Effect of Place of Birth on Knowledge of Neonatal Danger Signs and Associated Factors among Mother's in Meicha District, Northwest Ethiopia: A Community-based Comparative Cross-Sectional Study

Background: Danger signs in the neonatal period are non-specific and could be a manifestation of almost any newborn disease. Early recognition of these signs by mothers is a pre-request for increasing neonatal care-seeking behavior. There four this study aimed to assess knowledge of neonatal danger signs and associated factors among mothers who gave birth at home and health institutions in Meicha District, Northwest Ethiopia. Methods: A community based comparative cross-sectional study was conducted among Home and Health institutions delivered mothers two months before the survey. A simple random sampling method was used to select the participants. Data was collected by face to face interviews of mothers. Binary logistic regression analysis was used to identify associated factors. The odds ratio with 95% CI was computed to assess the strength of the associations. Result: A total of 650 (325 health institution delivered and 325 home delivered) mothers were interviewed. Among this, 50.7% (AOR=2.19, 95%, CI (1.594-3.003)) of the mother were knowledge of neonatal danger sign in the district. Mother knowledge of neonatal danger sign was higher among mothers who give birth health institution (60.1%) than home (41.1%). Age of the mother (AOR:3.99, 95% CI:(1.45-11.03)), age of new born (AOR:0.53, 95% CI:(0.36-0.78)), parity (AOR:1.27, 95% CI:(1.37-5.31)), postnatal care attendance (AOR=2.42,95% CI: (1.47, 3.96)), distance of health center (AOR:0.46,95% CI:(0.27,0.78)) were significantly associated with overall mother’s knowledge of neonatal danger sign. Whereas, residence (AOR: 3.09, 95% CI: (1.44, 6.64)) and occupational status of husband (AOR:0.23, 95% CI:(0.201,0.67)) were significantly associated with health institution delivered mother. Age of new born (AOR: 0.50, 95% CI:(0.28,0.896)), parity(AOR: 0.29, 95% CI:(0.113,0.74)), antenatal care (AOR: 12.04, 95% CI: (5.9,24.65)) and postnatal care attendance (AOR:0.27,95% CI:(0.138,0.51)) was significantly associated with home delivered mother. Conclusion: The overall mother’s knowledge of neonatal danger signs was low. However, health institutions delivered mother more knowledge about danger signs than home-delivered mothers. Therefore, it is better if the district Office should investigate the implementation of educational in the community and strengthen the health extension program.


Abstract
Background: Danger signs in the neonatal period are non-specific and could be a manifestation of almost any newborn disease. Early recognition of these signs by mothers is a pre-request for increasing neonatal careseeking behavior. There four this study aimed to assess knowledge of neonatal danger signs and associated factors among mothers who gave birth at home and health institutions in Meicha District, Northwest Ethiopia.
Methods: A community based comparative cross-sectional study was conducted among Home and Health institutions delivered mothers two months before the survey. A simple random sampling method was used to select the participants. Data was collected by face to face interviews of mothers. Binary logistic regression analysis was used to identify associated factors. The odds ratio with 95% CI was computed to assess the strength of the associations.

