Factors associated with caesarean sections in Phnom Penh, Cambodia

Abstract

C-sections are an increasingly performed medical practice which can save lives but may also lead to major complications. Through a mixed methods study conducted in 2015 in Cambodia, we aimed to analyze the reasons for requesting a c-section and to explore factors that are associated with c-sections. 60% of the women in the cohort who gave birth by c-section reported having requested it. Through 31 in-depth interviews, we determined the reasons given by women for requesting a c-section before and during labour. Before labour, reasons for requesting a c-section were: choosing the delivery date; bringing luck and joy to the family; protecting the genitals, and the belief that c-section is safer for the mother and for the baby. Reasons given during labour were fear, pain, and having no more energy. We also observed two major factors driving the women’s request for a c-section: family support for requesting a c-section, and the over-usage of ultrasound examinations. Our multivariate analysis of the interviews of 143 women before and after delivery showed that having a previous c-section, delivering in a private facility, being older than median at the time of sexual debut, residing outside of Phnom Penh and having the delivery costs covered by the family were all factors independently and significantly associated with a higher chance of c-section delivery. We conclude that women are not well informed to give consent for c-delivery, and that their request is often affected by false belief and poor knowledge.

Résumé

La césarienne est une pratique médicale de plus en plus fréquente qui peut sauver des vies, mais aussi provoquer de graves complications. Dans une étude à méthodologie mixte réalisée en 2015 au Cambodge, nous avons analysé les raisons de la demande de césarienne et exploré les facteurs qui sont associés à cette pratique. 60% des femmes de la cohorte ayant accouché par césarienne ont indiqué l’avoir demandé. Au cours de 31 entretiens approfondis, nous avons déterminé les raisons données par les femmes pour demander une césarienne avant et pendant l’accouchement. Avant l’accouchement, les raisons étaient: choisir la date de la naissance ; apporter chance et joie à la famille ; protéger le périnée ; et la croyance qu’une césarienne est plus sûre pour la mère et l’enfant. Les raisons données pendant l’accouchement étaient la peur, la douleur et l’épuisement. Nous avons aussi observé deux facteurs majeurs incitant les femmes à demander une césarienne : le soutien familial et la surconsommation des échographies. Notre analyse multivariée des entretiens avec 143 femmes avant et après l’accouchement a révélé qu’une césarienne précédente, le fait que l’accouchement se déroule dans un centre privé, que la mère soit plus âgée que la médiane au début des relations sexuelles, qu’elle réside en dehors de Phnom Penh ou que la famille prenne en charge les frais de l’accouchement étaient autant de facteurs indépendamment et significativement associés à une probabilité accrue de césarienne. Nous en concluons que les femmes ne sont pas bien informées pour donner leur consentement à une césarienne et que leur demande est souvent influencée par des croyances erronées et des connaissances inexactes.

Resumen

La cesárea es un procedimiento médico cada vez más practicado, que puede salvar vidas pero también puede producir complicaciones mayores. Por medio de un estudio de métodos combinados realizado en 2015 en Camboya, buscamos analizar las razones para solicitar una cesárea y examinar los factores que están asociados con las cesáreas. El 60% de las mujeres en la cohorte que dieron a luz por cesárea informaron haberla solicitado. Por medio de 31 entrevistas a profundidad, determinamos las razones dadas por las mujeres para solicitar una cesárea antes y durante el parto. Antes del parto, las razones para solicitar una cesárea fueron: escoger la fecha del parto; traer suerte y alegría a la familia; proteger los genitales; y la creencia de que la cesárea es más segura para la madre y para el bebé. Las razones dadas durante el parto fueron: temor, dolor y no tener más energía. Además, observamos dos factores importantes que impulsan a las mujeres a solicitar una cesárea: el apoyo de la familia para la solicitud de la cesárea y el sobreuso de exámenes por ultrasonido. Nuestro análisis multivariante de las entrevistas con 143 mujeres antes y después del parto mostró que haber tenido una cesárea anterior, dar a luz en una unidad de salud privada, ser mayor que la edad media en el momento de la iniciación sexual, vivir fuera de Phnom Penh y tener los gastos del parto cubiertos por la familia, todos estos fueron factores asociados independiente y significativamente con mayor probabilidad de parto por cesárea. Concluimos que las mujeres no están bien informadas para dar su consentimiento para una cesárea y que su solicitud a menudo es afectada por falsas creencias y conocimientos incorrectos e inexactos.

