Husbands’ Knowledge of Obstetric Danger Signs, and Level of Birth Preparedness and Complication Readiness and Associated Factors in Wara Jarso, North Shewa, Oromia, Ethiopia, 2019

Background : Most women have uneventful pregnancies and childbirth. Birth preparedness and complication readiness prevents these problems. Globally, 830 women die every day from preventable causes related to pregnancy and childbirth. Ethiopia is one of developing country with 412 maternal mortality rate. However different intervention was made still mothers preparation for birth is low while the husbands are decision makers and control resources. The aim of the study was to assess husbands’ knowledge of obstetric danger signs, level of birth preparedness and complication readiness and associated factors in Wara Jarso, Ethiopia. Methods : Community based cross sectional study was conducted in Wara Jarso from April 8 to 28 2019. The study participants were interviewed using simple random sampling and data were coded, cleaned and entered to Epidata version 3.1, exported to SPSS 23. Bivariate and multivariable logistic regression was used to identify association of independent variables with husbands’ level of birth preparedness and complication readiness at p<0.05, 95%CI. The results were presented using descriptive, tables and diagrams. Result: The data were collected from 593 participants, and 574 were completed the interview giving the response rate of 96.8%. The mean age of respondents was 36.5±7.8 years. Knowledge of obstetric danger signs of the respondents was 32.06% while level of birth preparedness and complication readiness was 22.30%. 4.478), p=.018), living in urban (AOR = 5.550 (95%CI 2.211, 13.933), p=.001) escorting their wives to health institution (AOR = 2.217(95%CI 1.095, 4.487), p=.027) accept buying material and clothes for baby before delivery (AOR = 3.599 (95%CI 1.995, 6.490), p=.001) and knowledgeable about obstetric danger sign (AOR = 4.957 (95%CI 2.726, 9.016), p=.001) were variables associated with husbands’ birth preparedness and complication readiness. Conclusion and recommendation : The husbands’ knowledge of obstetric dangers signs and their level of preparation was low. Occupation, residence, escort wife, accepting buying materials and knowledge were variables associated with preparation. Therefore, district health office, policy makers, planners, and HEW have to work on awareness creation about obstetric danger signs and how to increase level of preparation among husbands.

and materials necessary to bring to the facility, transportation to a facility for birth or in the case of a complication, and identification of compatible blood donors in case of complications [6].
BPCR needs the participation of both couple in planning which enables mothers to give birth in the presence of a skilled attendant and this effect is magnified when is carried out by the couple. This indicates that male participation can increase the BPCR practices and so should not focus on women only, as involving the couple is most likely lead to positive care-seeking practices [7].
An estimated global total of 10.7 million women have died in the past 25 years between 1990 and 2015 due to maternal causes [8]. It is also evidenced that 830 women die every day from preventable causes related to pregnancy and childbirth. Almost all maternal deaths (99%) occur in developing countries more than half occur in sub-Saharan Africa. Maternal mortality between 1990 and 2015 shows that 12 deaths per 100,000 livebirths for high-income regions, but 546 for sub-Saharan Africa. This indicates that there is high variation between high income countries and sub Saharan countries [9]. In Ethiopia report from EDHS, 2016 indicate that still the maternal mortality ratio related to pregnancy is high, 412 per 100,000 live births [10]. Sub Saharan region has maternal deaths that occur at high rates in all three risky periods [11]. By 2030 Sustainable Development Goal 3 target to reduce the global maternal mortality ratio to less than 70 per 100,000 live births [12].
But the death is mainly caused by direct causes that are due to pregnancy and its complication. Study from Eastern Ethiopia showed that only less than half of pregnant women were prepared for birth [13]. Most common causes of maternal mortality are obstructed labor, hemorrhage and hypertension related complication [14]. Men reported that issues related to pregnancy and childbirth are the domain of women according to study done in Uganda. But women were interested in receiving more support from their husbands [15]. Around half women are not prepared from findings of study from Mizan Tepi, even though 77.6% have information about birth preparedness and complication. This indicates that even though 77.6% of them know preparedness only half were prepared [16].
Husbands control over household and large purchase decision making affects the preparation of women for delivery. Women's decision making on core household and personal issues are very low [17]. Their involvement in making independent decisions on large purchase is almost none (0.1%) [18]. Therefore, the aim of this study was to assess knowledge of husbands about obstetric danger signs and their participation in birth preparedness and complication readiness among husbands in Wara Jarso District.

