Introduction
To reduce high rates of maternal mortality in low- and middle-income countries, the World Health Organization (WHO) advocates for universal skilled birth attendance (WHO, 2004). In much of sub-Saharan Africa, that is most easily achievable by encouraging women to deliver in health facilities. Yet numerous barriers prevent women from delivering in facilities, including logistical issues such as the cost of services (Ejembi et al., Reference Ejembi, Alti-Muazu, Chirdan, Ezeh and Sheidu2004; Mills et al., Reference Mills, Williams, Adjuik and Hodgson2008), distance to facility and difficulty finding transport (De Allegri et al., Reference De Allegri, Ridde, Louis, Sarker, Tiendrebéogo and Yé2011; Faye et al., Reference Faye, Mariama and Ibrahim2011, Gabrysch et al., Reference Gabrysch, Cousens, Cox and Campbell2011). Issues such as fear of being alone during delivery and fear of maltreatment upon arrival at the facility have also been described as barriers to facility-based delivery (D'Ambruoso et al., Reference D'Ambruoso, Abbey and Hussein2005; Mills & Bertrand, Reference Mills and Bertrand2005; Bazzano et al., Reference Bazzano, Kirkwood, Tawiah-Agyemang, Owusu-Agyei and Adongo2008; Spangler & Bloom, Reference Spangler and Bloom2010; Thwala et al., Reference Thwala, Jones and Holroyd2011). These latter issues are especially poignant in a culture where traditional childbirth practices involve tight-knit groups of women who support the pregnant woman through her delivery (Lori & Boyle, Reference Lori and Boyle2011).
One potential intervention to encourage more women to deliver in health facilities is the inclusion of lay support people during facility-based labour and delivery (Yuenyong et al., Reference Yeunyong, Jirapaet and O'Brien2008). The assumption is that if women can bring someone from their existing social network into the facility setting, that individual may be able to provide much needed social and logistical support to help bridge the gap between what women might expect from a home delivery and what they typically experience in a facility delivery.
The term ‘social support’ describes the functional content of social relationships (Story et al., Reference Story, Burgard, Lori, Taleb, Ali and Hoque2012) and differs from other relationship functions in that it is provided by the sender with the sole purpose of being helpful to its receiver. Social support is typically seen as an umbrella term that includes four discrete types of support: appraisal support (providing affirmation, feedback, social comparison and self-evaluation); emotional support (providing empathy, caring, love, trust, concern and listening); informational support (providing advice, suggestions, directives and information); and instrumental support (providing aid, money, labour, time or any direct help) (House, Reference House1981).
A Cochrane review of continuous support during childbirth that spanned low-income and high-income settings documented across sixteen trials and nearly 14,000 women found that women who had continuous intrapartum support – defined in a variety of ways – were less likely to have prolonged labours, undergo Caesarean section deliveries, use intrapartum analgesia and report dissatisfaction with their childbirth experience (Hodnett et al., Reference Hodnett, Gates, Hofmeyr and Sakala2007). Notably, results were best when the support person was not a member of the hospital staff, when support began early in labour and when support occurred in settings where epidural anesthesia was not routinely available (Hodnett et al., Reference Hodnett, Gates, Hofmeyr and Sakala2007).
Advocating for social support during labour is not a novel concept. In the 1990s Hofmeyr et al. published a series of studies in South Africa linking companionship during labour to such things as labour progression, perceptions of labour, breast-feeding and postpartum depression (Hofmeyr et al., Reference Hofmeyr, Nikodem, Wolman, Chalmers and Kramer1991; Wolman et al., Reference Wolman, Chalmers, Hoffmeyr and Nikodem1993; Nikodem et al., Reference Nikodem, Nolte, Wolman, Glumezoglu and Hofmeyr1998). One theoretical explanation posited for the relationship between social support during labour and improved outcomes rests with the ‘buffering hypothesis’ proffered by social support researchers in other settings (Wheaton, Reference Wheaton1985). This hypothesis suggests that having a companion present who can offer various types of social support, including emotional, informational, instrumental and appraisal support, can mitigate the impact of the stressful aspects of a facility-based birth. Women delivering in facilities are often subjected to unfamiliar routines, high rates of intervention, unfamiliar personnel, lack of privacy and a ceding of personal control – all of which may negatively affect labour progression and maternal sense of competence and confidence (Hodnett et al., Reference Hodnett, Gates, Hofmeyr and Sakala2007). According to this theoretical model, having a companion present may serve as a buffer between labouring women and the stressful events they are experiencing, thus reducing the impact of such events.