Keywords:

• caesarean section
• Cambodia
• gender
• social control
• overbiomedicalization
• mixed method

Introduction

Access to a caesarean section (c-section) is a crucial tool to reduce maternal mortality. However, a 2015 WHO statement mentions that c-sections are “effective in saving maternal and infants’ lives, but only when they are required for medically indicated reasons”. It goes on to say that, at a population level, c-section rates “higher than 10% are not associated with reductions in maternal and newborn mortality rates,” and that c-sections should be performed only when medically justified. ¹ C-section is a surgical intervention and, while it can save lives, it may lead to complications. ² C-sections, particularly when not justified by a medical condition, expose women to well-documented risks of maternal mortality, depression and higher morbidity. ^2,3 They lead to more hospital-acquired site infections, and delays in breastfeeding. ^4,5 C-section can also compromise future pregnancies, and have long term negative effects for the newborns, such as obesity, asthma, type-1 diabetes, and lead to unexplained stillbirths in the second pregnancy. ^2,6–12 These risks explain why c-section on demand is considered, “an expensive and dangerous luxury”. ¹³ Current evidence strongly advises against unnecessary c-sections. ²

Despite different recommendations ¹ and warnings about its risks, c-section rates have been constantly and globally increasing, particularly in middle- and high-income countries, reaching “epidemic” proportions in some countries, ^14,15 and the increase in c-section rates shows no signs of slowing down. ¹⁶ This rise has become a major public health concern and a cause for worldwide debate. The root causes for this dramatic increase are not yet fully understood and are often ascribed to the complex interplay of different governmental and health institutions with familial and social dynamics, sometimes including factors as peripheral as fashion and media. ¹⁶ Due to the complexity of this issue, it is highly unlikely that the responsibility for this increase lies exclusively with maternity healthcare workers. Since the 2000s, evidence shows that increasing numbers of women are requesting delivery by c-section. ¹⁷ A review of the literature on c-section requests highlights fear of poor or discriminatory care, lack of control in labour, pain, stress and anxiety as contributing to the desire for a c-section. ¹⁸

A WHO Survey conducted in 359 facilities worldwide found c-section rates in Asia ranging from 18.6% in Japan to 47.6% in China. ¹⁹ In Cambodia, estimates range from 14% to 30%. ^4,19,20 Of 21 countries surveyed in 2010-2011, Cambodia had the largest increase in c-section rate (+ 16.8% per year between 2007-2008 and 2010-2011). ¹⁹ Comparisons of successive Cambodian Demographic and Health Surveys (CDHS) also suggest a sharp increase from 0.8% to 6.3% in the country, and from 4.6% to 14.4% in Phnom Penh between 2000 and 2014. ^21,23

The situation in the province of Phnom Penh is of particular interest, as it reported the highest c-section rate in the country in 2014. ²¹ Our previous research in four maternity clinics in Phnom Penh showed a c-section estimate of 27% in 2015 (n = 5855 c-sections/21,673 deliveries). ²² In 2015, Calmette Hospital was the largest maternity center in Phnom Penh, with 11,080 deliveries, of which 26% were done by C-section. Calmette is a half-private, half-public university hospital, established by the French in 1959. This hospital is known for catering to clients of higher socio-economic class. The cost of an overnight stay ranges from 10 to 250 USD, allowing poorer women to deliver there also. For all patients, a vaginal birth costs 60 USD, a c-section 278 USD.