METHODOLOGY AND MATERIALS
The study was conducted in Wara Jarso Wereda/district, North Shewa, Oromia regional state that is located at 186km to the north of Addis Ababa. According to District health office data the Wereda has 31 kebeles/villages of which 6 are urban and 25 are rural. And its climatic conditions are dega 7.13%, woyna dega 43.73% and kola 49.5%. The Wereda had 40944 households with total population of 191237. Among these population 95535 were females while 95702 were males. The number of female in reproductive age group was 6820. The data also indicated that institutional delivery of the Wereda during the past year was 24%. The Wereda had 57 health extension workers, 7 HC and 25 HP during study period with health service coverage of 92%, ANC follow up 65%, institutional delivery 24% and PNC 24%. The data indicated that ANC utilization is high but with low institutional delivery. The study was conducted from April 8 -28 /2019 using community based cross-sectional study design.
The source of population was all men whose wives were pregnant or had given birth within the past one year from study period. All sampled men were included in the study and data were collected at an individual level. Husbands whose wives were third trimester pregnancy during study or had given birth within the past one year from study period was included while husbands who were critically ill and unable to respond to the interview during data collection period excluded from the interview.
The sample size required for the study was calculated based on a single population proportions statistical formula. The sample size for study was 593 after using 1.5-design effect. Wara Jarso District has 31 villages. Among them 10 were selected using rule of thumb, which is 30%. After data of husbands with third trimester pregnant women or husbands whose wives were given birth within the past one year. Totally within these ten villages there were 891 husbands whose wives were pregnant (528) or given birth within the past one year (368). Having their number then for each village the samples were allocated proportionally. Then the study participants were identified by taking their lists using simple random sampling (SRS). After the participants were identified, the data collectors used their list and the data was collected. Selected study participants who were not available during data collection was interviewed the next day.

Operational definition:
Knowledge of danger sign: Knowledgeable: if husbands knew obstetric danger signs more than mean average score during any of the three phases (pregnancy, childbirth, or post-partum period). Not knowledgeable: if husbands knew obstetric danger signs less than mean average score during the three phases [19,22].
Birth preparedness and complication readiness: Level of birth preparedness and complication readiness was categorized as: prepared: for respondents responded yes for 3 or more of the birth preparedness and complication readiness components. Not prepared for respondents responded yes to less than three of the components of birth preparedness and complication [20,24].
Vaginal bleeding: any vaginal bleeding irrespective of the amount during pregnancy or severe vaginal bleeding or not the same as previous deliveries during labor and delivery [21].

Data Collection Method and Instruments
The data were collected by face-to-face interviewer administered semi structured questionnaire. The questionnaire was composed of 3 components. These were sociodemographic and socio economic characteristics with 12 items. Among these items the average income was removed from analysis because of missing values (64.5%), questions related with obstetric characteristics with 13 items, questions on knowledge of obstetric danger signs, and BPCR. The knowledge of danger sign had components at each 3 stage. The possible danger signs during pregnancy were: vaginal bleeding, severe headache, blurred vision, convulsions, swollen hands/face, high fever, loss of consciousness, difficulty of breathing, severe abdominal pain, severe weakness, reduced fetal movement and water breaks without labour. Possible danger signs during labour were: severe bleeding, severe headache, convulsion, high fever, loss of consciousness, labour lasting greater than 12hrs, and placenta not delivered 30min after baby. Possible danger signs during post-partum period were: severe bleeding, severe headache, convulsion, blurred vision, swollen hand/face, high fever, malodorous vaginal discharge, loss of consciousness, difficulty of breathing, and severe weakness during postpartum. The participants were asked to mention any dangers signs that can occur during the three periods. Questions on birth preparedness and complication readiness includes 6 components such as: saved money for delivery, identified blood donor, identified skilled birth attendant, identified health facility, saved money for emergency, and identified transportation. The tool was adapted from Johns Hopkins Program for International Education in Gynecology and Obstetrics (JHPIEGO) [2]. It was prepared in English. Language experts translated the English version questionnaire into 'Afaan Oromoo' then back to English. Then the data collection was run accordingly. The data was collected using 5 diploma Nurses after one-day training on the objective of the study and tools. Two supervisors supervised data collection.