Understanding the relationship between desire for social support, the source of support and the type of support provided may help explain women's choices regarding lay companionship during facility-based delivery in developing countries. Figure 1, the conceptual framework that underpins this study, illustrates these issues. Each box reflects an empirical question that was asked: Do women desire social support during labour and delivery? If so, by whom? What type of social support was desired, and how did that vary by the source of the support? Sources of support were categorized by sex, with mothers, sisters and mother-in-laws grouped together as female helpers and husbands and partners grouped together as male helpers. The type of support desired was categorized as appraisal, emotional, informational or instrumental. Note that embedded in this framework is the notion that the sex of the lay companion may influence the type of social support that the woman receives. The birthing process in Ghana is viewed as an event that involves mainly women, and the male figure – typically a husband or boyfriend – often participants only prior to or following labour and delivery.
Fig. 1. Conceptual framework for social support during delivery.
Using the conceptual model illustrated in Fig. 1, this study was designed to: (1) increase understanding of women's desire for, and perceptions of, the role of social support during facility-based delivery in rural central Ghana; (2) identify key variables and qualitative themes related to a predisposition towards including or excluding labour companions during delivery in a facility setting; and (3) determine the relationship between the source of support women desired and the type of support they desired.
Study setting
Participant recruitment and data collection took place at Apam Catholic Hospital (ACH), located in the rural coastal town of Apam in the central region of Ghana. Formerly known as St Luke's Catholic Hospital, ACH is a part of the Christian Health Association of Ghana (CHAG), an umbrella organization that co-ordinates the activities of the Christian Health Institution and Christian Churches' Health programmes in Ghana. Apam Catholic Hospital collaborates specifically with the Ghana Health Service to offer basic medical and surgical services, ophthalmology, emergency trauma services and specialized services in obstetrics and gynaecology. This 135-bed facility also serves as the referral facility for the entire Gomoa West District, which is comprised of two clinics, one health centre, four community-based health planning and service (CHPS) zones and a private maternity home. It also serves as a referral facility for bordering districts and communities. Apam Catholic Hospital averages approximately 1900 deliveries annually, and is staffed with eight midwives and three doctors. Antenatal and postnatal education clinics are open to the public and are held at the hospital five days per week. The birthing centre at Apam Catholic Hospital is a labour ward that contains three delivery beds with two hanging partitions between them. One midwife attends to each bed along with support staff. Apam Catholic Hospital serves a projected district population of 109,975 people, and an actual catchment population of 45,090 people. Community health workers at ACH reach out to the community once per week and provide health checks and education to pregnant women in Apam and neighbouring communities.
Methods
Study design
A mixed-methods study design was used, including short-answer questions and in-depth interviews with pregnant women presenting for antenatal care. Quantitative analysis was conducted on key demographic and attitudinal variables assessed during the interview process and qualitative analysis techniques were used to explore the issues surrounding desire for social support during labour and delivery. This approach integrated the descriptive nature of quantitative data with the explanatory nature of qualitative data to provide a well-rounded understanding of social support during labour and delivery.
Participants
A convenience sample of 50 pregnant women was recruited for the study between May 23rd and July 25th 2011. An announcement was made to women waiting in the antenatal clinic regarding participating in a study about women's delivery preferences. Women who agreed to participate were taken to a separate area of the hospital for privacy. Each woman was then taken through a short-form consent document by the research assistant and translator, after which she participated in a 30–45 minute semi-structured interview. To be eligible for the study, women were to be pregnant, able to provide consent and willing to communicate in English or one of the local dialects known by the translator. Exclusion criteria were women facing an imminent health crisis, and those unable to communicate in English or one of the local dialects known by the translator.