To our knowledge, no social science study has been conducted in Cambodia to understand this increase. Cambodia has recently moved from a poor to a lower-middle income country and has entered a demographic transition. ²⁴ The total fertility rate decreased from around 7 to 2.7 children per woman between the early 1980s and 2014 respectively. ^21,25 The maternal mortality ratio dropped from 1020 to 161 maternal deaths per 100,000 live births between 1990 and 2015, allowing Cambodia to meet the Millennium Development Goal (MDG5) by reducing the maternal mortality rate by 75%. ²⁶ The number of institutional births jumped from 10% in 2000 to 83% in 2014. ^21,23 However, in 2014, only 16% of women were covered by health insurance, ²¹ meaning that more than 80% of women still have to pay for their delivery. We observed that institutional deliveries were linked with more routine episiotomies, delayed breastfeeding and an increasing rate of c-section on request. Both public and private health sectors are quickly expanding their medical services as ways of adding revenue sources. The Cambodian health system is currently witnessing an intense circulation of both medical professionals and patients. Most obstetricians work part time in several health facilities while women attend various maternity clinics during their pregnancy, often choosing the place where they will deliver at the last minute. Previous research in Phnom Penh found that some women were requesting c-sections to protect their genitals, or in fear of a painful delivery, and that families had an influential impact on the process of childbirth. ²² Using a mixed-methods study, our goal was to better describe women’s desire for c-section and the other factors associated with c-section in the largest public maternity clinic of Phnom Penh, Cambodia, in 2015.

Methods

Study sources

Analyses were performed using two sources of data collected from January to April 2015: a prospective cohort of women interviewed during their pregnancy and after delivery, and in-depth semi-structured interviews with skilled birth attendants, mothers in post-partum wards and Cambodian men.

Prospective cohort

A prospective cohort of 146 pregnant women was constituted by systematically inviting all 156 pregnant women with more than 30 weeks gestation presenting for antenatal care at Calmette hospital between 2 February and 5 April, 2015 to join the study. Previous research that we had conducted in this hospital helped us to build trust with its management and healthcare staff.

Of 156 women contacted during their last antenatal consultations, 10 refused to participate for lack of time, 21 finally delivered in a private clinic but not Calmette, but were still included in the final analysis; 143 were re-interviewed after delivery. Socio-demographics and obstetrical factors, knowledge and perceptions on c-sections, women’s preference for a route of delivery, and conditions of the actual delivery were explored using a standardized questionnaire including eight questions previously validated. ¹⁴ All interviews were conducted in a private room to ensure confidentiality by a socio-demographic researcher and an interpreter not affiliated with Calmette hospital.

In-depth interviews

We also conducted 31 in-depth interviews with two midwives, 12 obstetricians (three male and nine female), 10 mothers admitted to the post-partum ward at Calmette hospital, and seven men living in Phnom Penh. Midwives were selected because of their ability to speak French. Interviewed obstetricians were working daily in maternity wards. Mothers were selected because of their availability in postpartum wards during the days of the research or because of interesting quotes during the quantitative interview. Men were recruited using the “snowball” method. The first one was the tuktuk driver of the first author. He introduced her to the other men incrementally.

Questions asked of midwives and obstetricians related to the practice of c-sections. The questions for the mothers related to their pregnancy follow-up and delivery. Questions asked of men related to body, birthing, and sexuality. Interviews were conducted until saturation of the main topics was obtained, focusing on key points of the interview but not limited to pre-determined topics, and allowing the generation of unexpected information and themes. Interviews lasted on average 40 minutes.

Mixed-methods analysis

Analyzing the factors associated with a c-section

Bi- and multivariate comparisons were done between women who underwent a c-section and those who did not to identify factors associated with a delivery by c-section. The three women who could not be re-contacted after delivery were excluded from these analyses. Quantitative comparisons were done using Stata 13 (Stat Corp., College Station, Tx, USA) software, in bivariate analysis first. The Student test, Wilcoxon rank sum test, Chi-squared test or Fischer’s exact tests were used when appropriate. Significance of statistical tests was assessed at P < 0.05.

