Maternity waiting homes as component of birth preparedness and complication readiness for rural women in hard-to-reach areas in Ethiopia

Background In rural areas of Ethiopia, 57% of births occur at home without the assistance of skilled birth attendants, geographical inaccessibility being one of the main factors that hinder skilled birth attendance. Establishment of maternity waiting homes (MWH) is part of a strategy to improve access to skilled care by bringing pregnant women physically close to health facilities. This study assessed barriers to MWHs in Arba Minch Zuria District, Southern Ethiopia. Methods A community-based cross-sectional study was undertaken from February 01 to 28, 2019. Study participants were selected by computer-generated random numbers from a list of women who gave birth from 2017 to 2018 in Arba Minch Health and Demographic Surveillance System site. Data were collected using a pre-tested and interviewer-administered questionnaire. Stata software version-15 was used for data management and analysis, and variables with p-values ≤ 0.2 in bivariate analysis were considered for multivariable logistic regression analysis. Level of statistical significance was declared at a p-value < 0.05. Qualitative data were analyzed manually based on thematic areas. Results MWH utilization was found to be 8.4%. Wealth index (lowest wealth quintile aOR 7.3; 95% CI 1.2, 42), decisions made jointly with male partners (husbands) for obstetric emergencies (aOR 3.6; 95% CI 1.0, 12), birth preparedness plan practice (aOR 6.5; 95% CI 2.3, 18.2), complications in previous childbirth (aOR 3; 95% 1.0, 9), history of previous institutional childbirth (aOR 12; 95% CI 3.8, 40), residence in areas within two hours walking distance to the nearest health facility (aOR 3.3; 95% CI: 1.4, 7.7), and ease of access to transport in obstetric emergencies (aOR 8.8; 95% CI: 3.9, 19) were factors that showed significant associations with MWH utilization. Conclusions A low proportion of women has ever used MWHs in the study area. To increase MWH utilization, promoting birth preparedness practices, incorporating MWH as part of a personalized birth plan, improving access to health institutions for women living far away and upgrading existing MWHs are highly recommended.

Plain english summary One of the ways to reduce maternal and neonatal deaths is to ensure that there is a skilled birth attendant (SBA) at every child-birth. However, in rural Ethiopia, more than half of all births occur at home without the help of SBAs. In remote and hard to reach areas, geographical and logistical barriers are among the main factors hindering skilled birth attendance. Maternity waiting homes (MWH) have been implemented in rural districts of Ethiopia with the aim to address these deterrents. The purpose of this study was to assess the barriers to MWHs use in the rural district of southern Ethiopia.
Face-to-face interviews were conducted among 807 women who gave birth two years before the study period. A structured questionnaire was used to collect quantitative data. We have also conducted focus group discussions to get a better insight into the barriers to MWHs utilization.
Overall, we found a low MWH utilization; women who prepared for birth and its possible complication had higher rates of MWH use. Moreover, women in the low wealth quintile, who made decisions jointly with their partners, who had a history of previous institutional childbirth, who were residing within two hours walking distance, and had the ease of access to transport were more likely to use the existing MWH services. To increase MWH utilization with the aim to improve access to SBA, promoting birth preparedness practices, incorporating MWH as part of a personalized birth plan, promoting MWHs for women living far away, and upgrading existing MWHs are highly recommended.