Data collection and compensation
A study-specific, semi-structured interview guide was used. The interview guide included questions about participant demographics, previous and current pregnancy experiences, attitudes towards delivery location, attitudes surrounding the need for family support, perceived need for support in the delivery room and attitudes and perceptions regarding having a family member with them in the delivery room during delivery. All interviews were administered verbally in English when possible, and with the help of the translator when necessary. All responses were audio-recorded, with the translator immediately translating the woman's responses back into English when interviews were conducted in a local dialect. Research assistants recorded additional notes on hardcopies of the interview guide during the interview. Each interview was later transcribed verbatim in English, and any identifying information was excluded from the transcription. Women did not receive monetary compensation for their participation. However, participants were offered a complimentary meal and beverage in the hospital cafeteria upon completion of the interview.
Key variables
Social support during labour and delivery was characterized by four components: emotional support (giving encouragement verbally or simply with presence), instrumental support (providing tangible assistance such as money, travel, washing clothes, etc.), informational support (translating to a native language or dialect, or helping the woman understand something related to childbirth) and appraisal support (helping the woman to evaluate her state or actions in a given situation). Each of these components was considered a key variable of interest during data analysis.
Data analysis
All quantitative data were analysed using STATA/IC 11.1 for Windows. Frequencies and descriptive statistics were computed for each of the key quantitative variables collected. These included age, education, gestational age at the time of the interview, total number of antenatal care visits, number of previous deliveries, location of previous deliveries, previous infant deaths, travel time to clinic and desire for companion during labour and delivery. Bivariate analyses compared key variables against the desire for a companion during labour. Due to the small sample size, multivariate analysis was not conducted, and a p-value of 0.10 was taken as statistically significant. All qualitative data were analysed using NVivo 9.0 for Windows. Conventional content analysis was used to describe social support during labour and delivery. Two members of the research team (AA, AM) read each transcript and utilized a specific coding scheme, which involved identifying emergent macro-themes from the interview transcripts by marking common potential thematic areas in the margins. The researchers then discussed the potential themes with the principal investigator and arrived at the initial coding scheme. Using the initial coding scheme, sub-themes were identified and organized. A codebook was then generated from the resultant coding structure, and definitions for each theme were identified from the data. The final codebook was then entered into NVivo 9.0, and each interview transcript was imported, coded and analysed. The research team met periodically to review the coding, discuss any coding discrepancies or additional codes required and ensure consistency in coding.
Ethical statement
At the time of consent each woman was asked if she understood the consent form. Women were reminded that participation was not obligatory nor would it affect their health care or employment, and that they may stop at any time. Non-monetary compensation was offered only after the interview was completed to maintain strictly voluntary participation. The study protocol and all instruments were reviewed and either approved or exempted from ongoing review by institutional research review boards and both the University of Ghana and the University of Michigan.
Results
A total of 50 pregnant women were interviewed. Table 1 illustrates the demographic characteristics of the sample. More than three-quarters had already delivered at least one child, and 28 of the 50 women (56%) had never delivered in a facility before. The mean years of education across the sample was 7.4 years. Eighteen per cent of the sample had no formal education. Fifty-eight per cent of women interviewed said that they would like to have a companion accompany them during labour and delivery at a facility, while 42% of women indicated they would rather not have a companion present. Bivariate analysis suggests that women who reported desiring a companion were younger (p=0.04), had a greater number of antenatal care visits (p=0.03), were more likely to have formal education (p<0.1) and were more likely to be experiencing their first delivery (p<0.08) (see Table 1).
Table 1. Demographic and health-related variables by women's desire for companionship during labour, rural central Ghana
a Means compared using t-test; categorical variables compared using logistic regression.
Figure 2 illustrates a breakdown of women's desire for social support during labour and delivery. Specifically, two main facets of social support are investigated: from whom women preferred to receive social support, and what type of support they desired from their chosen companion. Of the 29 women who desired support from a lay companion, 21 (72%) desired a female companion. Of the female companions desired, the woman's mother was most preferred (n=15, 68%), followed by the woman's sister (n=6, 27%), and finally the mother-in-law (n=1, 5%). Those who reported wanting a female companion present during labour and delivery also reported a range of desired supportive behaviours from those companions; including emotional, instrumental, appraisal and informational support. Of the 29 women who desired support from a lay companion, seven (24%) desired a male companion. Among these women who preferred a male companion, fewer types of social support were expected or desired. Most women desired emotional support from their chosen male companion (n=5), followed by instrumental support (n=1). One of the women who desired social support from a lay companion had no preference for the sex of the companion, but desired emotional support from the companion. These findings are explored further in the qualitative data.