Background
Neonatal-mortality is one of the world's most neglected and still a significant public health problem. It is estimated that globally, four million deaths during the first 4 weeks of life occur annually, and 75% of neonatal deaths happen in the first week of life [1,2]. Out of the whole neonatal deaths, 99% occur in low and middleincome countries, mostly in sub-Saharan Africa[1, 3,4,5]. A majority of these newborn deaths occur at home, indicating that few families recognize signs of newborn illness, and/or a majority of the neonates are not taken to health facilities when they are sick [4]. Demographic and health survey data from 40 countries collected between 1995 and 2003 reported that more than 50 percent of neonatal deaths occur after home birth [6][7][8].
Different tools to facilitate the identification of these health problems and reduce neonatal mortality have been introduced into health programs in several countries [9]. Integrated Management of Newborn and Childhood Illness (IMNCI) developed by the WHO focuses on the assessment of general danger signs in the examination of neonates presenting with illness at health care centers [9]. The danger signs of severe illness included are history of difficult feeding, convulsion, respiratory rate of over 60breaths per minute, severe chest in-drawing, body temperature above 37.5 o C, body temperature below 35.5 o C, movement only when stimulated, yellow soles (sign of jaundice), umbilical redness or draining pus, skin boils, eyes draining pus, and, Vomiting [10]. Over the past several decades, the global incidence of under-five mortality has steadily decreased. Death during the neonatal period accounts for almost two-thirds of all deaths in the first year of life and 40 percent of death before the age of five [1, 11,12]. Evidence shows that decreasing the proportion of deliveries conducted at home could reduce prenatal death by nearly half [1, 13,14]. Despite the high burden of neonatal mortality, Africa has scored the slowest decrement in NMR with a decline of only 19% from 1990 to 2010, when it compares with 43%decline in these high income and developed countries and still accounts for 39% [10]. Nigeria is the leading country in Africa with the highest number of neonatal deaths. Annually, more than a quartermillion neonates die [15]. Ethiopia is one of the ten countries with the highest number of neonatal deaths globally, with an estimated 122,000 newborn deaths per year [4]. Close to 90% of delivery in Ethiopia takes place at home, this indicating that lack early recognition on the danger sign [11,13]. Childhood mortality levels are decreasing in Ethiopia. According to the Ethiopian Demographic Health Survey 2016, Neonatal Mortality Rate is 29 per 1,000 live births. Infant mortality rate (IMR) is 48 deaths per 1,000 live births and Under-five mortality was 67 deaths per 1,000 live birth [16]. Recent studies identified that predictors for knowledge about neonatal danger signs were educational status, Health Institutions related and others, the consumption of maternal and child health (MCH) services, obstetric status, and place of births [17-20, 22, 26, 29]. As people attribute high neonate deaths to low utilization of neonatal care services at crucial stages of pregnancy, delivery, postpartum and post-natal periods, neonatal complications during birth or after the birth of a child greatly contribute to neonate mortality in developing nations [13]. If those factors affecting neonate danger signs are not well known and identified by the mothers, neonatal mortality and morbidity might continue and a challenge to the community as well as the government in the future. Therefore, improving maternal knowledge concerning neonatal danger signs is a key entry point.

Study setting
A Community based comparative cross-sectional study design was conducted from March to April 2018 in Meicha District, northwest Ethiopia. It is located 35km and 543 Km far from the capital cities of the Amhara Region and Ethiopia, respectively. The District has 12 urban and 41 rural Kebeles. Based on Meicha District health office report, there are 1 hospital,13 health centers, 3 ambulance service, 53 health post, 14 private lower clinics, and 8 private drug stores which deliver comprehensive health services to the communities in the district. There was a total population of 614,022 (male 308,765 and female 305,254), 115,225 households, 14,933 estimated live births, surviving infants of 6102,<5 years of 42028, estimated women with childbearing age of 131,717, estimated pregnancy/delivery of 7257and new pregnant in fertile age of 47764. There are several 317 all types of health professionals of different disciplines in the district including HEWs. The local communities in the district largely depend on agriculture [21].

Participants
All mothers who were living in the selected kebeles of the Meicha district, and gave live birth two months before the data collection were considered. Mothers who lived at least 6 months in Meicha districts were included, while mothers who were not mentally competent or who had any psychiatric disorders and/or had serious illness were excluded.

Sample Size Determination
The total sample of the study was determined by using a double population proportion formula. Mother's knowledge of neonatal dangers sign was determined by calculating for factors with Epi-info version 7.0 by taking factors from previous findings in Welkite town 41% [18], 95% confidence level (1.96), Power 80%. The sample size without design effect and non-response rate was calculated as -206 +206 = 412.By considering the 1.5 design effect it was calculated as -412X 1.5 =.Then considering 5% non-response rate:-618+32 = 650. This sample size was divided into two groups, therefore, the final sample size for home-delivered was 325 and for Health institution delivered was 325.

Sampling Procedure
A simple random sampling technique was employed to select the study participants. From the 53 kebeles in the Woreda, 25 kebeles were selected by using simple random sampling (lottery method) having the total mothers who gave birth the last 2 months was 917(381 at Home and 536 at Health institution). For each kebeles sampling frame was developed by taking mothers who gave birth in the last two months before this study. Then based on the population, the sample size was allocated for each selected kebeles proportionally. The allocated sample size for each kebeles was selected using simple random sampling and study participants were interviewed at their home.

Operational Definitions
Good knowledge: A mother who asked to mention danger signs of the neonate, if she can mention three of the ten danger signs for neonate without prompt and three and greater danger signs with prompt [23,26]. Institutional delivery: Is any delivery service, mothers who gave birth in the health facilities, including health posts attended by health extension workers. Recently delivered mothers: Mother who delivered before 2 months of the survey irrespective of how many births gave before.

