Women’s satisfaction and perception of control in childbirth in three Arab countries

Abstract Women’s satisfaction and perceived control in childbirth are important attributes of the childbirth experience and quality of care indicators. This article presents findings from the pre-intervention phase of a multi-centre implementation study in Egypt, Lebanon and Syria, to introduce a labour companionship model in these countries. A sample of 2620 women giving birth in three public teaching hospitals from November 2014 to July 2015 in Beirut and Mansoura, and from November 2014 to April 2015 in Damascus were interviewed by trained field workers. Additional information was abstracted from medical charts. An adapted version of the Mackey Childbirth Satisfaction Rating Scale was used to measure women’s satisfaction and the shortened version of the Labor Agentry Scale was used to assess perception of control. The total satisfaction score was high in all sites with the lowest being in Egypt. Perceived control was directly related to satisfaction. Women with low education levels had higher levels of childbirth satisfaction. Women who had fewer children from Egypt and Lebanon, and those who received care by a team including both male and female physicians in the Syrian hospital were more likely to be dissatisfied than their counterparts. Variations in the management and provision of care between the three countries may explain the differences in satisfaction levels observed. Further qualitative research is needed to deepen our understanding of the concepts of control and satisfaction in the Arab culture as well as to establish the factors associated with women’s positive childbirth experiences to inform the provision of quality maternity care.


Introduction
Childbirth is an important life-changing event and women's birth experiences have been reported to have long and short-term influences on women's well-being. [1][2][3] Reviews indicate that women's sense of control during labour is an important determinant of their satisfaction with the childbirth experience, 4,5 which is considered an important indicator of the quality of maternity care services. 6 A woman's satisfaction with the birth experience has been shown to influence her relationship with her infant, to affect her self-esteem and self-image, and influence her future childbirth expectations. 1,3,7 Perceptions of being in control during childbirth have been recognised as the strongest component of women's birth experiences, [8][9][10][11] of their own behaviour during labour and their interaction with care providers, 10 contributing largely to women's feelings of fulfilment and postpartum well-being. 10,12 Satisfaction is also related to the caregiver's attitude, good communication with care providers, and the responsiveness of staff to women's needs. 5,13 One report indicates that dissatisfaction with care and perceptions of diminished control over the process of childbirth have led to a preference for caesarean sections for future births. 14 A number of individual-level and system-level factors pertaining to the policies, organisation and management of health care have been reported in the literature to be associated with positive childbirth experiences, such as women's involvement in the decision-making process, access to information, familiarity with the care provider, support during labour and type of birth. 5,11,[15][16][17] Demographic factors like education, age, socio-economic status, marital status have been found to have little relationship with childbirth satisfaction in a systematic review, except for higher levels of satisfaction with multi-parity 5 and high education levels. 9 Despite the increased recognition of the importance of women's positive experiences with birth as an essential attribute of quality of care and women-centred care, [18][19][20] research on quality of maternity care has focused mainly on avoidance of adverse events and minimal attention has been given to women's childbirth experiences. 21 In 2016, the World Health Organization (WHO) put forward a framework of quality of care for pregnant women and newborns, highlighting the importance of women's experiences including the provision of companionship during labour and birth and ensuring respectful maternity care. 22 Also recently, social movements such as the humanisation of childbirth and the promotion of women-centred approaches aim to prevent disrespect and abuse and improve women's experiences with care. 23 The provision of emotional support through a birth companion is a component in models advocating respectful maternity care. 24,25 The objective of this article is to describe the levels of satisfaction with the childbirth experience and perceptions of control of women giving birth in public hospitals in three middle-income Arab countries, Egypt, Lebanon and Syria, and to determine the service delivery factors associated with their satisfaction. The data analysed here is from the findings of a pre-intervention phase of a multi-centre implementation research study to develop and introduce a labour companionship model in public hospitals. Baseline levels of satisfaction and perceptions of control were established during this phase in order to inform the implementation of the intervention and measure their changes during implementation.