Fig. 2. Social support deconstructed: sources and type of support desired. Women were allowed to indicate multiple types of social support as desired from their chosen companion. Also, some women indicated the sex of a desired companion, but did not indicate which type of social support the companion could provide.
Desire for companionship during labour and delivery
Among the 29 women who desired a lay companion, sixteen (55%) had delivered in a facility previously. Three dominant themes emerged from the interviews with this group. The first two themes focused squarely on the desire for provision of social support by the companion, specifically emotional and instrumental support. They were also most often described in reference to a female helper. The third emergent theme was women's desire for a male companion to ‘witness her pain’ during labour and delivery. Each of these themes is expanded below.
Emotional support, including encouragement and comfort
Many women indicated that the primary role of a lay companion in the delivery room would be to provide emotional support for the labouring woman:
My mother [would come], and if my mother is there, she will give me encouragement. My mother will comfort me.
(35-year-old woman, four previous deliveries; all facility-based)When asked how they felt about having people they knew around them when they deliver, some women said that they would feel encouraged to give birth.
I'm talking about the encouragement that the person I know would give me because if I'm screaming or I'm in pain the person would say you should take it easy. So that's how I would feel when someone I know is around me.
(35-year-old woman, four previous deliveries; all facility-based)… they should let the friends and relatives come stand by me because… I'm saying that maybe the nurse, the midwives won't have time and patience to give you the kind of sweet talks and encouragement that your mother being around can give you. Nice words and all that.
(29-year-old woman, one previous delivery in facility)Instrumental support: ‘In case the nurse/midwife is too busy’
Some women talked about a shortage of midwives, nurses or aids in the labour and delivery ward. Some women discussed wanting a companion present to do tasks staff were too busy to complete, while others desired a companion to be present in order to draw the attention of staff when it was medically necessary.
… let's say my relative is standing with me, and then maybe the nurses go to get something, and then there is something the baby needs or something should happen, my mother can scream and call the nurse for her to come and attend to me and the baby.
(28-year-old woman, no previous deliveries)I agree that they should teach the relative these things because sometimes the nurses they are busy and sometimes even when they are not busy they can't do [it] all and if I have soiled myself, my mother can change me and other stuff. So I think they can be trained in these stuff.
(32-year-old woman, two previous deliveries; both facility-based)In addition, women listed a variety of tasks that they perceived a companion could do to support them during labour and delivery (see Table 2).
Table 2. Types of social support women described that a labour companion could provide, rural central Ghana
Witness my pain
When talking about male companions, many women said they wanted their husband present during delivery so that he would see how difficult and painful it was for the woman.
[My husband] doesn't know much about delivery. I think when he is in the delivery room … I will have some confidence to push. And I can feel that maybe when he is with me in the delivery room, I can feel it and see that he is also sharing my pain.
(23-year-old woman, no previous deliveries)I want him to be there to see [chuckles] … to see how women suffer during delivery.
(32-year-old woman, two previous deliveries; both facility-based)Notably, women who reported wanting their husband present rarely expected him to do anything aside from bear witness to the challenges of delivery.
Do not desire a lay companion during labour and delivery
Of the 21 women who did not desire social support during delivery, sixteen (76%) had delivered in a facility prior to the interview. Three main themes were identified related to the desire to forego a labour companion. Specifically, some women did not want a companion present for fear that an evil-spirited companion would harm them or the baby; some women thought that a companion was unnecessary, as it was the nurse's duty to take care of them; and some women said they would be ‘shy’ of a companion being present during delivery. Each of the three themes identified is expanded below.
Fear of an evil-spirited companion
Some women said that if a companion was present, it would need to be somebody the woman trusts very much. Other women said that regardless of who the person was, the person could have unconscious ill intentions towards the woman. Such ill intentions could take the form of being possessed by an evil-spirit or a supernatural being intending to cause harm. Some women feared that if the companion was possessed by an evil-spirit, the companion could harm the baby or the mother if present during delivery. Thus some women did not desire a companion present in order to protect themselves and their baby.