Data Collection Methods and procedures
Data was collected by using face to face interviews with 8 interviewers who completed diploma Midwifery's and two Health officers supervised them. Data was collected using a structured interviewer-administered questionnaire. The questionnaire consisted of information on socio-demographic characteristics, Obstetric characteristics, Place of delivery, Maternal health service exposure, Health Institutions related factors and other questions. The data collector was organized into four pairs to minimize bias and errors during the data collection.

Data measurement methods
Mothers' knowledge was measured by mothers' capability of mentioning neonatal danger signs with and without prior prompt by the interviewer. Then knowledge was categorized into good knowledge and poor knowledge.

Data Quality Control
The data collection tool (questionnaires) initially prepared in English and translated to the local language and translated back to English by independent language expert was used to maintain consistency of meaning. Three days of training was given to data collectors and supervisors on data collection techniques, how to approach and collect the required information from the respondents. Data were collected under the supervision of the supervisors and the overall data collection process was supervised by the principal investigator daily. The pretest was conducted outside selected kebeles with 5% of the study subjects to test the quality of the data collection tool. Data were cleaned and checked for completeness daily.

Data Processing and Analysis
After checking data for completeness, coded manually data were entered into Epi -Data version 3.1 and exported into SPSS version 21 software for analysis. The results were expressed by text, tables, and charts Binary logistic regression model including Bivariate and multivariable logistic regression were fitted All independent variables having p-value <0.2 in the Bivariate logistic regression analysis were entered into a multivariate logistic regression analysis to control confounding effect. Appropriate model diagnostics and goodness of fit tests were done. Multicollinearity was checked to test correlation among predictor variables and Hosmer and Lemeshow test P-value (>0.2) were conducted to see model fitness. Odds ratio with 95% confidence interval was used to determine the presence, strength, and direction of the association between dependent and independent variables.

Socio-demographic characteristics of the participant's
A total of 637 mothers' (321 home delivered and 316 institutions delivered mothers have participated in this study representing a response rate of 98% (98.8% and 97.2% for home-delivered and institution delivered mother, respectively). Out of the total participated women 316(49.6%) were from Health Institution delivered and 321(50.4%) from Home deliver women's mean age of the study participants was 29.59(SD = + 6.52) years. (Table 1).

Maternal and Child Health service and obstetric factors
Approximately three-fourth of participants 471(72.5%) had at least one ANC follow up for their recent birth. Out of the total participants, 384(53.5%) have four and more ANC visits. (Tabel 2)

Health Institutional factors related and others
Out of the total participants, three hundred eighty-four (53.5%) have four and more ANC visits. Among the interviewees, six hundred one five (94.6%) of mothers have the accessibility of modern health facilities in their local. (table 3)..

Mother's knowledge of Neonatal danger Signs between Health Institution and Home delivery
Health professional was the most common source of information were 392(73.7%) mother's heard or counseled about neonatal danger sign (figure 1)..
The majority of participants five hundred twelve (83.4%) had information heard about neonatal danger signs for their recent birth. Out of the total participants, 80.5% of mothers knew at least one danger sign. (See table 4

