The extent of Knowledge and practice toward neonatal resuscitation among nurses and midwives in public hospitals of South Wollo, northeast Ethiopia: Cross-sectional study

Background: The neonatal period is the most vulnerable time for child morbidity and mortality. Asphyxia due to poor resuscitation techniques contributed significantly to this vulnerability. Therefore, this study is aimed to assess the extent of knowledge and practice towards neonatal resuscitation among nurses and midwives in public hospitals of south Wollo northeast Ethiopia. Method: Institutional based cross-sectional study design was employed among 143 study participants selected by a simple random sampling method from each hospital. Self-administered questionnaire and interview-based guide were used to collect data. Data were coded and entered into EPI data software version 3.1 and was exported into SPSS version 20 for analysis. Logistic regression with backward LR method was performed to see the possible associations of factors with the outcome variables. Finally, p-values of less than 0.05 in multivariate analysis were declared having a significant association with the outcome variable. Result: One hundred and forty-three participants were included with a response rate of 100%. Only 32.9% and 24.5% of the participants had good knowledge and practice toward neonatal resuscitation respectively. After adjustment was made for covariates; lack of training (AOR: 3.44, 95% CI:1.54-7.68), absence of guideline (AOR: 3.8, 95% CI: 1.68-8.58) and lack of supportive supervision (AOR: 2.49, 95% CI:1.1-5.6) were significantly associated with poor knowledge score, whereas absence of guidelines (AOR: 2.83, 95% CI:1.12-7.15) and lack of supportive supervision (AOR: 5.28, 95% CI: 1.86-14.96) were significantly associated with the poor practice of the participants. Specialization with neonatology (AOR: 0.19, 95% CI: 0.047-0.8) was associated with good practice level of neonatal resuscitation.

Conclusion: Knowledge and practice of nurses and midwives toward neonatal resuscitation were suboptimal. The training was found to be a single predictor for knowledge while supportive supervision and availability of resuscitation guidelines were predictors for both knowledge and practice toward neonatal resuscitation. Therefore, regular training and supportive supervision need to be strengthened to fill the identified gaps. Other observational and qualitative studies are recommended to explore factors that contributed to poor knowledge and practice toward neonatal resuscitation.

Background
The first 28 days of life is called the neonatal period and incontrovertibly, it is the most vulnerable and high-risk time in life because of the highest mortality and morbidity that occurred in this period [1][2][3]. The vast majority of these deaths took place in developing countries and approximately 70% occurs in South-east Asia and Sub-Saharan Africa (SSA) [4,5]. Sub-Saharan African countries (SSA) are the region of home to most of the highest mortality countries in the world with 1 child in 13 dying before the first birthday compared to 1 in 189 in high-income countries [3,4].
Asphyxia due to poor resuscitation techniques had remained one of the leading causes of neonatal mortality and morbidity globally [3,[6][7][8]. According to the world health organization (WHO) report of 2012, about one-quarter of the global neonatal deaths were caused by asphyxia [2,9]. Furthermore, the 2013 Report of Global Development Alliance (GDA)showed thatout of 139 million newborns 17 million babies need help to breathe with resuscitation and close to 700,000 were died from asphyxia [10]. Asphyxia could also result in the short term and long term neurological disorders among survivals [11]. The burden of asphyxia is disproportionately concentrated in low and middle-income countries (LMICs) with the highest percentage in Sub Saharan countries (SSA) [11,12]. Studies in Nigeria and Zambia indicated that asphyxia had contributed to a greater proportion of neonatal mortality [11,14]. According to the 2011 Ethiopian Demographic Health Survey (EDHS), asphyxia is the second most common cause of neonatal mortality in Ethiopia [15]. A study in southwestern and northern Ethiopia indicated 47.5% and 31% of neonatal mortality was attributed to asphyxia respectively [7,8].
Despite the severity of asphyxia on neonatal mortality and morbidity, competency toward neonatal resuscitation had remained a significant challenge [16,17]. Many neonates in developing countries were died unnecessarily due to poor resuscitation techniques [18]. Studies conducted in Sub Saharan Countries had revealed that nurses and midwives had a considerable knowledge and skill gap in all areas of resuscitations [19][20][21][22]. However, studies conducted to date to identify barriers contributing to poor resuscitation had remained scarce in low resource countries including Ethiopia [24,25]. Therefore, this study was aimed to assess the extent of knowledge and practice toward neonatal resuscitation among nurses and midwives in public hospitals of northeast Ethiopia.