Data analysis and Processing
After data collection, the data was checked for completeness before entry to computer. Then the collected data was entered to computer using software, Epidata 3.1 and then exported to SPSS version 23. Husband's knowledge was computed using above average mean score at each phase. After the mean was computed then using this mean the it was as knowledgeable for those who score above average mean score, and not knowledgeable for those who score below average mean score knowledge items. Level of BPCR was categorized as prepared for respondents responded yes to 3 or more of BPCR components, otherwise it was coded as not prepared. Bivariate logistic regression was conducted to explore association of each variable with outcome variables to check which variables had association with the dependent variable individually and multivariate logistic regression (stepwise backward likelihood ratio method) was conducted to analyze factors that were associated with husbands' knowledge of danger signs, and birth plans and complication readiness. All associated variables with the main outcome variables by having odds ratio that reach statistical significance in the bi-variate model < 0.05 was candidate for the multivariate model at 95% C.I (p-value < 0.05). The data was summarized and the adjusted odds ratios (AORs) estimated; and their corresponding 95% confidence intervals (95% CI) was computed. The result was presented using tables, figures and narratives.

RESULT
The data were collected from 593 participants, and 574 were completed the interview giving the response rate of 96.8%. The 19 questionnaires were incomplete and excluded from the analysis. The result was presented as descriptive and table for each components.

Obstetrics Characteristics of the Respondents
The obstetric characteristic of the respondents was identified based on their wives' condition by asking the husbands. Among the respondents' wives of 327(57.0%) were breastfeeding during data collection. Regarding number of pregnancy wives of 544(94.8%) respondents were being pregnant more than once and 252 (43.9%) husbands were escorted their wife to health institution in their recent pregnancy. Concerning cultural acceptance of buying clothes and materials for baby before delivery only 107(18.6%) were responded as it was accepted. Regarding place of delivery of their wives 349(64.2%) were delivered their last recent pregnancy at home, and 122(21.3%) were developed obstetric complication. As to number of children 199(36.6%) of respondents were in the category of 0 to 2 children while 170(31.3%) were in the category of 3 to 4 children. About 284(49.5%) of the respondents were heard information of obstetrics danger sings and birth preparedness and complication readiness, and 106(37.3%) were heard from health professionals. 403(70.2%) of the respondents mentioned that there was no health facility in their kebele, and 168(29.3%) need to walk distances of 90 minutes and above. Among the respondents 404(70.4%) were responded that ambulance were used as transportation in their community ( Table 2).
Knowledge of obstetric danger signs were measured at three periods. There were components used to measure at each period. The knowledge of obstetric danger sign of the respondents were computed at each period using average mean score and categorized as knowledgeable or not for each. The knowledgeable respondents about obstetric danger signs during pregnancy, delivery and postpartum period was 178(31.0%), 186(32.4%) and 170(29.6%) respectively. Then for the three again average mean score was computed to categorize the overall knowledge of respondents. After computing average mean score the overall knowledge of respondents was 184(32.06%) which was categorized as knowledgeable about obstetric danger signs for those above average mean score or not for less than average mean score.

Birth Preparedness and Complication Readiness
The respondents were asked to mention the steps they have made during the current pregnancy if their wives were pregnant during study or during the last recent delivery if their wives were given birth within the past one year. Birth preparedness and complication readiness components that were prepared by the respondents were saving money 344(59.9%), identified skilled birth attendant 61(10.6%), identified health facility 141(24.6%) and identified transportation 132(23.0%). Only 51(8.9%) were identified the blood donor while only 93(16.2%) were saved money for emergency cases ( Table 4). Level of birth preparedness and complication readiness was categorized as prepared or not prepared. And those who prepared three of BPCR item was categorized as prepared while those less than three were not prepared. Among the respondents 128(22.30%) were prepared while the rest were not.