Background
Maternal mortality continues to be a public health problem globally despite the existence of effective interventions to curb it. In 2017, an estimated 295,000 women died from preventable obstetric complications worldwide, 94% of the deaths occurring in low-income countries. The sub-Saharan region accounted for 66% of the deaths, with 59% occurring in Ethiopia alone, a share that was among the top ten by country in the world [1]. Moreover, for every woman who dies of pregnancy-related complications, about 20-30 others experience morbidity globally, and the burden is adjudged to be highest in low income-countries [1,2].
Ensuring skilled care at delivery, backed up by easy and available transport in case of emergency referral, is vital in reducing maternal morbidity and mortality [3]. Yet, the lack of timely planning for skilled care for normal births and inadequate preparations for obstetric complications is common in low-income countries, especially in Ethiopia where more than 50% of all births are not assisted by skilled providers [4]. It has been proven that making a birth plan during pregnancy reduces delay in obtaining care and promotes the timely use of emergency obstetric care [5][6][7][8]. The major components of birth preparedness include: selecting a desired place of birth, identifying a skilled provider and making the necessary arrangements to receive skilled care for normal births, and preparing for prompt action in the event of an obstetric emergency [5].
Yet, transportation issues, such as the poor condition of local roads and the distance of health institutions (HIs) from the homes of pregnant women, appear to be important factors influencing access to timely obstetric care and may lead to poor pregnancy outcomes [9]. To overcome the major logistic barriers in low-income countries, the establishment of maternity waiting homes (MWHs) at or around HIs has been widely adopted in addition to the common birth preparedness plans. MWHs are homes or shelters built nearby or in the compound of HIs, where pregnant mothers can lodge so that, when they go into labor, they can be quickly transferred to the facilities for safe delivery [5,[10][11][12][13]. Regarding the contributions of the homes, several studies have reported that MWHs improve attendance to antenatal and postnatal care and reduce the odds of homebirth [12,[14][15][16][17]. However, as MWHs have not been considered as one component of a personalized birth plan, there is still a very huge proportion of homebirths in rural and hard-to-reach areas [18]. Poor quality of care at both the MWHs and the adjacent facilities, structural barriers, costs associated with transportation, and psychosocial aspects of the care also result in limited use of the existing MWH services [19][20][21][22][23][24].
However, to the best of the researchers' knowledge, there is no study in the literature which addresses the association between MWH use and birth preparedness plan in Ethiopia. This study, therefore, considers birth preparedness practice as a factor which may affect stay at MWHs. Although MWH intervention has been promoted in the study area to decrease delays in access to maternal care, previous studies have reported that only 9.7% of the women completed the continuum of care [25] and 18.5% of births were assisted by skilled birth attendants [22,26], with about 30% prevalence of birth preparedness plans in 2017 [27]. Evaluation of contextually Keywords: Maternity waiting home, Ethiopia, Birth preparedness plan, Access, Logistic barriers, Low-income, Pregnancy, Maternal health related barriers and constraints to access to MWH services using mixed methods design may improve maternal health. Therefore, this study aimed to explore the barriers to the use of MWHs, taking into account the socialcultural and economic factors of the participants, birth preparedness practices, and geographic specifics of the setting.

Study area and setting
The study was conducted at Arba Minch Health Demographic Surveillance site (HDSS), which is located at Arba Minch Zuria District of Gamo Zone, Southern Ethiopia. The district has a total of 29 kebele (the smallest administrative structure in Ethiopia): 2 semi-urban and 27 rural. Out of the 29 kebele in the district, 9 (eight rural and one semi-urban) were selected randomly based on their geoclimatic zone (high-, mid-or lowland). The surveillance site uses an Open Data Kit (ODK) tablet computer-based application for data collection with updates conducted biannually. Since the HDSS establishment in 2009, there has been continuous recording of vital events and utilization of health services, including verbal autopsy for the cause of death. According to the 2007 Ethiopian national census, the estimated projected total population of the district for 2017 was 195,858, of which 50% were female. There are seven health centers in the surveillance site; six of the health centers are attached each with an MWH as a physically separate room. At the health center level, basic emergency obstetric care services are offered. The health center is typically staffed by health officers, midwives, and nurses and when they are faced with obstetric complications, they refer the women to Arba Minch general hospital where comprehensive emergency obstetric care is offered.

Study design and period
A community-based cross-sectional study using mixed methods design was conducted between February 1 and 28, 2019.

Population of the study
The source population for this study comprised of all women in Arba Minch Zuria District who had given birth within two years to the beginning of the survey. The study population was women in the nine kebele of Arba Minch HDSS site who had given birth within two years to the survey. During data collection, women who were seriously ill or unable to give information were excluded from the study.

Sample size determination and sampling technique
A single population proportion [28] formula was used to estimate the minimum sample size required for the study, with an estimated level of MWH utilization of 31.5% adopted from a previous study which was conducted in Zambia [16], a maximum tolerable error of 5%, and a value of standard normal distribution (z-statistic) of 1.96 at 95% confidence level. The calculated sample size using the above assumptions was 663. In addition, the sample size for the barriers to MWH use (women's occupation, household index, and travel time to obstetric care facility) [29] was computed yet the result yielded a smaller sample size than the level of utilization. An additional 10% from the estimated sample size was considered to compensate for potential nonresponses and multiplying the design effect by two, and the final sample size for the study was 814.
A sampling frame, a list of 2187 women who gave birth between 2017 and 2018, was obtained from the HDSS database and used to select the study unit. The required sample size was proportionally allocated to each kebele based on the total number of eligible women per kebele. The eligible women for interviews were then selected using computer generated random numbers. For qualitative data, purposive sampling technique was used to select focus group discussants. Community health extension workers were contacted first to recruit women who used MWHs (users) and those who did not (non-users). One of the reasons for using the qualitative approach in this particular study was to provide a more in-depth picture of the situation, as qualitative data would reflect the cultural beliefs and practices that need to be considered for MWH service improvement in the future. Four focus group discussions (six to eight participants per group) were conducted among MWH users (2) and MWH non-users (2).