Study setting and study participants
The institution-based cross-sectional study design was employed among 143 nurses and midwives working at delivery and NICU unit from 1 st February to 30 th April 2018 in public hospitals of south Wollo northeast Ethiopia. All nurses and midwives from neonatal intensive care unit and delivery units of each hospital during the study period were included; whereas nurses and midwives who were seriously sick and on leave allowance were excluded. The study participants were proportionally allocated to each study hospital and then to their working units (NICU and delivery) based on the number of their source populations. Study participants were recruited by a simple random sampling technique using the list of nurses and midwives from the salary payroll of each hospital as a sampling frame.
Data processing and management Data collection tool was prepared from similar kinds of literature, WHO guidelines and Ethiopian Pediatric Association Guidelines and Training manuals. The tool consists of a structured self-administered questionnaire addressing provider and institutional characteristics; structured questions addressing the knowledge of nurses and midwives and interview-based guide addressing the practice level of the participant. The tool was validated by 8 experts and its content validity index was 0.89 (CVI = 0.89) which is greater than 0.78, the cut-off point for the content validity index. Before the actual data collection time, the tool was pre-tested in a separate hospital and the necessary changes were made. The three-day training was given for data collectors and supervisors regarding the data collection instrument and procedure. Trained data collectors distribute selfadministeredquestionnaires to the participants in the respectivedepartments and then interviewed the participants for their level of practice after collecting selfadministered questionnaires. Completeness of the data was checked by supervisors and principal investigators daily. Data were verified, coded and entered into Epi-Data Software version 3.1 and was exported into SPSS version 20 Software for analysis. Results were described using descriptive statistics like frequency, percentage and mean using tables, graphs, and texts. To measure the possible association of factors with the outcome variables binary logistic regression using a backward method was performed and factors with p-values of less than 0.05 in multivariate analysis were considered statistically significant. respectively with an inter-quartile range of 7(25.93%). Out of 92 midwives, more than two-thirds of 69.6% had poor knowledge score similarly out of 30 BSC nurses two-third (66.7%) had poor knowledge score toward neonatal resuscitation. From a total of 13 neonatal nurses, more than half 53.8% had poor knowledge scores similarly from a total of 8 pediatric nurses more than half (62.5%) had poor knowledge scores toward neonatal resuscitation. In terms of educational status from a total of 110-degree holders, more than two-thirds (68.2%) had poor knowledge scores towards neonatal resuscitation (Table 1) Nurses and midwives who had no neonatal resuscitation guidelines in the workplace had almost 4 times less knowledge than those who had a guideline. Participants with no supportive supervision had 2 times less knowledge than those who got supportive supervision (Table 3).

The practice of the participant toward neonatal resuscitation
This study revealed that the majority of the participants (75.5%) had poor practice towards neonatal resuscitation. The minimum and maximum practice scores of the participants about neonatal resuscitation were 8(26.67%) and 27(90%) respectively with an inter-quartile range of 7(23.33%). In this study from a total of 92 midwives, the majority (81.5%) had poor practice similarly from a total of 30 BSC nurses, the majority (70%) had poor practice towards neonatal resuscitation. From a total of 13 neonatal nurses nearly half (53.8%) had poor practice similarly from a total of 8 pediatric nurses, two-thirds (62.5%) had poor practice toward neonatal resuscitation. Among 110-degree holders, 75.5% had poor practice toward neonatal resuscitation. Out of 43 participants who had a poor attitude towards neonatal resuscitation majority (88.4%) had poor practice towards neonatal resuscitation. In terms of neonatal resuscitation training out of a total of 78 participants with no neonatal resuscitation training majority (83.3%) had poor practice toward neonatal resuscitation (Table 5). In this study, the majority of the participant in all governmental hospitals had poor practice towards neonatal resuscitation. About  (Table 5).