Maternity care services in the three settings
The structure of health systems in the three Arab countries (Egypt, Lebanon and Syria) involved in this project is different. In this regard, Egypt and Syria have both active private and public sectors for service delivery. In Lebanon, the private sector is dominant. Fragmented care is characteristic to each of these settings with lack of continuity from prenatal to postpartum care. Out of pocket money spent on health is high in the three countries, although a health insurance system is better established in Lebanon. In addition, the ongoing political instability in the region continuously influences the provision and access to health care services in these three countries. We note that the situation in Syria has changed since the beginning of the conflict in 2011, with around half of the public health care facilities reported being closed or non-functional with consequent disruption in access to health care. 26 Egypt and Syria are characterised by total fertility rates of around 3, whereas Lebanon has a total fertility rate of 1.5. 27 The Arab region, including these three countries, has witnessed a decrease in maternal mortality levels during the last decades 28 coupled with high levels of skilled birth attendants. 29 The most recent estimates point to seven out of ten births happening in health facilities in Syria 30 and Egypt, 31 98% of births occur in hospitals in Lebanon. 32 Childbirth is highly T Kabakian-Khasholian et al. Reproductive Health Matters 2017;25(S1):S16-S26 S17 medicalised in hospital settings in these three countries where a substantial discrepancy between routinely followed practices and best practices in maternity care exist with a number of unnecessary or even harmful practices, such as non-indicated labour induction, and enemas and episiotomies for primiparas being routinely applied. 33 Reported Caesarean section rate in these countries have also followed the worldwide increasing trends with 41% in Lebanon 34 based on hospital-based studies, 52% in Egypt 35 and around 27% in Syria 30 based on population-based studies. Earlier studies from the region indicate that there are no national or hospital policies supporting labour companionship in these countries and women's choices for having a labour companion is not always considered in Lebanon, and almost never considered in Egypt and Syria. 33 This comes in contrast to women's needs for support and experiences of isolation and fear as documented previously in these countries. 36

Study setting
One public teaching hospital was chosen in each of Mansoura, Egypt; Beirut, Lebanon; and Damascus, Syria as the sites to conduct the project. All three are university teaching hospitals that provide services to a population of low to middle socio-economic level, and have large number of births compared to other hospitals in their geographic areas. The hospitals in Beirut and in Damascus also serve a large number of Syrian refugee or internally displaced women. The hospitals in Mansoura and Damascus have both a public and a private section, while the hospital in Beirut is only public. However, this study only included participants using the public section of the hospitals in Mansoura and Damascus. Labour care, in all three sites, is provided mainly by midwives, nurses and residents with slight variations between settings. Birth is attended by senior residents in Syria and Egypt, and by residents and obstetricians in Lebanon. Both hospitals in Damascus and Mansoura have shared labour rooms accommodating four to twelve women in each room, while in the selected hospital in Beirut, the labour and birthing rooms are designed to house only one woman. The policies at the Damascus and Mansoura hospitals do not allow labour companionship and continuous labour support is not offered by the staff. In the Beirut hospital, labour companions are allowed, however, this practice is not institutionalised and is usually left to the discretion of the care provider.

Methods
A total sample of 2620 women (709 in Mansoura, 705 in Beirut and 1206 in Damascus) giving birth in the three selected hospitals participated in the pre-intervention phase of the study. The sample size was calculated based on an estimated reduction of 4% on the average C-section rates in the three countries (35%), the primary outcome of interest in the intervention study, considering a power of 85% and a significance level of 5%. Women who gave birth and consented to participate in the study during the period November 2014 to July 2015 in Beirut and Mansoura, and November 2014 to April 2015 in Damascus, were approached and recruited in the study. Recruitment was continuous throughout different time periods on all days of the week. Women who were classified as high-risk by health care providers upon arrival to the hospital, those who suffered from intrauterine foetal death and those below 18 years of age were excluded from the study. The age restriction was applied to include women who could provide informed consent without the need to seek the consent of a guardian considering that a legal guardian for women below the age of 18 might not be present at the time of recruitment. The overall refusal rate was 1.4% in the three sites.