Factors associated with the mother's knowledge of neonatal danger sign
We have fitted three different models to assess the mother's knowledge of neonatal danger signs. The first model was fitted to assess the overall factors of the mother's knowledge of neonatal danger signs. Variable such as Age, postnatal follow-up, parity, age of the newborn and far from health center was significantly associated with mothers' knowledge of neonatal danger sign for the whole mothers irrespective of Age. Mothers who were 15-19 years had 4 times more knowledgeable (OR: 3.99, 95%CI (1.45, 11.03)) than that of >35 years mother and mothers' who attend postnatal follow up were 2.42 times more knowledgeable (OR: 2.42, 95% CI (1.47, 3.96)) than that of not attend postnatal follow up. mothers who give birth four and above were 1.27 times more knowledgeable(OR:1.27,95% CI (1.37,5.31)).than that of who gives birth two up to four times. Additionally, mothers' who were >30 days age newborn had decreased knowledge by 47% (OR: 0.53,95 % CI (0.36, 0.78)) when compared with that of <30 day age of newborn. Mothers far more than 10 km had decrease knowledge by 54% (OR: 0.46, 95% CI (0.27, 0.78) when we compared that far less than 10KM. (Table 5) The second model was fitted only for Health institution delivered mothers. Accordingly,the residence and occupational status of the Husband showed a significant association. Health institution delivered mothers with relatively being urban residence was 3 times (AOR = 3.09, 95% CI (1.44, 6.64 more likely to know neonatal danger sign compared to counterparts. Mothers had farmer occupation were 77% less likely (AOR = 0.23 95% CI (0.207, 0.67) to have good knowledge of neonatal danger sign than government employed ( Table 6).
The third model was fitted for Home delivered mother's age of newborn, parity, ANC and PNC. As a result, Age of newborn, being below 30-day mothers' was 50%(AOR = 0.50, 95% CI = 0.28,0.896) less likely knowledgeable about neonatal danger sign than those age 30 and above day. Mothers had one-time giving birth decrease knowledge by 71% (0.29, 95% CI = 0.113,0.74) when compare that giving birth four times and above. Mothers with ANC follow up were twelve times (AOR = 12.04, 95% CI = 5.9, 24.65) more likely to have good knowledge of neonatal danger signs than mothers who had no ANC. Mothers who did not attend PNC follow up were 73% less likely(AOR = 0.27, 95% CI = 0.138, 0.51)to have good knowledge of neonatal danger signs than mothers who had PNC follow up (Table 6).

Discussion
Mother's early identify neonatal danger sign has a great potential to reduce neonatal mortality. In this study, the overall Knowledge of neonatal danger sign was 50.7% (AOR = 2.188, 95%, CI = 1.594, 3.003). There was a significant difference in mothers' knowledge of neonatal danger signs with the place of birth at health institutions (60.1%) and home-delivered (41.1%). Mothers who give birth at health institutions were 46 times knowledgeable about neonatal danger signs than mothers who give birth at home. The possible reason might be health institutions may provide an adequate education for mothers to deliver theirs about neonatal danger signs. It could be also due to the accessibility of information about neonatal danger signs.
Overall mother's knowledge of neonatal danger sign (for total 637 mothers) was 50.7%. The overall mother knowledge of neonatal danger sign this study, is lower than the study conduct in Egipt and Ghana were (68-70%), 71.9% respectively [23,27]. but this result was in line with the study conducted in Bangladesh, Mekele Ethiopia and Chincha district southern Ethiopia which showed mothers knowledge neonatal danger sign of 51, 50.6 and 50.3% respectively [17,20,25][ However, this finding is higher than the study conducted in Nigeria  [15,19,26,28]. The possible reason for the observed difference might be due to the extensive work of HEWs in the community and different health care institutions in awareness creation on neonatal danger signs. The other possible reason might be study period difference, time gap(this study included less than two-month newborns, but in the other study included more than 4 months which leads to recall bias), residence and socio-demographic and cultural variation which may negatively or positively affect the level of knowledge neonatal danger sign. Among the socio-demographic variables, only residence and occupational status husband of mothers makes a significant association with knowledge neonatal danger sign at Health Institution delivered while none of them showed a statistically significant association at Home delivered mothers. The reason for this difference might be mothers who live in urban were more seek to health care and health information from different sources as compared to mothers who lived in rural parts. For health institution delivered mothers, urban women were 3 times more likely to know neonatal danger signs when compared to rural mothers. In agreement with this, studies revealed that the mother's residence determines their level of knowledge of neonatal danger signs [17,29]. For example, the study conducted in Chincha towards mother knowledge neonatal danger sign with residence showed that mothers who lived in urban were two times more likely known as compared to mothers who lived in a rural area. For health institutions that delivered mothers, the husband occupation level was associated with knowledge of neonatal danger signs. Those respondents whose husband occupation farmer was 77% less likely to know neonatal danger signs as compare to Governmental employees. This may due to Governmental employ husband are more informed about neonatal danger signs. In this study parity of mother was statistically associated with the level of knowledge neonatal danger sign. Mothers who were giving birth one times had 71% less likely knowledgeable as compared to mothers who give birth four and above at the home-delivered. Comparably, in the overall model of mothers' knowledge neonatal danger sign, mothers who had to give birth four and above had two times more likely knowledgeable about neonatal danger sign than mothers who gave birth two up to four times. This finding was supported by studies in Egypt showed, mothers being primi Para are more likely to have poor knowledge about newborn care than multi parity [18]. This study has shown the age of the newborn was found to be a determinant to affect knowledge of neonatal danger sign at home delivered mothers'. Whose age below 30 days newborn 50% less likely knowledgeable about neonatal danger sign than whose age was 30 and above days newborn. This might be due to mothers exposure to Immunization package, might get information on neonatal danger sign. Comparably, in the overall model of mothers knowledge neonatal danger sign, mothers who had newborn age 30 and above day had decreased knowledge by 47% when compared with mothers who had newborn age below 30. In the present study having ANC follow up was twelve times more likely to know neonatal danger sign as compared to women had no ANC follow up. This finding is supported by finding in Uganda and Gonder, north Ethiopia [15,19]. The possible and the best reason for this could be mother's exposure to ANC package, might get information on neonatal danger sign.
In this study is no attendance PNC follow up was a significant factor to affect knowledge of neonatal danger sign of Home delivered mothers those who had no PNC follow up in their last child were 73% less likely knowledgeable when compared to those who attend PNC follow up care. Similarly, with finding studies in Mekele city, Tigray [20]. Those mothers who had postnatal follow-up and who received advice regarding neonatal danger signs were 4 times more likely to know the defined neonatal danger signs as compared to their counterparts. The possible reason might be mothers have no attended PNC follow up are did not receive counseled about neonatal danger signs; this decrease knowledge of the mother concerning the neonatal danger signs. Comparably, in the overall model of mothers knowledge neonatal danger sign, mothers who had to attend PNC follow up had two times more likely knowledgeable about neonatal danger sign than mother who had not to attend PNC follow up. Mothers' those who were knowledgeable about neonatal danger sign by age 15-19 years old were four times more knowledgeable than that of counterparts in the overall model of mothers knowledge neonatal danger sign. This study is far from the health center was a significant association in the overall model of mothers' knowledge of neonatal danger signs. Those who reside in greater than 10 km 54% less likely knowledgeable when compared to those who raised less than 10 km. This is due to the physical proximity of health facility affects the utilization of services. In this study to the overall model of mothers' knowledge neonatal danger sign, residence, frequency ANC visit, and place of birth were not showed statistically significant association which is supported by studies done in rural Wardha, Uganda, Mekelle Ethiopia [15,20,30].