Discussion
In this study the overall knowledge and practice of comprehensive nurses, pediatric nurses, neonatal nurses and midwives toward neonatal resuscitation were sub standardize. The overall mean knowledge and practice score of the participants was 68.56% (SD = 16.37%) and 62.96% (SD = 15.89%) respectively. This study had revealed that only 32.9% and 24.5% of the participants had good knowledge and practice towards neonatal resuscitation respectively.
The overall mean knowledge score of the participants was 68.56% (SD = 16.37%).
This score was higher than the study from Gondar teaching hospital (50%) [28]. The difference in the scores might be due to the difference in the study period and study setting. From this study, it was found that the mean knowledge score of nurses was insufficient (64.69%). This finding was higher than the study from India (57%) and another study from Gondar teaching hospital (43.9%) [18,19]. The discrepancy could be due to the difference in the quality of training on neonatal resuscitation and the availability of resuscitation materials. The mean knowledge score of midwives was 68.68% (SD = 16.72%)). This finding was higher than the study from Gondar teaching hospital (42.8%) [19]. The discrepancy could be due to the difference in the study period and the study setting.
This study showed only 32.9% of the participant had good knowledge (a score greater than 80%) toward neonatal resuscitation. This finding was in line with the study from Kenya (only 35.4% score greater than 85%), a study from India (only 34% score greater than 85%) and another similar study from India (only 33.33% had good knowledge) [20,29,30]. The overall mean practice score of the nurses and midwives was 62.96%(SD = 15.89%). This finding was in sharp comparison with the study from Gondar teaching hospital 56.7% [19].
In this study, 75.5% of the participant had poor practice toward neonatal resuscitation. The finding was in sharp contrast to the study from Iran (84.6%) [31].
The discrepancy might be due to the difference in the quality of neonatal resuscitation training, adequate exposure to real resuscitation cases and hospital set up for resuscitation. This study showed that 30% of the nurse had good practice toward neonatal resuscitation. This finding was inconsistent with the study from Nigeria only 10% of the nurses involved in the management of birth asphyxia adopted appropriate high levels of practices [32]. This discrepancy might be due to the difference in the quality of the training, supportive supervision or refreshment training.
This study showed that resuscitation training (p = 0.002) and supportive supervision (p = 0.027) were associated with the good knowledge of nurses and midwives. This finding was consistent with that of the study from Afghanistan in which training complemented with supportive supervision was significantly associated with a higher knowledge [33]. In this study, a lack of resuscitation guidelines in the workplace was significantly associated with a low level of knowledge. This might be probably because of the lack of guidelines in the workplaces impedes nurses and midwives from updating their knowledge on neonatal resuscitations.
In this study lack of neonatal resuscitation guideline (p = 0.028) and supportive supervision (p = 0.002) were significantly associated with a low level of resuscitation practice. This could be because a lack of resuscitation guidelines in the workplace could result in nonadherence to the resuscitation guideline which in turn resulted in poor resuscitation practice. Similarly, if there is no supportive supervision, the practice level of nurses and midwives deteriorate which may have contributed to poor neonatal resuscitation practice. In this study, being a neonatal nurse is associated with a higher performance of neonatal resuscitation (p = 0.024).
This might be because neonatal nurses received adequate training on neonatal resuscitation during the undergraduate study than other fields of study.
Limitation of the study: The major limitations of this study were an assessment of practice using an interview guide rather than an observational checklist.<\p>

Consent to the participate
Verbal consent was taken from the participants.

Consent for publication
Note applicable.

Availability of data and material
All of the relevant data related to this study are incorporated into the body of the manuscript. Furthermore, for clarity never hesitate to contact the corresponding author.

Competing interests
The authors declare they have no competing interests.

Funding
The authors had declared there is no specific funding to carry out this study.
All authors conceived and designed the study. GB, KB, TH developed the proposal, collected the data, analyzed the data and wrote the draft. GB, AW and MB performed the interpretation of the data and draft of the manuscript. All authors involved in paper writing and manuscript preparation.    Key: * = Significant, ** = highly significant, 1= reference, COR=Crude odds ratio, AOR=Adjusted odds ratio.   Figure 1 Knowledge level of the nurses and midwifery towards neonatal resuscitation in public hospita Practice level of nurses and midwifes toward neonatal resuscitation in public hospitals of nor