Multivariate logistic regression was conducted by initially introducing into the model all variables significantly associated with a c-delivery in bivariate analysis with P-values < 0.2. We followed a backwards stepwise selection procedure to leave only the factors associated with a c-section with a P-value < 0.05 in the final model. Models were compared one after another during the backward procedure using the log-likelihood test. Final model fit was assessed by its log likelihood test and the Akaike’s information criterion (AIC). The area under the ROC curve was also calculated to estimate the overall discriminatory power of the model. Odds ratios (OR) and their 95% confidence interval (CI) are only presented for variables remaining independently and significantly associated with a c-section.

Understanding the socio-demographic reasons for c-section

Reasons explaining the c-section rate were also explored using a qualitative analysis of the 31 in-depth interviews. Qualitative interviews were recorded and transcribed. Social processes that led to women’s demands for a caesarean delivery were analyzed using N-Vivo software. Content analysis involved reading and examining transcripts to develop a sense of the themes and sub-themes. This inductive thematic approach allowed us to identify 9 noodles and 16 sub-noodles based on 122 references. We aimed to understand whether women and their families were asking for a c-section, how they saw it and why they requested the procedure. We also explored how birth-practitioners address these demands and how negotiations between them and the mothers happen. We then articulated our quantitative and qualitative data to understand more precisely the construction of the c-section requests.

Ethics

The study was approved by the Cambodian National Ethic Committee for Health Research in May 2014. All study participants gave informed consent before any information was anonymously recorded.

Results

Cohort characteristics

The socio-demographic characteristics of our prospective cohort are shown in Table 1 . The median age was 27 years (IQR: 25–31) and 29 years (IQR 27–31) among women who had a vaginal and a caesarean delivery, respectively (p = 0.03). Most women (86.7%) lived in the capital city. Almost all of them (98.6%) were living with the father of the expected child at the time of the first interview. The two groups were also comparable in terms of education and occupation: 25.9% of all women enrolled had limited incomes, although 85.3% had reached at least secondary education. Obstetrical characteristics of enrolled women are presented in Table 1. The analysis did not show any difference in parity and gestational age between the two groups (53.1% were nulliparous, 62.9% delivered at term).

Table 1 Socio-demographic and obstetrical characteristics of the study population, bivariate analysis, Cambodia, 2015.

	All		Delivery by c-section
				No	Yes
	n = 143	% or IQR	n = 94	% or IQR	n = 49	% or IQR	P
Sociodemographic
Age (median, IQR)	28	[26–31]	27	[25–31]	29	[27–31]	0.03
Residence in PP	124	86.7	85	90.4	39	79.6	0.06
Province of origin							NS ^†
Phnom Penh/Kandal	88	61.5	57	60.6	31	63.3
Adjacent provinces	39	27.3	27	28.7	12	24.5
Distant provinces	16	11.2	10	10.6	6	12.2
Marital status
Currently with a partner	141	98.6	93	98.9	48	98.0	NS
Currently without partner	2	1.4	1	1.1	1	2.0
Occupation							NS
Employee	37	25.9	24	25.5	13	26.5
Seller, shop keeper	48	33.6	30	31.9	18	36.7
Worker	23	16.1	13	13.8	10	20.4
Other	35	24.5	27	28.7	8	16.3
Education							NS
Primary school	20	14.0	14	14.9	6	12.2
Secondary school	76	53.1	51	54.3	25	51.0
University or equivalent	47	32.2	29	30.9	18	36.7
Place of delivery							0.02 ^*
Public hospitals	122	85.3	85	90.4	37	75.5
Private clinic	21	14.7	9	9.6	12	24.5
Price of admission night							NS
< 15 USD	50	35.0	33	35.1	17	34.7
15 to < 45 USD	55	38.5	34	36.2	21	42.9
45 USD and more	38	26.6	27	27.8	11	22.4
Obstetrics
Nulliparous	76	53.1	49	52.1	27	55.1	NS
Gestational age							NS
Premature < 37 weeks	52	36.4	32	34.0	20	40.8
At term	90	62.9	62	66.0	28	57.1
Post term > 41 weeks	1	0.7	0	0.0	1	2.0
History of c-section	14	9.8	2	2.1	12	24.5	< 0.001
Pathology during pregnancy	3	2.1	2	2.1	1	2.0	NS
C-section desired during pregnancy	11	7.7	3	3.3	8	16.3	0.007
Newborn’s weight (IQR)							0.007
Low birth weight (< 2500g)	4	2.8	3	3.2	1	2.0
Normal birth weight	132	92.3	90	95.7	42	85.7
Macrosomia (>=4000g)	7	4.9	1	1.0	6	12.2