Trained female field workers checked the labour and birth registry daily and approached women eligible for the study in the postpartum ward before their discharge from the hospital which ranged from 6 hours to a maximum of 48 hours after giving birth, to obtain informed consent and conduct an interview using structured questionnaires. Socio-demographic data, women's satisfaction with the childbirth experience and perception of control during labour were assessed through women's interviews using the questionnaire. In addition, a member of the staff abstracted information from medical charts on a daily basis using a form prepared by the study team. Information about type of birth, Apgar score and length of labour was obtained from the medical charts.
To measure women's satisfaction, we used the Mackey Childbirth Satisfaction Rating Scale, a 34item scale of established reliability and validity, consisting of six sub-dimensions capturing aspects related to self, partner, baby, nurse/midwife, T Kabakian-Khasholian et al. Reproductive Health Matters 2017;25(S1):S16-S26 S18 physician and general rating scale. 9 Women indicate their degree of satisfaction with each item using a 5-point Likert scale. This scale was adapted to a 31-item scale, translated and used in Arabic in a previous study. 37 Additional adaptations, pertaining to the wording in Arabic, were made for this study. The items removed pertained to women's expectations, which were not relevant for women in these contexts. The scale was pilot tested in the three sites for comprehension of the wordings used in Arabic. Internal consistency reliability coefficients for this adapted scale was calculated from data used in a previous study conducted in Lebanon 37 (total scale 0.95; self 0.93; husband/family 0.97; baby 0.83; nurse 0.95; physician 0.95) and were found to be comparable to the reported ones for the English version of the original scale. 9 The Labor Agentry Scale (LAS) was used to assess women's perception of control. The LAS is a 29item scale with a shortened 10-item version that measures personal control during childbirth with established robust psychometric properties of validity and reliability. 38 Women's degree of agreement or disagreement with each item is measured on a 7-point Likert scale. We used the shortened 10-item version, which was translated into Arabic and back-translated into English to check for accuracy of translation and was pilot tested in the three sites. Coding was reversed for the negatively worded items so that a higher score on the scale reflects a higher personal control.
The independent variables were identified from the literature (education, parity, type of birth) and from the formative research conducted by the study team prior to the intervention where women's preferences for female providers were noted. 37 Ethical approval was obtained from the Institutional Review Board at the American University of Beirut, the University of Damascus, the University of Mansoura, the hospital in Beirut and the World Health Organization Ethics Review Committee. The approval and the collaboration of the management of the three selected hospitals were ensured. Written informed consent was obtained from women prior to interviews.
Descriptive analysis was conducted initially for the variables of interest to this study. Chi-squared statistics and one-way analysis of variance (ANOVA) test were used to compare socio-demographic and birth-related characteristics among the three country sites. Sub-scores were created for the sub-dimensions of the satisfaction scale in addition to the total satisfaction score. Multiple regression analysis was performed to identify the determinants of the total satisfaction in this sample. For this purpose, the variables of education and gender of the obstetrician were grouped into dichotomous variables. Low level of education included the categories of illiterate and primary education and high level of education included those of elementary education and above. The variable for the obstetrician's gender was grouped as having both genders in the team versus having only one gender.

Results
The average age of women giving birth in these hospitals was 26 years and the vast majority reported not working outside the home. There was a variation in the education level of women between the country sites with more women in the Syrian and Lebanese hospitals having low levels of education and more women in the hospital in Egypt with higher levels of secondary education. The average number of children was found to be higher in Syria (3.53) than in Lebanon (2.68) and Egypt (2.19) ( Table 1).
Thirty-three percent of women in Lebanon reported having both male and female obstetricians within the team providing labour and childbirth care, whereas the majority of obstetricians were males in the Egyptian site (72.5%) and females in the Syrian site (60.6%). C-section rate was the highest in the hospital in Egypt (32.4%) and lowest in the Syrian hospital (16.3%). Only 14% of women from Egypt reported having less than four antenatal visits, whereas 23.7% of women in Lebanon and 32.5% of women from the Syrian site reported it ( Table 1).