There are evil people around and maybe that person is evil. So when she stands by me, she can make my delivery complicated and that will affect me and the baby as well.
(25-year-old woman, three previous deliveries; all home deliveries)… it will be better that I go into the delivery room alone with the nurses and the doctors [because some] people have evil spirits and bad intentions. You never know who you are going there with. If the person is an evil person or has any bad intention toward you they can even make you lose your life, or maybe probably anything, do anything bad to you so, with me, it would be better if I go there alone…
(37-year-old, one previous delivery in a facility)It's not necessary – it's the nurse's job
Some women said it was not necessary to have a companion present during delivery, as that person could not help them to deliver the baby in any way.
With me, when I come to give birth, it is the nurses that are there, so I don't want a relative or anyone to come because my mother will not … do anything for me in the delivery room.
(33-year-old woman, six previous deliveries; three home births and three facility-based deliveries)…it's between me and my God, so there's no way my mother would help me in any way while I am giving birth. Only I can push…
(35-year-old woman, four previous deliveries; all home deliveries)I am seriously [wanting to] giving birth so whether someone is standing there or not I can just do it on my own.
(38-year-old woman, six previous deliveries; two home deliveries, four facility-based deliveries)Some women said that it is the duty of the nurse to care for the woman while delivering. Even when provided with examples of tasks that a companion could perform for the woman, some women cited that such a task was the responsibility of the nurses, or that they felt comfortable with only the nurses present:
If she came, my mother wouldn't be helpful. It's only the nurses and those who are working that can help me when I am giving birth. My mother cannot help me in any way.
(34-year-old woman, four previous deliveries; one home delivery and three facility-based deliveries)‘Too shy’ for a companion
Some women said that they would be ‘too shy’ to have a companion, even her husband, present during delivery. When asked to explain, women did not describe modesty or fear of being naked in front of a companion as their primary concern. Instead they expressed concerns that the companion would go outside of the delivery room and tell what happened during delivery to others.
When you are in pain, you talk anyhow. So if my sister is there with me and I say some things … when we go home my sister will laugh at me … [about what happened], and maybe when the child is growing up and my sister sends the child to do something and the child refuses to go, my sister will say that when your mother was giving birth she was in so much pain and [tell what the woman said] … So I don't want that.
(42-year-old woman, three previous deliveries; one home delivery and two facility-based deliveries)There's no one I want to stand by me. That person can go and stand somewhere and say whatever she saw during the … delivery, but the nurse would never do that.
(30-year-old woman, six previous deliveries; all home deliveries)One respondent also said that in addition to being shy of someone being present, typically labour and delivery is something that concerns only women. When asked whether she would like her husband to accompany her in the labour ward she replied:
The husband, no, I'm shy of him, also he is a man. Normally, birth, delivery, is all about women, so it doesn't really concern men. But… I'm really shy of him being there.
(34-year-old woman, four previous deliveries; one home delivery and three facility-based deliveries)Other women flatly refused to consider having anyone with them in labour.
… I don't want someone to stand by me, [even] my mother. My mother doesn't have courage and besides I'm shy of my mother and don't want anyone there.
(33-year-old woman, six previous deliveries; three home deliveries and three facility-based deliveries)Discussion
In this study, 58% of pregnant women presenting for antenatal care in rural Ghana indicated they would prefer to have a companion with them during labour and delivery in a facility setting, whereas 42% indicated they would prefer to deliver alone with the nurses and midwife in a facility. Bivariate analysis of quantitative data suggests that women who reported desiring a companion were younger, had a greater number of antenatal care visits, were more likely to have formal education and were more likely to be experiencing their first delivery. Women desiring support were most likely to desire support from a female companion (her mother, her sister or her mother-in-law) and they expected more varied types of social support from female companions than they did from male companions. There were three themes in support of inclusion of a lay helper: provision of emotional support and provision of instrumental support, as well as witnessing the labouring woman's pain. There were also three themes that ran counter to the inclusion of a lay helper: fear of evil-spirited companion, a companion not being necessary or helpful and being ‘too shy’ of a companion.
