neonates against the eyes of devil. But it resulted in hypothermia as it interferes with exclusive breast feeding. The more the neonate breast feed, the more they would have adequate glucose to cope up with their energy expenditure. This is due to; breast milks are full of different vitamins and Calorie. Additionally, there is skin to skin contact during breast feeding and neonates can share a mother body heat and take advantageous over their counterparts [20,24].There was no previous study finding with the same variables to discuss our finding.

Sleeping near cold object (metal) was significantly associated with hypothermia among neonates in rural, urban, and overall, in this study by 1.75, 2.01.and 4.2 times more likely related to their Counterparts parts respectively. This object was placed at the head of the bed near the baby's head. This metal placement may be due to people's Spiritual beliefs and lack of awareness of environmental thermal care in rural community. On the Contrary, the neonate lost heat through conduction (neonates body contact with cold objects) and radiation (Loss of heat to the cold metal surface even if not in contact). By this mechanism, the neonate may lose internal heat and result in hypothermia [1,11]. As there was no other study done on it previously, it was difficult to compare our result with the other for discussion.

Neonates who were slept in a house where a humans and animals house were not separated were 2.3 times more likely to develop hypothermia related to those, whose house were separated among neonates in rural. This may be due to animals made wetted environment with high humidity and, result in dropped room temperature. It may be resulted due to heat loss in the cold environment by radiation from the infant to a cooler environment [1]. Additionally, those neonates were easily exposed to neonatal disease, which again result in neonatal hypothermia. This variable is also new finding and, we couldn’t able to discuss it. The current study found there was a significant association between time of delivery and neonatal hypothermia among neonates in urban. The possible justification for this may be due to, temperature difference at night and daytime. Additionally, there is no added warmth during cold nights, and the new-born is at risk to lose heat and, develop hypothermia [1,21]. On the other hand, the work overload during night-time is not equal to the daytime, for those neonates who were delivered at the health institution. This finding is in line with another study conducted in Public Hospital in Addis Ababa [12], in Dessie Referral Hospital [29], Northwest Ethiopia [20]. The possible justification for this similarity may be due to the majority, 60.1% of new-borns were delivered at night in our study, which is almost the same as the above study. For example; in the Dessie referral hospital [25] and in Northwest Ethiopia 69.8% were delivered at night [20]. 

The overall study revealed, neonates who had resuscitation at birth were 3.88 times more likely to be hypothermic when compared to those who had no resuscitation. This may be due to neonates who need resuscitation are those who had birth asphyxia. For those neonates there is no enough oxygen which is needed for mitochondrial oxidation in the brown adipose tissue, for heat production. Additionally, during resuscitation thermal care may not be properly done without wrapping the baby in cold table. This finding is supported by study conducted in Gonder University teaching hospital [20], Dessie referral hospital [25], study done in Bangladesh [26], and a study done in Iran [27]. Another significant factor in overall analysis was obstetrical complications like premature rupture of membranes, Hypertension, DM and antepartum haemorrhage during pregnancy and/or labour. Those women need special managements like caesarean section or instrumental deliveries. Due to this, the mother might be too sick to be with their neonates in close contact during the time of delivery, especially for skin-to-skin contact. Additionally, neonates delivered after premature rupture of membrane might be develop neonatal sepsis and fall in body temperature. This finding is supported by research conducted in Southern [10], and Eastern Ethiopia [14]. This similarity may be due to the socioeconomic status of the society across the country, which leads to obstetric complications and may end up in neonatal hypothermia. 

With our maximum search engine, this is the first purely community-based assessment of neonatal hypothermia in Ethiopia and the Second in sub-Saharan Africa after a study conducted in Northern Uganda 2021 [16]. We found new variables un like that was previously done at hospital level. Additionally, the obtained finding is generalizable to all neonates in the woreda including home births.