The mean childbirth satisfaction score as measured with the Mackey Childbirth Satisfaction Rating Scale in the total sample was high (X = 113.3; range 45-155). It was highest in Lebanon (X = 127.4; SD = 22) and below the mean score in Egypt (X = 99.5; SD = 20.5) and in Syria (X = 104.2; SD = 16.2). The satisfaction scores for the sub-dimensions differed also with the highest scores obtained in the sample from Lebanon and the lowest scores from Egypt except for the subdimensions of baby and nurse/midwife where the levels were lowest in Syria. The perception of labour control as measured by the Labour Agentry T Kabakian-Khasholian et al. Reproductive Health Matters 2017;25(S1):S16-S26 S19 Scale (LAS) was around the average score (X = 44.9; range 10-70) and lowest in Egypt ( Table 2).
The associations between the variables and with sub-dimensions of the satisfaction scale as well as the total childbirth satisfaction level were calculated. Women's education, number of children, the country site, gender of the physician and perception of control were found to be significantly associated with childbirth satisfaction (p = .000) whereas type of birth was not shown to be statistically associated with satisfaction (p = .220). Women's education, gender of the physician, perception of control and number of children were entered into a multiple regression model to identify the determinants of overall satisfaction separately for each country site whereas the gender of the physician was removed from the model combining the sample from all three sites. Table 3 shows that in the sample from Egypt, women's satisfaction increases with an increase in their number of children (p < .01) and their perception of control during labour (p < .000). Similar predictors are identified for the sample in Lebanon (p < .000). As for the sample from the Syrian site, an increase in the level of perceived control during labour predicts an increase in satisfaction level (p < .000) and having physicians from both genders in the team decreases women's satisfaction levels with childbirth (p < .05). When the sample from all three sites is combined, perceived control during labour (p < .000) and number of children (p < .01) remain significant predictors of increased levels of

Discussion
This study assessed satisfaction and perceived control in childbirth among a large sample of women giving birth in three public hospitals in Egypt, Lebanon and Syria. Our findings revealed a high level of satisfaction and an average level of perceived control in labour in each country and in the overall sample of women giving birth in the three public hospitals. Perceived control during labour was a significant predictor of improved satisfaction levels among women in the three sites of the study. In addition, having few children in the sample from Egypt and Lebanon and receiving care by a team including both male and female physicians in the Syrian hospital decreased satisfaction. The lack of choice given to women in the care received during labour and birth and lack of awareness about alternative options are known to result in women's low expectations during childbirth 39 and might partly explain the high level of satisfaction found in this study when considering the type of care provided in these hospitals. The fact that women's interviews were conducted in the early postpartum period at the hospital can also play a factor in the reported high satisfaction levels. The initial relief and the feelings of physical and emotional vulnerability have been previously reported to mask any negative feelings that women might have. 20 Similar high levels of satisfaction were reported in a sample of women with hospital birth in Belgium and Netherlands using the same scale. 40 Contrary to our findings, a study in Jordan found high levels of dissatisfaction with maternity care when women were interviewed during their seven-week postnatal visit to public clinics. 41 Hodnett 5 argues that there is no optimum time to measure satisfaction in childbirth and that it is highly dependent on the purpose of the study. As mentioned earlier, our purpose in this study was to observe differences in satisfaction due to the planned intervention, and the data reported here was used as the baseline. Although the mean satisfaction score was high, there were slight variations in the levels between the three sites, something that might be attributed to the variations in the process of care between the hospital in Lebanon, where women are offered individual labour and birthing rooms, and the hospital in Egypt where crowding and lack of privacy were found to be major complaints expressed by women. 36 These differences might also be explained by the context of conflict and displacement that affect Syria and Lebanon.