Conclusion
Place of birth affects the mother's knowledge about neonatal danger signs.
In this study mothers' knowledge of neonatal danger signs at Meicha district in both health institution and homedelivered as well as for the overall model of mothers' knowledge, neonatal danger signs were to below.
The percentage of mothers' knowledge of neonatal danger signs at health institutions and home-delivered mother of the district was statistically different.
Age of the mother, antenatal care and postnatal care attendance showed a significant associated with homedelivered mother's knowledge neonatal danger signs. Were as a residence and occupational status of Husband showed a significant associated with health institution delivered mother's knowledge neonatal danger signs. It is better to give infancies, especially for home-delivered mothers to create awareness regardless of neonatal danger signs and to strengthen the health extension program. Ethical clearance was obtained from the Institute of Public Health, College of Medicine and Health Sciences, University of Gondar. Formal letter of permission was obtained from administrative bodies of the West Gojam zone Heath Department, Meicha district and selected kebeles. A letter of cooperation from kebeles administrators was also secured. Finally, Written and verbal consent was obtained from every study participant included in the study during data collection time after explaining the objectives of the study and the right to withdraw from the study at any time. The confidentiality of respondents was also assured by excluding personal identifiers like the name. Written informed consent was obtained from a parent or guardian for participants under 16 years old.

Availability of data and materials
The datasets used during the current study available from the corresponding author on reasonable request.

Competing interest
The authors declare that they have no competing interests.

This research project is funded by the University of Gondar
Authors' contributions TG designed the study, performed statistical analysis, and drafted the paper. AT, AB and TG participated in the paper writing. All authors contribute to the data analysis and read and approved the final paper.      Note: *-significant results, 1-reference category **p-value<0.001, *p-value≤0.05 Note: *-significant results, 1-reference category **p-value<0.001, *p-value≤0.05 Figure 1 Diagramatic presentation of source information with mother the last two month giving birth at home and health institution in Meicha Woreda, north west Ethiopia2018 Figure 2 Diagramatic presentation of mother's knowledge of neonatal danger sign between Health Institution and Home delivered mothers in Meicha Woreda, north west Ethiopia2018.