Consumer demand for caesarean sections in Brazil: informed decision making, patient choice, or social inequality? A population based birth cohort study linking ethnographic and epidemiological methods

Introduction

Of 12 Latin American countries recently reviewed, Brazil had the second highest rate of caesarean sections1; 55% of women from families earning more than $1000 (£700, 1120) per month had a caesarean section, above the 15% recommended by the World Health Organization. 2 3 Factors other than clinical indications play a major part in this inequitable distribution of caesarean sections.

Women's requests for caesarean section may be an important determinant of birth outcome, particularly in countries with growing privatisation and options for patient choice. 4 6 7 8 Most research focuses on women's fears of the physiological consequences of vaginal delivery and their desire to keep their sexual performance intact. 5 9 For this reason the debate has focused on providing consumers with knowledge on the risks associated with vaginal and operative deliveries so that decisions about birth may be rationally informed.10 11 12 13 14

Such approaches tell us little about how individual preferences for operative deliveries have grown from particular socioeconomic processes. How do women and obstetricians conclude that the risk of attempting a vaginal delivery is too high? 13 14 We investigated why some women seek out medicalised care. In particular, we explored how medicalisation is linked to social power in a society where the everyday experience of inequalities is profound.

Methods

We used a linked ethnographic and epidemiological approach for our study. For the ethnographic study we selected 80 mothers from a birth cohort of 5304 women who gave birth in 1993 in any of the hospitals in Pelotas, Brazil.2 We stratified these mothers by education and age, and we chose them at random until each quota was completed. We interviewed women two or three times during the first five months after birth, using a semistructured guide to explore women's health seeking behaviour, interactions with medical institutions, social integration and activities, and dynamics within the household.

We also interviewed 19 clinicians: six obstetricians, six paediatricians, four general practitioners, and three nurses. We taped and transcribed all interviews. Our analysis combined general overviews of thematic areas and a comparative approach. Gestational risk was measured with the score developed by Chamberlain and previously used in Pelotas.15

Qualitative research is based on the close observation of locally meaningful practices and their patterning. We created two scores by using simple dichotomous (yes or no) variables that, according to previous ethnographic understanding, were deemed to relate to a common underlying theme. These scores were not constructed in reference to an absolute gold standard but rather were used for their relative values as simple tools in the comparative analysis of women's experiences.

The maternal score for medicalisation is the sum of several variables: woman went to the hospital as soon as she felt any signs of labour, woman actively sought ultrasonography during the pregnancy, woman explicitly did not want to know what was going on during the birth, and woman stated she had expected a caesarean section or wanted one before entry to hospital. The score for socioeconomic power was based on several variables: woman stated she worked because she liked to have her own money, woman took part in decisions about the household, woman took money or goods without asking partner or family members, woman did not complain about how money was distributed in the home, woman acquired goods or services informally through exchange with others, and woman appeared more independent from partner or parents than other mothers in the sample. These two constructs represent underlying empirical issues that are important in several societies.16 17 18

Results

In Pelotas over 99% of births occur in hospital, most of which are dealt with by an obstetrician.19 No significant differences were found between the epidemiological and the ethnographic samples. In the epidemiological sample, 17% (923) of the mothers were under 20, 19% (984) had a family income under $100 (£70) a month, 28% (1492) had had less than five years of schooling, 35% (1860) were primiparous women, and 19% (639) of those with one or more previous deliveries had had a caesarean section. Around 30% (1619) of births were by caesarean sections and 32% (1695) were induced.

Results from the epidemiological study are similar to those from the ethnographic subsample (table 1). Caesarean sections were more common among wealthy and educated women, those with more antenatal attendance, primiparous women and, paradoxically, those with a lower gestational risk. Overall, 83% of women who had had a caesarean section had repeat procedures. Apart from primiparous women, those with the greatest need for caesarean sections were the least likely to receive one. The ethnographic study investigated how this socioeconomic reality was experienced by both physicians and women.