Digital thermometers might slightly over or underestimate temperature readings as compared with mercury thermometers. We have used it as it is easily available and good for field study unlike the mercury thermometer [28]. Measurement of temperature was only based on twice the measurement record at the same time. The instrument used by the person who took the measurement, the site, and the time of measurement taken might not be similar for all neonates, which may bias the result. Our study was done in one season, and considerations such as seasonal variations were not taken into account. On the other hand, hospital-related characteristics Such as; the qualifications of healthcare personnel working in delivery rooms and NICUs were not taken into consideration as they may have been related to our dependent variable. The other limitation of our finding was recalling bias. To decrease this possibility proper definition and articulation of the research questions, and administering the interview properly and consistently was done. The outcome of the neonate was unknown including those referred to health institutions.

The overall prevalence of neonatal hypothermia in this study was 51.8% (95% CI: 47.2%-56.3%). The prevalence of neonatal hypothermia was higher among rural neonates, 55.1% when compared to those of urban neonates 48.6%. Among the factors, placing a cold Object (Metal) near a neonate's bed, Maternal obstetric problems during pregnancy, and taking amessa for traditional medication are the important factors contributing to neonatal hypothermia

Based on our study findings, the following public health measures were recommended

For woreda Health Sector management.

It is better to give periodic training for HEW on cost-effective thermal care. This is to end up with good awareness, knowledge, and skills of HEWs to endeavour prevention mechanisms such as; room warming, neonatal wrapping (head covering), Continues skin- skin contact, separating humans and animals house, and separating a cold object from a neonate's bed. Again, society teaching on traditional practice done on neonate is needed.

Proper counselling should be required before discharge to home for those born to mothers with obstetric complications. Priority should be given to night-time delivery room man power. Additionally, counselling for cost-effective thermal care such as a warm environment where neonates can sleep, and endeavouring pregnant women for hospital delivery for proper thermal care at health institutions is recommended. Counselling on birth preparedness during ANC for neonates covering material should also be mandatory. 

Its good if NGO working in this area are alert and run for solutions to prevent hypothermia in the woreda based on our study findings and address the gaps. Future researchers; Researchers should focus on Qualitative and Prospective cohort study design in a different season to address some factors like seasonal variation.

AOR; Adjusted odd ratio, CI; Confidence interval, CBNC; Community based neonatal care CFR; Case fatality rate COR; Crude odd ratio, EDHS; Ethiopia demographic health survey EMDHS; Ethiopia Mini demographic health survey ENC; Essential new-born care, HEW      Health extension worker, IRB; Institutional review board KMC, Kangaroo Mother care, MDG Millennium Development goals, WHO; World Health organization.

Ethics approval and consent to participate; Ethical clearance was obtained from the institutional review board (IRB) of Hawassa university college of Medicine and Health Sciences with a reference number of IRB/193/14; Date:21/06/2023. After the letter of permission was obtained, the letter was taken to the head of Shebadino Woreda health office, and consent was obtained from the Woreda health officer and then from the head of each health posts catchments. At the time of data collection respondents were informed about the purpose of the study and informed written Consent was obtained from the study participants. The data for this study was collected following the declaration of Helsinki. We confirm that all methods were performed per the relevant guidelines and regulations by including a statement in the “ethics approval and consent to participate” section under ‘Declarations’ to this effect.

Consent for Publication 

Not applicable

Availability of data and materials

The dataset used/or analysed during the current study are not publicly available. Because we did not have consent from all participants to publish raw data, but are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests. 

Funding

There was no fund received for this research 

Authors’ contributions

Mr. Gizu Tola developed the draft of proposal and performed the statistical analysis and result write-up under the supervision of Mr. Shambel Negesse. Mr. Dagne Deresa, Bashatu Berkessa and Shimalis Tadesse were participated in manuscript preparation. All authors made a significant contribution to the conception and conceptualization of the study. All authors read and approved the final manuscript.

Acknowledgements

First, we would like to thank Hawassa University, the college of medicine, and the health science department of midwifery for giving us a chance to conducts this research. Next, our gratitude goes to Hawassa university comprehensive specialize Hospital for providing us with the necessary materials for data collection. Additionally, I would like to thanks my lovely friend Mrs. Konjit Habtamu for her being my rock, my confidante, and my best friend.