Data collection methods and procedures
A structured interviewer-administered questionnaire was developed after reviewing the relevant literature in the field (see Additional file 1). Socio-demographic information, obstetric and reproductive health information, and data related to MWHs awareness and utilization were collected using ODK (smart phone or tablet computer-based application for data collection) through face-to-face interviews. Ten HDSS data collectors and 6 site supervisors working in the district participated in the data collection. Individual and household identity numbers (every household/individual has a unique identification number in the surveillance site) which were generated along with the list of eligible women from the HDSS database were used to access houses of the sampled women.
For the qualitative data, semi-structured open-ended focus group discussion (FGD) guides were developed to facilitate the discussions (see Additional file 2). The principal author moderated the discussion in the presence of a facilitator from each of the kebele. Group discussions were conducted in health post halls. Idea saturation was used to determine the number of FGDs and each discussion was tape-recorded in order not to forget any issue discussed.

Data quality management
Two days of intensive training was given to the data collectors and supervisors by the study team with a particular emphasis on the objective of the study and the content of the questionnaire. The quantitative questionnaire used to obtain the data was initially prepared in English and translated to Amharic language by an expert in the language, and finally translated back to English to check its consistency with the original meanings. Prior to data collection, pre-testing was conducted outside the study area on 5% of the sample and some important improvements were made on the content of the questions. The subjects were interviewed in separate private places to minimize social desirability bias. Since we used a mobile Open Data Kit application to record responses to structured interviews with participants, data entry related errors were nil.

Data analysis
Participants' responses which were recorded in mobile Open Data Kit applicationwas exported to Stata version 15 software package for further management and analysis. Descriptive statistics were used to describe the level of utilization of MWH services and explanatory variables. The findings were presented in narration, tables and graphs as frequencies, proportions, and means, with their corresponding standard deviations. The principal component analysis (PCA) was used to construct the socioeconomic status (wealth index) of the households where the study participants lived. The questions were adapted from the Ethiopian Demographic Health Survey 2016 [30], which included information on household ownership of durable assets, house and livestock, housing characteristics, and access to utilities and infrastructure. All variables that supposedly measure household wealth were coded into binary variables (0/1). Wealth quintiles (five equal groups) were then constructed based on weights for each asset with the 1st quintile representing the bottom (poorest) households and the 5th quintile representing the top (richest) households in the sample.
Binary logistic regression analysis was performed to identify factors affecting the use of MWH services. Primarily, the association between each explanatory variable and the dependent variable was checked using bivariate logistic regression analysis. Then, explanatory variables which showed association at the bivariate level with p-values of less than 0.25 were considered using multivariable logistic regression analysis. Association between the dependent variable and the explanatory variables was reported using an adjusted odds ratio at a 95%-confidence interval. Finally, statistical significant association was declared at a p-value of ≤ 0.05. With regard to multicollinearity, Variance inflation factor was calculated to check the collinearity between independent variables and we found the maximum value of 2.16 for all explanatory variables included in the multivariable logistic regression. For the purpose of analysis, the recorded audio discussion was transcribed verbatim and categorized according to the themes that emerged, and narrative quotes have been presented to support the quantitative findings.

Definition and operationalization of variables
The dependent variable for this study was the use of MWH services. Those mothers who had stayed in MWHs during their last pregnancy were considered as MWH users while those that had never used MWH were regarded as non-users. The explanatory variables included socio-demographic variables, obstetric-related variables, MWH awareness, accessibility to health institutions' , and birth preparedness and complication readiness (BPCR) practice.
To determine their level of preparedness, the study subjects were asked about the five basic components of BPCR: whether, for their last pregnancy, they saved up for their delivery, arranged ahead for transport, and identified in advance a place of delivery, a skilled attendant and a blood donor; and a woman was considered as "prepared" if she practiced at least three of the five components [31][32][33].