Factors Associated with husbands' level of BPCR Factors Associated with level BPCR in Bivariate Logistic Analysis
In Bivariate logistic analysis: age of husband, age of wife, educational status of husband, educational status of wife, occupation of husband, occupation of wife, place of residence, number of pregnancy, escort wife to health facility, cultural acceptability of buying materials for unborn baby, obstetric complication on the last recent pregnancy, source of information, presence of health facility, number of children, mode of transportation in community and time it take to health facility were found to be significantly associated at (p-value ≤ 0.05) with level of BPCR.

Factors Associated with level of BPCR in Multivariable Logistic Analysis
After controlling confounding factors in multivariable logistic analysis of sociodemographic and obstetric characteristics with husbands' level of preparedness among variables significantly associated in bivariate logistic regression analysis using backward LR the following variables were left in the final model.
The variables were occupation of husband, place of residence, escort to health institution, cultural acceptance of buying materials before delivery and knowledge of obstetric danger signs. Merchant husbands were 2.272 times (AOR = 2.272 (95%CI 1.153, 4.478), p=.018) more prepared compared to farmers. Husbands living in urban were 5.550 times (AOR = 5.550 (95%CI 2.211, 13.933), p=.001) more likely to be prepared than those living in rural. And also husbands who escort their wives to health institution were 2.217 times (AOR = 2.217(95%CI 1.095, 4.487), p=.027) more likely prepared compared to did not escort. Those husbands who accepts buying material and clothes for baby before delivery were 3.599 times (AOR = 3.599 (95%CI 1.995, 6.490), p=.001) more likely to be prepared than those who think that buying materials for un delivered baby was not good. Knowledgeable husbands about obstetric danger sign were 4.957 times (AOR = 4.957 (95%CI 2.726, 9.016), p=.001) more likely to be prepared compared to not knowledgeable husbands (Table 7).