Notes:

† NS: Not significant.

* p < 0.2, variables included in the multivariate regression analysis

Factors associated with c-sections in the prospective cohort

The bivariate analysis of the cohort showed that women who underwent a c-section were more likely to have expressed a desire to deliver by c-section during their first interview (during the pregnancy) than women who delivered vaginally (16.3% vs 3.3%, p = 0.007). Of the 14 women who had a previous c-section, 12 underwent an iterative c-section for this pregnancy (p < 0.001). Although 60% of women who delivered by c-section had requested it, only 49% of them were finally satisfied by this mode of delivery. They regretted and spoke of a negative experience whereas 99% of women who gave birth vaginally were satisfied. Women who had a c-section were more likely to be financially supported by their family for the delivery costs than women who delivered vaginally (18.4% vs 2.1%, p = 0.001). Women who had their first intercourse later than the median population were more likely to deliver by c-section (67.3% vs 48.9%, p = 0.04).

However, the multivariate analysis showed that five factors remained significantly associated with a higher chance of c-section (Table 3): a previous c-section; the woman’s family covering the delivery costs; delivering in a private facility; having the first sexual intercourse above median age; and residing outside of Phnom Penh.

Table 3 Factors independently associated with a caesarean delivery, multivariate logistic regression, Cambodia, 2015, n = 143 women.

Factors	Odds Ratio [95% IC]		P >
History of c-section	27.1	[4.8–152.6]	< 0.001
Family covering all or a part of the delivery costs	14.4	[2.5–84.7]	0.003
Delivery in a private facility	3.3	[1.1–10.2]	0.036
Above media age at 1st intercourse	2.9	[1.1–7.4]	0.031
Residing in Phnom Penh	0.3	[0.1–1.0]	0.046

Note:

ORs are only reported for factors independently associated with a c-delivery

AIC: 146.02; Log likelihood chi-square test: 53.81, p < 0.001; Area under ROC curve 0.8299

Knowledge and perceptions of the cohort

Of all enrolled women, only 44.8% had some knowledge of caesarean delivery (Table 2 ). The source of information was the family (25%), friends (23%), their own experience of a previous c-section (20%) and the neighbors (15%). Only 10% of them had received information from health workers. However, 91.6% of all women imagined that such a procedure would prevent them from getting a “floppy vagina”, 82.5% believed that it would prevent them from feeling pain, 52.4% and 46.2% stated that it would be beneficial for their baby’s and their own safety respectively, while 39.9% thought that it would preserve their beauty.

Table 2 Knowledge, perceptions and gender issues related to c-section, bivariate analysis, Cambodia, 2015.