These results dovetail well with findings in other low- and middle-income country settings. In a study conducted in Mexico City, researchers found that women who were not assigned the support of a doula during labour and delivery desired a companion such as their husband or mother, and said that having such a companion would make them feel important, supported and cared for (Campero et al., Reference Campero, Garcia, Diaz, Ortiz, Reynoso and Langer1998). Another study among 224 Nigerian women revealed that 80% of women desired emotional support during labour, 18% desired spiritual support and 9% desired instrumental support (Morhason-Bello et al., Reference Morhason-Bello, Olayemi, Ojengbede, Adedokun, Okuyemi and Orji2008). Such results are consistent with the findings presented here of emotional support being the most important form of support a companion can offer during labour and delivery. Researchers in Zambia reported results similar to those presented here with regard to why women would not desire a companion if the option existed in the facility setting. Maimbolwa et al. (Reference Maimbolwa, Sikazwe, Yamba, Diwan and Ransjö-Arvidson2001) found that some women felt it was the health staff's role to take care of labouring women. According to one newly delivered mother, ‘Relatives have no role to play in hospital apart from escorting me here. My own mother is not trained to take care of labouring women, and she does not know what to do,’ (p. 229). Other Zambian labouring mothers cited concerns about the companion interfering with care, spreading rumours about the woman, the labouring women being embarrassed when the companion was present and the hospital staff being responsible for the woman's care (Maimbolwa et al., Reference Maimbolwa, Sikazwe, Yamba, Diwan and Ransjö-Arvidson2001). In another study assessing the attitudes of 70 Russian women towards the presence of a support person during labour, 68.8% of women declined to have a partner present during labour. The most common reasons were that they felt their husbands would be afraid (15.7%); felt personally embarrassed (17.1%); wanted to have a private experience (22.9%); and thought it would adversely affect their sex life (8.6%) (Bakhta & Lee, Reference Bakhta and Lee2010). Similar results were found in a study assessing the attitudes of Nigerian women towards the presence of their husband or partner. In a study of 197 women, 39% did not desire social support of any kind. Of the other 128 participants who did not object to having a companion present, 55% declined to have their husband/partner present during labour because of: personal embarrassment, fear of loss of sexual attractiveness, concern for their husband/partner and desire for privacy (Oboro et al., 2010).
The results presented here challenge the assumption that labouring women in low- and middle-income countries (LMICs) – where traditional home deliveries may include several female relatives – prefer to have a companion with them during labour and delivery in facility settings. These finding may appear counterintuitive, but for those familiar with rural deliveries in western Africa, they may not be surprising. One phrase that was repeated throughout the interviews was the sentiment that delivery in a facility was ‘between me and God’. Women did not want other people with them in the labour and delivery room, but that did not mean women felt alone. They felt as though God was with them, and God would take care of them far better than their mothers or sisters or husbands could. The spiritual component of childbirth in Ghana is very important, and for many women human companionship was simply not necessary in the face of the companionship of God.
These findings also challenge the increasingly common emphasis on paternal involvement in childbirth. For much of rural Ghana, male involvement in childbirth is more of the exception than the rule. It is notable that women who indicated a desire for her husband to be with her during labour and delivery did not desire him to do anything in particular to support her, but rather for him to validate her pain and discomfort by bearing witness; in hopes of better treatment or more empathy from the husband when she returned to life at home.
For women who desired companionship, the results of this study suggest they are discerning in the source and type of support they seek. What is not possible to disaggregate from these results, however, is the direction of the relationship. Do labouring women seek a certain type of support provider (male vs female) because they associate that type of provider with different types of social support? As indicated in Fig. 2, the data suggest that female support providers offer a wider range of supportive behaviours than male support providers. However, are women choosing the companion they would like to have and then adjusting their expectations for the type of support they will get from that companion? Or are women thinking about the type of support they would like, and then choosing the companion who will be able to offer that support? Further research is needed to explore the direction of this relationship.