Socio-demographic characteristics of the study participants
A total of 807 women, who had recently given birth, participated in this study. The mean age of the participants was 31.9 years (± 5.1). The majority of study participants were from the rural area (91%), housewives (85.9%), married (99.6%) and had no formal schooling (70%) ( Table 1). Four FGDs were conducted with 29 women: 15 women who admitted to HIs via MWHs and 14 non-users (either directly admitted to HIs to give birth or gave birth at home or en route to HIs.)

Obstetric characteristics and birth preparedness plan
Nearly all (98.6%) participants were multiparous, 706 of the women (87.48%) had antenatal visits, three out of the five mothers (62.4%) had given birth without the assistance of skilled providers, and only one in five (21.9%) women had prepared for birth and its possible complications during their last pregnancy ( Table 2).

Utilization of the existing MWHs
Nearly two out of three (65.55%) participants have heard about the existence of MWHs at nearby healthcare facilities. The main source of this information was community health extension workers (61%). Only 68 (8.43%) of the participants used MWHs during their last pregnancy. Out of the 68 women, transportation problem (46women (67%)), absence of food catering at the MWHs (51 women (75%)), and poor availability of utensils (11 women (16%)) were the main challenges they faced during their stay (Table 3).
Corroborating the quantitative findings (Table 4), the FGD discussants highlighted that substandard service at the MWHs and adjacent health centers are among the reasons most pregnant women do not make use of the existing MWHs, as illustrated below: One of the discussants, a 28-year-old lady, said: I decided to stay in this home [ [TBA]. For some reason I don't know, I had started bleeding shortly after the child was born and the blood was a and then the household's possession of the assets or materials was coded into binary variables (No = 0/Yes = 1). Then households' wealth quintiles (five equal groups) were constructed based on the weights for each asset from the 1st quintile (poorest) to 5th quantile (richest) in the study population.)

Factors affecting MWH utilization
The women's economic status, decisions made jointly with male partners (husbands) for obstetric emergency, history of previous institutional childbirth, birth preparedness plan practice, history of previous childbirth complications, < 2hrs travel distance to the nearest HI, and ease of access to transport in case of obstetric emergency were independent factors significantly associated with MWH use after adjusting for the effects of other variables. Accordingly, those women who live in households that fell within the 1st (AOR = 7.3; 95%CI: 1.2, 42) and 2nd wealth quintiles (AOR = 8.9; 95%CI: 1.8, 42) were more likely to use MWHs than those women grouped under the 5th quintile. Those women who made decisions jointly with their husbands to seek obstetric care (AOR = 3.6; 95%CI: 1.1, 12) had 3.6 times higher odds of using MWHs than those women whose husbands alone took the decisions. Concerning BPCR practice, those women who had prepared for birth and its possible complications were 6.5 (AOR = 6.5; 95%CI: 2.3, 18) times more likely to use MWHs than those women who did not prepare for birth and its possible complications.
Compared to those mothers who had no history of institutional childbirth, the odds of staying in MWHs by those women who have a history of previous health a Number of individuals in a household b BPCR plan: To measure birth plan practices of the study subjects, participants were asked the five basic components of birth plan practice based on their importance (planned to save money to the childbirth, planned to arrange transport, identified place to give birth, identified skilled attendant and identified blood donor for their last pregnancy) and a woman was considered as "prepared" for birth and its complication if she reported at least three of the five basic components institutional childbirth is 12-times higher (AOR = 12; 95%CI: 3.8, 40). Those mothers who experienced complications during their previous childbirth were 3 times (AOR = 3; 95%CI: 1.0, 9) more likely to use MWHs during the following pregnancy compared with those mothers who had no such experience. Those women residing in areas within 2 h of walking distance to the nearest HI were 3.3 times (AOR = 3.3; 95%CI: 1.4, 7.7) more likely to stay in MWHs compared to those women residing in areas farther than 2 h of walking distance. Similarly, those women who had easy access to transport in case of obstetric emergency were 8.8 times (AOR = 8.8; 95%CI: 3.9, 19.8) more likely to stay in MWHs compared to those women who had difficulty in accessing transport for obstetric emergency (Table 5).