DISCUSSION
However, husbands are the decision makers and heads of the households in this community their knowledge of obstetric danger sign was about one third while their preparedness was less than a quarter. There were variables associated with husbands' birth preparedness. Occupation of both husbands and wives, acceptance of buying materials and clothes before delivery and source of information were factors associated with knowledge, while occupation of husband, place of residence, number of pregnancy, acceptance of buying materials and clothes for baby before delivery and knowledge of obstetric danger sign were associated with birth preparedness.
This study shown that husbands preparedness for delivery and complication was low even though around one third have had knowledge of obstetric danger sign. This finding was higher than finding from southern Ethiopia, Tanzania, northwest Ethiopia [20,21,24]. This difference could be explained as time gap and the wives' utilization of maternal care may be improved and initiated the husbands for preparation. study in Kenya was conducted on one factory and participants were selected purposefully. Study from north west Ethiopia were self-administered and this could lead to under scoring may be due to omission of questions.
However level of husbands' preparedness was lower compared to findings from Wolaita Sodo southern Ethiopia and Ambo town Ethiopia [22,23]. The difference could be explained as study conducted in Ambo, Ethiopia was conducted in urban. There may be exposure to information as they were from urban and probability of wives ANC follow up leads to good preparation. And the finding from southern Ethiopia indicated that the data were collected by reading the options for the respondents while in this study data was collected by asking respondents to mention the steps they have made in preparing themselves.
Bivariate logistic analysis indicated that there were variables significantly associated with preparedness of husbands. Among those variables the following were significantly associated in multi variate logistic analysis. With preparation of husbands: occupation of husbands, place of residence, escort wife to health institution, cultural acceptance of buying materials before delivery, and knowledge of obstetric danger signs.
Place of residence were significantly associated with preparation and husbands living in urban were four times more likely to be prepared than those living in rural. This may be due to accessibility to information and health institution as they may contact with different persons. But the study conducted in southern Ethiopia indicated that place of residence was negatively associated with preparedness of husbands [21]. Study from Tanzania shown that place of residence was significantly associated in bivariate logistic analysis, while not in multivariable logistic analyses [20].
Culture has also its effect on husbands' preparation, husbands who think that preparing material for baby before delivery as good were three times more likely to be prepared than those who think that buying materials for un delivered baby was not good. Those accepting buying clothes and other materials as good could prepare themselves as they have no bad perception while those with bad perception could not. Those who escorted their wives to health institution were two times more likely to be prepared compared to those who were not. This could be due to the fact that those who escorted have education from health care providers and initiated to for preparation. Knowledge of obstetric danger sign is significantly associated with preparation. Knowledgeable husbands about obstetric danger sign were three times more likely to be prepared than not knowledgeable. This may be due to their knowledge of the impact of pregnancy related complications. The finding from Tanzania, ambo, Ethiopia and southern Ethiopia indicated that knowledge of obstetric danger sign were significantly associated with husbands' birth preparedness and complication readiness. Husbands with good knowledge were participated than those with poor knowledge [20,21,23]. Husbands' awareness of obstetric danger sign made them participate in preparing necessary materials for complication prevention and early management of the complication. But those husbands having poor knowledge were less likely to be prepared for prevention of the complication. Husbands who interacts with others were knowledgeable about obstetric danger signs and prepared for birth and complication. Strength and limitations of the study Strength of the study The study was conducted using standardized tool. The data collection was conducted at community level and husbands were studied. This study presents evaluating it from the husbands' perspective. Data collectors were trained to teach those who did not prepared for delivery and also about the complications. Limitations of the study This study was done using a cross-sectional design, which may result in difficulty of providing causal relationship between husbands' knowledge of obstetric danger signs and other variables and husbands BPCR and other variable. And also there could be a problem of recall bias even though reduced to one year, as the husbands were expected to remember events that occurred up to one year before the study and data were collected by asking husbands to mention the obstetric danger sign they know and activity they made without reading the options for them. To minimize his one year was used. Additionally, there could be social desirability bias especially when husbands were asked regarding cares they given. The husbands could respond as they were done without performing the activities.

Conclusion
More than three fourth of husbands in study area were not prepared for delivery and complication. Their level of preparation for delivery and complication was determined by occupation of husbands, escorting wives to health institution, cultural acceptance of buying materials and clothes before delivery of baby, place of residence and husbands knowledge of obstetric danger sign.
The finding of the study indicated knowledge of participants and their level of preparation was low and they were affected by different factors. Husbands were decision makers with low knowledge of obstetric danger signs and birth preparedness and complication readiness. To manage this problem awareness creation among them is needed. Therefore, is better if district health office, media and health extension workers give health education for husbands about obstetric danger sign, and birth preparedness and complication readiness at community level. Needs further qualitative study using focused group discussion including both husbands and wives. During data collection all selected study participants were asked their permission and verbal consent was obtained prior to the interview from selected respondents. In addition, confidentiality of information was assured C. Consent for publication The study does not include images or videos relating to individual. But concerning other collected and used data in this study; while obtaining consent from each participant, information related to publishing the study finding were addressed and participants were agreed on that.

D. Availability of data and materials
The datasets and tools used to support the findings of this study are available from the corresponding author upon request.

E. Competing interests
The authors declare that they have no any financial or non-financial competing interests.

F. Funding
Funding for this research was gained from Wolkite University to conduct the study. G. Authors' contributions GT: -The principal investigator designed the study, collect, analyses and interprets the data, and also drafted the manuscript. SB: -Equally participated in conceptualization of the study, design, analyses and interpretation of results as well.

MB: -Participated in conceptualization and design of the study H. Acknowledgments
We would like to thank Wereda health office and health extension workers of selected kebeles for their support in giving us data of the kebele. Our thanks also go to district administration office and village leaders, for their cooperation and facilitation during data collection. We would also like to thank study participants for their participation in the study, and data collectors for participating in collecting data for study. Finally, our thanks go to Jimma University Institute of Health Science, School of Nursing and Midwifery.