This research has several implications for practice, policy and research. First, it cannot be assumed that all women presenting at a facility desire an accompanying companion during labour and delivery. However, nearly six out of ten women in this sample would like to have a labour companion. In practice, this means that facilities may need to determine creative ways to make room for family members in the labour and delivery ward. It may also mean that midwives and nurses may need to think creatively about non-medical tasks they can ask family members to assist with to help maximize the benefit for facility staff of people being present in the clinic or hospital setting. From a policy perspective, this research suggests that, logistical and privacy concerns notwithstanding, women ought to have the option to bring someone with them during labour and delivery if they so desire. While creating such a policy has many logistical ramifications – not the least of which is space, given the cramped quarters of many delivery wards – it is likely that the potential benefits may outweigh the initial challenges.
This study also suggests several areas for further research. First, as noted, very little empirical research has been conducted on social support during labour and delivery in LMICs. Further research is needed that explores differences between female labour companions and husbands or male partners in terms of the type of social support offered and the effect it has on the labouring woman. In addition, further research is needed that explores delivery and birth outcomes associated with the provision of social support in facilities in developing countries. While data in high-income countries indicate a link between social support and improved birth outcomes, such studies are lacking in countries like Ghana. In addition, research that disaggregates the impact of various types of social support is needed to improve our understanding of the types of social support that are the most effective for women in labour and delivery. Moreover, further research that explores the counterintuitive notion that women may desire social support during home deliveries, yet prefer to go without such support when delivering in facility settings, is also needed. Finally, research is needed that explores the impact of the inclusion of lay companions during labour on the prevalence of midwifery maltreatment in facility settings. While this topic is extremely hard to quantify, research that explicitly addresses observations of verbal abuse, physical abuse and discrimination in facility settings that do and do not allow companions in labour and delivery is long overdue.
This study has several notable strengths. First, it includes 50 in-depth interviews with pregnant women in rural Ghana. This sample, while somewhat small for a quantitative study, is extremely rich for qualitative research. It includes a broad sample with diverse perspectives, yet the quantitative and qualitative data were in agreement with regard to a general divide in preference for companionship during labour. Another strength of this study is the mixed methodology approach. The variables associated with a preference for a companion (younger age, greater number of antenatal care visits, greater likelihood of formal education and fewer previous deliveries) were explored, as well as asking women what they perceived to be the pros and cons of having a companion in the delivery room. This provides a unique perspective on potential drivers of their preferences of which women may not be consciously aware (their own education level and the number of antenatal care visits they have had), as well as those that they mention with little prompting (such as desire for emotional support or fear of evil spirits inhabiting their companion). An additional strength of this study is that women were specifically asked what they thought a companion might be able to do to help them during labour. Both emotional and logistical support were important to women, suggesting that labour companions may be able to do more than hold a woman's hand and make her feel better. They may also be able to participate in the labour and delivery process in a way that makes things easier for women or midwives.
Despite its strengths, this study has several limitations. First, this study employed a convenience sample of women presenting for antenatal care at one facility in rural Ghana. The study was designed to obtain a wide variety of qualitative perspectives, thus the quantitative results cannot be generalized widely. Additionally, women were questioned about their desire for social support during labour in an environment where lay companion support is not the normal practice. Typically women deliver alone in this hospital. Thus, questions regarding the inclusion of a companion seemed rather hypothetical and difficult for some of the respondents to contemplate. Nonetheless, women appeared to have fairly well-developed opinions on the topic, mitigating concerns about the validity of the data. Finally, this research involved the use of real-time translation between English and Fanti and other local dialects. It is possible that nuances in meaning were lost in translation. However, one of the authors (SE) served as translator and assisted in the verification of all quotes.
Conclusion
In summary, the research presented here challenges the assumption of a unilateral underlying desire for social support during delivery. Not every woman wants lay support during her delivery. Nonetheless, nearly 60% of women in this study reported desiring a female companion to accompany them during facility delivery. This research also suggests that women make conceptual distinctions between support from their husbands and from female relatives, and while emotional support may be a dominant driver of the desire to have a labour companion, logistical support is also cited as an important motivating factor. In the quest to decrease the worldwide maternal mortality rate in part by increasing facility-based deliveries, these data encourage health care facilities to strive to accommodate women's preferences for or against the inclusion of a lay companion to provide social support during labour and delivery.
Acknowledgments
The study was supported by Grant No. T37 MD001425-15 from the National Center of Minority Health & Health Disparities, National Institutes of Health and administered by the Center for Human Growth & Development of the University of Michigan. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of any of its sponsors.