Discussion
Using mixed methods design, this study tried to explore the barriers to the use of MWHs at Arba Minch Zuria District. Although all health centers in the study area have MWHs, the study revealed that only 8.4% of the participants used the service for their last pregnancy. The women's economic status, decisions made jointly with male partners (husbands) for obstetric emergency, history of previous institutional childbirth, birth preparedness plan practice, history of previous childbirth complications, < 2hrs travel distance to the nearest HI, and ease of access to transport in case of obstetric emergency were independent factors significantly associated with MWH use. The level of MWH use in agreement with a previous study in Jimma Zone, Western Ethiopia in 2016, where only 7% of the study participants had ever used MWHs [34]. In rural and hard-to-reach areas, the WHO recommends bringing all pregnant mothers, particularly those who are at risk of childbirth complications, physically close to obstetric facilities. The current study was based entirely on the rural area, with the findings showing that the current use of MHW services is very low, which is also supported by other previous studies [35,36]. This might be due to the low level of awareness regarding the importance and benefits of using MWH, inadequate referral from health posts where pregnant women get in contact with the health system, poor encouragement of pregnant women to make prior arrangements for MWH services during antenatal care as part of their personal birth preparedness plans, and the perceived poor condition of existing MWHs [37].
The findings from a study conducted in Zambia showed a 31.5% MWH utilization, which is higher than for the current study. This is probably because promotional and awareness-creation activities might have been done, apart from the routine care, as part of the Saving Mothers Giving Life project in the Zambian study. Moreover, one possible reason why a lower MWH use was recorded in the current study might have been because most MWHs are inaccessible due to the mountainous nature of the setting and lack of faster transport options [38].
This study showed that women in the lower wealth quintiles (1st and 2nd) were more likely to use MWH services than those women in the highest (5th) quantile. Similar findings have also been reported by studies conducted in Gurage zone, Central Ethiopia, Tanzania, Zambia and Malawi, all of which highlighted that poorer women were more likely to use MWH services than women in the wealthiest quintile [14,[39][40][41][42]. The possible reason might be that staying in MWHs may be the only means the poorer women could access obstetric care and overcome delay in reaching the HIs as a result of exorbitant transportation cost and poor road condition. It is also likely that women from households in the lowest quintile had no choice than to stay in MWHs even though the existing MWHs are substandard. At the same time, the poor quality of MWHs service might have discouraged the richest women from utilizing the existing MWHs.
On the contrary, a cross-sectional study conducted in Jimma Zone, Western Ethiopia reported that women from wealthier households were more likely to be MWHusers than the poorest women [34], notwithstanding that the wealthy women (5th quintile) would likely have the economic means for emergency transport to the hospital once they went into labor. This result suggests that women in the wealthier quintiles can be encouraged to use MWHs as far as the problem related to the quality of service is resolved. Moreover, there is the need to upgrade the condition of the existing MWHs to a level suitable for women from different socioeconomic backgrounds.
The current study also found that women who made decisions jointly with their husbands had higher odds of staying in MWHs compared to those women whose husbands alone took the decisions. This finding is also supported by the qualitative finding that most MWH users were accompanied and financially supported by their husbands. Also, MWH users explained that it would not be possible for a woman to stay in MWHs without a joint decision with the husband, as the woman is expected to leave the house for a longer period; the husband and other adult care-takers would have to take care of the home chores till the mother returns back home. A number of studies indicated that decisions made jointly with husbands/partners were associated with improved maternal health outcomes, including maternal service utilization [43,44]. This shows that household decisionmaking to seek obstetric care from the HIs, particularly to stay in MWHs, in collaboration with husbands, could increase the use of MWHs.
Another factor which was statistically associated with use of MWH services was the practice of birth preparedness plan. The finding showed that even though staying at MWH is not a component of personalized birth preparedness plan, women who had prepared for birth and its possible complications were about 6.5 times more likely to use MWH services than those who had not [5]. This is due to the fact that in rural areas like Arba Minch Zuria District, lack of transportation options and poor accessible roads are major hindrances to access to lifesaving obstetric care in case of emergency. As a result, those pregnant women who prepared to give birth at HIs preferred to stay at MWHs until they were due for childbirth. It is also likely that those women who practiced birth preparedness plan received enough counseling from the healthcare providers, which might include the use of MWH services. Hence, the inclusion of MWHs as a component of birth preparedness plan in the rural areas may be necessary.
Because birth complications are unpredictable, pregnant women who are at a risk of complications related to childbirth are encouraged to use MWH services [5]. Previous studies have reported that most intended MWH users had experienced previous childbirth complications [40] or anticipated delivery complications and so needed to be close to HIs [34]. The result of the present study is consistent with those of these previous studies, highlighting that those mothers who had experienced childbirth complications were more likely to use MWH services during the following pregnancy.