Physicians noticed an increase in their patients' awareness of medical knowledge and the possibility of legal action, previously only found among patients from higher social classes. For some, this led to a loss of control over patients (box). Before arriving at the hospital 32 women (40%) from the ethnographic study said they expected to have a caesarean section. A vaginal birth was considered a risky and negative experience, whereas caesarean sections represented the best quality care. When asked why women preferred caesarean sections, almost all recounted problems with fetal distress and mortality, excessive pain, or trauma to the vagina (see box).

According to some women, a traumatic vaginal birth often occurred because of medical negligence based on social and economic prejudice. Some explained that poor or uneducated women, teenage mothers, women with few antenatal attendances, those with “too many” children, and women who were believed to not want their child were least likely to receive any kind of medical attention, including interventions. These women were said to have to plead for induction by “making a scandal” during the birth to attract medical attention. For medical staff, however, women's “excessive” screaming indicated a lack of psychological preparation for birth and even desire for the child.

Attempts to avoid a vaginal birth were partly due to antagonistic relationships with medical staff. Women who did not have the money to pay for a caesarean section resorted to indirect methods, such as going to the hospital early in labour to pressure the on-call obstetrician for interventions, seeking an obstetrician who was known to perform caesarean sections (paying for the occasional private consultation), attending hospital outpatient care rather than local facilities to increase the chances of getting to know obstetricians (despite increased costs due to transportation), requesting ultrasonography, despite it being indicated only in high risk pregnancies in women in the public sector, and paying the on-call obstetrician half price for a caesarean section.

Discussion

Women's requests for caesarean sections in the absence of clear biological risks may seem irrational. However, traumatic birthing experiences often result from an antagonistic relationship with healthcare providers and a poor hospital environment—two aspects that are particularly acute for women who feel marginalised from society. What seems to underlie fear of vaginal delivery is not simply a lack of information on how to prepare for a vaginal birth but real issues relating to class based on differences in the quality of care provided.

Several studies exploring underlying influences with open ended methods found similar results.20 21 22 23 Indeed, many of the factors influencing maternal behaviours, such as fear of pain, are meaningful precisely because they are understood to differ by socioeconomic status and to be embedded in discriminating practices. In Brazil's plural healthcare system, there are alternative routes for women with marginal purchasing power to acquire a caesarean section other than by direct payment. After controlling for income and education, women's power to acquire a medicalised birth continued to have an effect on birth outcome.

In Brazil, one public health response has been to provide more education on the birth process, including risks associated with caesarean sections. This approach does not address the reasons for women's preferences nor that biological, institutional, and social variables are interdependent, as suggested by the holistic methods we used. Other studies have shown how health risks associated with birth progress are widely variable and sensitive to both the birthing environment and the nature of the doctor-patient relationship, making it difficult to standardise biological variables for “normal” births.14 18 24 25

It is questionable whether a clinical trial comparing deliveries by elective caesarean section with vaginal delivery would help weigh the risks of each birth type.26 Results from such a trial would have limited applicability because variables associated with actual hospital environments and sociobiological judgments made by physicians would not be taken into account. Indeed, research should explore the medical culture, including both in-depth ethnography of the obstetric subspecialty and observations of interactions between physicians and patients.

If poor quality care, stimulated by socioeconomic inequalities, is behind many women's preferences for a caesarean section, then arguments supporting interventionist birthing “for all” on the basis of equal access to health care must be reviewed. Debates about the reduction of unnecessary medicalised procedures should include, in addition to biological risks, questions of economic inequality, quality of care, and the social determinants of caesarean section rates. What initially seems to be a democratic or “liberal” stance in awarding all the same access to healthcare technology could in actuality be a conservative and medically radical solution to the social inequalities that have led to an increase in caesarean section rates. Additional research is needed in settings where the economically based distribution of caesarean sections is not as discrepant as in Brazil. Even modest inequalities in healthcare provision might create a market for unnecessary interventions among women who feel marginalised from access to medical technology.