	All		Delivery by c-section
			No		Yes
	n = 143	%	n = 97	%	n = 49	%	P
Knowledge and perceptions
Knows what is a c-section	64	44.8	40	42.6	24	49.0	NS ^†
Believes that preserves a woman’s beauty	57	39.9	35	37.2	22	44.9	NS
Believes that prevents ‘floppy’ vagina	131	91.6	86	91.5	45	91.8	NS
Believes that prevents pain	118	82.5	75	79.8	43	87.8	0.2 ^*
Believes that c-section is good for baby’s safety	75	52.4	44	46.8	31	63.3	0.06 ^*
Believes that c-section is good for mother’s safety	66	46.2	39	41.5	27	55.1	0.1 ^*
Gender issues
Age difference with partner							NS
5 years younger than partner or more	70	49.0	51	54.3	19	38.8
0 to 4 years younger than partner	64	44.8	38	40.4	26	53.1
Older than partner	7	4.9	4	4.3	3	6.1
Age of father unknown	2	1.4	1	1.0	1	2.0
Who pays for the delivery
Male partner pays all or a part	82	57.3	59	62.8	23	46.9	0.06 ^*
Woman pays all or a part	69	48.3	46	48.9	23	46.9	NS
Family pays all or a part	11	7.7	2	2.1	9	18.4	0.001 ^*
Others pay all or a part	6	4.2	3	3.2	3	6.1	NS
Husband’s education							0.07 ^*
Primary school	5	3.5	1	1.0	4	8.2
Secondary school	54	37.8	38	40.4	16	32.7
University or equivalent	65	45.5	40	42.6	25	51.0
Unknown	19	13.3	15	16.0	4	8.2
Age at first intercourse							0.04 ^*
< 23 y	64	44.8	48	51.1	16	32.7
>=23 y	79	55.2	46	48.9	33	67.3
Woman’s sexual debut at wedding time	106	74.1	67	71.3	39	79.6	NS
Woman decided when to resume intercourse after childbirth	47	32.9	30	31.9	17	34.7	NS
Male partner decided to resume intercourse	103	72.0	67	71.3	36	73.5	NS

Notes:

† NS: Not significant.

* p < 0.2, variables included in the multivariate regression analysis

Reasons for requesting a c-section

Of the 143 enrolled pregnant women, only 11 expressed a desire for a caesarean birth before the actual birthing event. In the end, 29 of the 49 women (60%) who ended up having a c-section reported having requested it during delivery. The sociological context and the women’s reasons for choosing a c-section are summed-up in the theoretical framework shown in Figure 1 . During their pregnancy, before the onset of labour, women made the following statements:

Figure 1. Factors constructing and driving the request for caesarean section in Phnom Penh.

A c-section allows a choice of the delivery date:

“As the Chinese, sometimes Cambodian people look for a date of birth. We ask for a specific date.”

(Birth Practitioner 11, female, 57 years old, Phnom Penh)

A c-section brings luck and joy to the family:

“[This woman requests a c-section because] it is « hong suy », the day of happiness … No [the monk did not choose the date], it is the predictor, it is the cards, the spirits, the mediums, etc. There are many different people!!”

(Birth Practitioner 8, female, 34 years old, Phnom Penh)

A c-section protects the genitals. All women, men and birth practitioners interviewed mentioned that giving birth vaginally is very likely to enlarge the vagina:

“More women want to do the c-section now. Because of the beauty of the perineum. Women think that the fetal head is large and that will enlarge their vagina. And get vaginal tears. It is not aesthetic. And it is not good for the sexuality neither.”

(Birth practitioner 4, male, 25 years old, Phnom Penh)

Investigator: And when we have a c-section, do we have a tight vagina?

Interviewee: Yes.

Investigator: That means that men like when women have a c-section?

Interviewee: Yes, for sure. But this is more for the rich and the celebrities.

(Male interviewee 1, 36 years old, Phnom Penh)

Delivery by c-section is not only considered safer for the mother and for the baby by women and their family, but also by some birth practitioners:

“Because normally for a c-section we can’t say that there are risks (…) Because for the delivery there are contractions, the fetus engagement, there are risks of uterine rupture, risks of hemorrhage, we do not know what to do for a prolonged pregnancy. A c-section is safer for the mother and the baby.”

(Birth practitioner 3, 29 years old, Phnom Penh)

During labour, others reasons were frequently given, such as fear of delivery and pain:

“I did an ultrasound during the ninth month and the baby’s head was not turned yet, I decided that if I could do a vaginal delivery, I would. But yesterday, I had contractions, from 4 am in the morning till 10 am. The doctor told me to wait. She told me that the uterus was opening. But I cannot wait. I feel too much pain. My baby stays high in my belly. So I am afraid. I feel that my baby stays high, I feel pain, his head did not turn, and the umbilical cord is around his neck. I am afraid for my baby. So I spoke with my husband. I told him we have to do a c-section.”