S21
The hospitals from these two countries included in the study provide services to refugees and to internally displaced populations, therefore it is expected that these women will be reluctant to express criticism about the care they received while feeling "grateful" for accessing a hospital and giving birth to a healthy infant. Perceived control during labour was found to be at an average level in the overall sample with slightly higher scores in Lebanon and Syria than in Egypt. It is known that personal control can be achieved by providing professional or lay support with the aim of empowering the woman to endure pain and feel a sense of accomplishment and self-efficacy. 2,5,8,12,40 In this regard, the lack of opportunities to participate in the decision-making process in these settings 33,[41][42][43] as well as the absence of continuous professional or lay support during labour and birth might have played a role in reducing women's feelings of control. In fact, higher scores observed in Lebanon compared to the other sites may be explained by the availability of lay support during labour.
The finding that women with high perceived control reported more childbirth satisfaction in this study is similar to what has been reported in the literature where personal control is identified as the most important predictor of childbirth satisfaction in different settings. 1,2,[8][9][10]12,40 Women's satisfaction with childbirth improved with an increased number of children, specifically in the samples from Egypt and Lebanon. These women would have prior experience with the system of care and therefore have already shaped their expectations according to the routines followed at these facilities. The lack of this relationship in the sample from Syria could be attributed to the lack of previous experience with birthing in a hospital for women displaced from rural to urban areas due to the prevailing conflict.
Satisfaction scores in this study tended to be lower with increased education levels. Women with a low level of education were more satisfied than their counterparts with a higher level of education. This trend, although not statistically significant, was mainly observed in the sample from the Lebanese site. As mentioned earlier, low expectations from the system of care among this group of women can explain this finding, as well as their reluctance to express dissatisfaction particularly when being interviewed in the hospital setting. We had previously documented women's low expectations through a qualitative study conducted at these sites 36 revealing that women arrive at these hospitals aware of the little that is being offered in the public system and accept the policies and procedures without challenging them. Similarly, women's low expectations and their reluctance in reporting dissatisfaction have been documented in a previous study in Lebanon, where women related satisfaction to the outcome of the birth. 44 Women will report negative birth experiences when their expectations and their actual experiences do not match. 5 We believe that this mismatch was not experienced by women in this study due to their initial low expectations. There might be some variation in expectations by educational level in this group of women that warrants further research. The lack of association of satisfaction with the type of birth in this study is similar to some other settings. 45,46 The rising caesarean sections in the three countries might have shaped women's expectations by normalising this practice and increasing its acceptance, therefore lowering expectations of having a vaginal birth.
The gender of obstetricians present during birth was another variable that was found to be associated with women's satisfaction with their childbirth experience. It is to be noted that women in the hospitals in this study are not entitled to choose their health care providers. The hospitals in Lebanon and Syria were found to have teams of both male and female obstetricians, residents, medical and nursing students attending the birth. This was a factor inducing dissatisfaction among women in the Syrian hospital compared to being offered care by either a male or a female physician. This finding needs to be considered together with the fact that the majority of the women in our sample from the Syrian site received care from a female physician. Women in Arab cultures prefer female obstetricians as they feel more at ease communicating about personal issues and undergoing vaginal exams. 44,46 The presence of a male provider in the team together with female providers is considered disrespectful in this context and influences their satisfaction with care.
There are a number of limitations identified in this study. The fact that this study is reporting from data gathered for an intervention project did not allow us to explore all potential elements related to women's satisfaction. Moreover, interviewing women at the hospital in the very early postpartum hours or days has most probably shifted satisfaction levels towards more favourable ones. It is recommended that future research in Table 3. Simple regression analysis for the variables predicting total childbirth satisfaction by country and for the total sample Variable Egypt N = 709 this area accommodate the expected changes in women's childbirth experiences over time. Qualitative studies, offering in-depth understandings of the concepts of control and satisfaction in Arab culture would make important contributions to the efforts on humanisation of childbirth conducted in other regions 23 as well as to the global discourse on disrespect and abuse in childbirth. 47 This comparative multi-country study provides an insight into women's experiences of childbirth and their satisfaction in Arab countries. The study provides contextualised information on the importance of women's feelings of control in shaping their experience with labour care and may have important implications on policy, practice and research. More research is needed to understand the different dimensions of women's experiences of childbirth in different health care systems and different regions of the world.