This implies that women who have had birth complications may be afraid that they are at a risk of similar events during the following childbirth, so that they decide to be close to the HIs by staying at MWHs. On the contrary, one study carried out in Gurage Zone, Ethiopia reported that pregnancy complications were not statistically associated with the MWH utilization. This might be due to the differences in the study setting, as the current study was entirely based in a rural area, where the study population might not have had alternative means to overcome the second delay but to stay at MWH [45].
Earlier studies have shown a significant association between the use of MWHs and an increase in facility delivery, antenatal and postnatal attendance [16,46,47]. This result is somehow related to our finding that those women who had a history of previous health institutional childbirth were more likely to use MWHs than those who used to give birth outside the HIs. Women who have a history of previous health institutional childbirth (usually due to difficulty during labor) are likely to get adequate counseling and advice from skilled birth attendants if they stay at MWHs during their next pregnancies. Lessons the women learned from their past experiences might have informed their decisions to use the MWHs. The role of the health professionals in counseling the women may also be significant in this regard.
MWHs were established to bridge the geographic gap in accessing obstetric care in remote areas [5]. However, our findings showed that those women residing in areas within 2 h of walking distance to the nearest HI were more likely to use MWH service than those women whose homes were more than 2 h of walking distance from the nearest HI. Similarly, those women who had easy access to transport in case of obstetric emergency were more likely to use MWH service than those women who had difficulty obtaining means of transport. A study conducted in rural Timor-Leste in 2007 also found a comparable result that women who lived within 5 km of the health centers were the group most likely to use MWHs [48]. The potential barriers to the utilization of MWHs, including long distances and the unavailability of means of transport to and from the MWHs, could be the reason why women in remote areas are not motivated to come to MWHs. Moreover, women in remote areas may not have adequate information about the benefits of MWHs and the adverse consequences of home delivery [38].
This conveys a message to the concerned bodies to act urgently as women from remote and hard-to-reach areas, who are the supposed candidates of the MWHs, do not make use of the existing homes probably because of lack of awareness regarding MWHs, difficulty in accessing the homes because of geographical barriers, or total absence of contact with HIs. The results of the current study might be an indication that for improved MWH use, good infrastructure and awareness-creation campaigns are needed to attract women who reside in remote areas [51]. This finding, however, does not corroborate the results of previous studies conducted in Zambia [49], Gurage Zone, Ethiopia [45], and Zimbabwe [50], which found that women who stayed in MWHs were more likely to live further away from the nearest HIs.
This study has two major limitations. First, since the study was cross-sectional, it cannot be proven beyond doubt that the identified factors are causal. Secondly, the women's self-reported travel time estimates might not have accurately reflected the physical inaccessibility of the MWHs. Although we have tried to encourage participants to actively engage in the FGDs, many of them either remained silent or answered a reply sign to indicate that they share the answer that was given beforehand. One of the reasons could have been their literacy level, being from rural and shyness. Except for these limitations, the study used both quantitative and qualitative methods of data collection, which allowed the exploration of factors that a single method would have not been able to capture adequately and provided up-to-date evidence on the current level of MWH use and its associated factors in the study area.

Conclusions
A low proportion of women use MWHs in the study area. The socioeconomic status of women, joint decision making involving both the women and their partners, the degree of preparation for birth and possible complications, history of previous obstetric complication, history of previous health institutional childbirth, residence in areas within 2 h of walking distance to the nearest HI, and ease of access to transport in case of obstetric emergency were the independent positive factors significantly associated with the use of MWHs.
Strategies to improve MWH utilization, including health promotion activities, are very crucial to increasing awareness about the importance of MWHs in the health system, thus enabling women to recognize danger signs during pregnancy and encouraging them to take an active role in household decision-making.
Efforts should be made to incorporate MWHs as part of personalized birth preparedness and complication readiness plans for women living very far away from HIs, and to upgrade the existing MWH services so that they are suited to women from all socioeconomic status. As the attitude of healthcare providers has implications for the initial and continued use of MHWs by pregnant women, taking advantage of all opportunities to address the women, especially when they visit the HIs, and counsel them on the importance of MWHs in client-oriented manner are highly recommended. Further researches assessing MWH users experience to better understand the challenges in MWHs use by employing qualitative methods were recommended to bring further evidence in the efforts to end preventable maternal death.