(Woman 2, 33 years old, Phnom Penh)

Furthermore, the notion of “having no more energy” (“oas kamlang”, in Khmer language) was brought up, which is part of a symbolic anthropology concept based on a system of bodily humors in Cambodia (ayurvedic influence from India).

“Some women tell us that they don’t have any energy anymore ‘oas komlang’.”

(Birth practitioner 7, female, 28 years old, Phnom Penh)

Women’s request for a c-delivery was also driven by demand from the family, as illustrated by this quote:

“I want to give birth vaginally. But I feel pain. Hence, my mother-in-law sees that it is very difficult for me; she […] tells me that I have to deliver by c-section. My mother-in-law asks the doctor to perform a c-section. And he accepts. She was afraid, really concerned. Because I had already a c-section in the past, I have a recent scar, she was really concerned.”

(Women 7, 33 years old, Phnom Penh)

Another contributing factor observed was frequent ultrasound examinations. Some women underwent up to ten ultrasound examinations during their pregnancy, the results of which frequently provided the justification for a c-section, for instance, a suspected “baby’s big head” or “having the cord around the neck”.

“I did eight ultrasounds and seven or eight consultations during the pregnancy. When I have some pain during the pregnancy, I go to do an ultrasound. As soon as there is something wrong, I go to do an ultrasound.”

(Woman 5, 28 years old, Phnom Penh)

Discussion

Beside the “usual” medical or socio-economic factors often associated with a higher chance/risk of c-section, such as a previous c-section, or having a family able to cover the costs of delivery, we have identified various other social considerations which drive women and their family to request a delivery by c-section.

Some of these social reasons are also found in other Asian countries. For instance, being able to choose the date of birth was decisive for the women we interviewed and their families. Chinese people also believe that choosing the right days for certain life events, like birth, can change a person’s life course. In Taiwan, c-sections are performed significantly more frequently on auspicious days. ²⁸

Having sexual debut later than the median age was another factor associated with c-section in our study. Our qualitative data also suggested that childbirth was strongly linked to concerns about sex in the interviews, a vaginal birth seen as potentially “reducing the women’s sexual capital”. Similar concerns have been reported in other countries. ^14,15,29 In Thailand, for instance, some women fear that their husband would be unfaithful after a vaginal birth and prefer caesarean deliveries. ³⁰ In Brazil also, the vagina is demarcated as the domain of sexuality rather than reproduction, and c-sections are thought to preserve the woman’s bodily integrity, saving her from distension of the vagina and damaging the perineum. ³¹ However, much of the vaginal damage following a delivery is caused by procedures performed to accelerate the birthing process, such as forceps, vacuum extractions and episiotomies. ¹³ Damage during vaginal birth has recently been observed in Phnom Penh where routine episiotomy is practiced in order to “help” women deliver faster. ²⁷

Lumbiganon and colleagues have shown that women often do not know that, when performed without medical justification, this surgical intervention may increase maternal and perinatal morbidity, causing an unnecessary additional health burden. ² In our study, many women mentioned that they felt a c-section would be safer for their baby and themselves. This was also the belief of some of the gynecologists we interviewed. In China, nearly half of the pregnant women deliver by c-section. ¹⁹ With the one-child policy, families do not want to take any risks, and non-clinical factors are the main drivers fueling the rise of caesareans in China. ²⁸ There has been a decrease of the total fertility rate in Cambodia ³² and this might bring families to adopt new strategies regarding their children and birthing process.

Inability to deal with delivery pain was another reason for women requesting c-section during their labour. Indeed, routine epidurals for vaginal birth are not available in most maternity wards in Phnom Penh. Many women, including a large majority of our cohort, think that a vaginal birth will be more painful than a c-section. ³³ Women and their families often do not have enough information about c-sections and do not have the capacity to choose. A woman choosing a c-section in order to avoid pain is misinformed:

“she exchanges 12 hours of labour pain for severe postoperative pain and debility and a longer recovery period with weeks or even months of pain.” ¹³

As shown in Brazil, ³¹ medically assisted childbirth with sub-optimal quality of care and quasi absence of anesthesia leads women to desire a c-section. Considering why c-sections were so quickly and easily adopted in Cambodia’s maternity wards while epidural anesthesia for vaginal birth was not, we hypothesise that c-sections address social expectations, such as choosing the delivery day and maintaining a sexual attractiveness for the husband, while a women’s labour pain is considered to be bearable.

As in other countries, we showed that c-sections more often occur in private health facilities. ¹⁸ Since the 2000s, the private sector has been flourishing. Because of their very low wages in the public sector, doctors are looking for additional sources of income in private clinics or through private practice in public health facilities. ^34,35 The private sector health market is mainly unregulated ³⁶ and this makes it susceptible to the promotion of unnecessary, but highly profitable, medical services like c-sections.

Finally, this research pointed out another unexpected issue, the over utilization of (also profitable) ultrasound examinations, requested by women to balance the sub-optimal quality of antenatal care and the lack of prenatal classes. Women pay for ultrasound to be reassured about their pregnancy and also to “test” different places and different medical doctors, in a context of medical consumerism and attractiveness of new technology. However, these ultrasound services do not bring satisfying answers to women who are in search of a human relationship. Ultrasound services often increase their misinformation and fear, and can even increase health workers’ fear of childbirth. Our finding on how ultrasound contributes to c-section requests is highly suggestive and needs to be further explored by other researchers in similar contexts.

Limitations of the study

This study has several limitations. First, it was conducted in a maternity clinic, which is not representative of Cambodia or even of Phnom Penh. However, it is one of the leading hospitals in the country, shaping national policies and teaching cohorts of medical, midwifery and nursing students. Second, our results may be biased toward more c-sections since the study site is a referral center for obstetrics. Indeed, residence outside of Phnom Penh was a factor independently associated with c-section. Third, the sample size of our cohort and of our group of interviewees was limited due to time and budgetary constraints. Findings were presented to and discussed with the obstetric team of Calmette in depth, to strengthen and refine our conclusions and to raise awareness on the c-section safety issues.

Conclusions

In the 2000s, making c-sections available in every obstetrical referral center has certainly contributed to the significant reduction in maternal and newborn mortality recently witnessed in Cambodia. However, the recent dramatic rise in the c-section rate raises the question of whether c-sections are being diverted from their medical purpose and used to address a social demand. Many sociocultural factors (choosing a day, preserving sexual attractiveness) influence Cambodian women to choose more c-sections. Women surrender their bodies to biomedical technology because they think that they are not able to give birth by themselves. Women seem to ask for a c-section under specific circumstances: their body controlled by pain, and they themselves controlled by fear and their family. Ultrasound may also be making women more inclined to ask for a c-section both before and during labour, thereby actively participating in the “culture of c-section” ¹⁴ in Phnom Penh. Information must be given and the decision of delivering a baby by c-section should be taken in light of the documented risks. Everything should be done to avoid medically unnecessary primary c-sections.

Contributors

All authors declare no conflict of interest.

CS and SG prepared the study protocol, wrote the paper and analyzed the qualitative (CS) and quantitative (SG) data. CS collected all quantitative and qualitative data. KLS and HR facilitated the data collection. VP commented on the article. All co-authors commented on, and approved the final version of the report.

Acknowledgements

This study was part of PhD work by Clémence Schantz, supported by the Université Paris-Descartes, France, and the Center for Khmer Studies. We thank Myriam de Loenzien, Leila Srour, Eric Opigez, Hubert Barennes for their advice. We thank the Ministry of Health in Cambodia (National Ethics Committee), the hospital staff in Calmette; Dr Pech Sothy, Dr Ek Meng Ly, Dr Prak Somaly and Pr Tung Rathavy for sharing their data with us and all the study respondents.