Caesarean deliveries in China

This study of 112,138 births is one of the largest-scale surveys including indications for delivery conducted in China to date, covering three geographical regions, and focusing on non-indicated caesarean sections. It showed an overall caesarean section rate of 54.5% in 2011 which is substantially higher than the 46.2% found in the 2008 WHO study [1]. As an example, at the Beijing Obstetrics and Gynecology Hospital, the largest tertiary specialty hospital in Beijing, the caesarean section rate steadily increased from 19.5% in 1980–1984, to 25.4% in 1985–1988, then 35.3% in 1989–1992 [14], and the rate was the highest ever recorded at 47.9% in 2011 in this survey. The high caesarean section rate in China is concerning and efforts to analyze caesarean section rates and indications are important first steps in designing programs to decrease these rates.

In this study, 38.4% of all caesarean section were non-indicated, with 85.6% scheduled and performed before labour. Non-indicated caesarean section in this study accounted for 20.9% of deliveries compared to only 11.2% in the 2008 WHO study.

In China as early as 1996, caesarean section on maternal request or caesarean section for social factors received attention from researchers [15]. Previous studies also demonstrated that caesarean section on maternal request or caesarean section for “social factors” accounted for 35.9–46.2% of all caesarean section in certain areas of China [7]. In this study 11.9% of deliveries were caesarean section on maternal request compared to rates estimated at 2.5% in the USA or 1-2% in the United Kingdom. Caesarean section on maternal request was the most common recorded reason for caesarean section in this study. The non-indicated caesarean section rate varied greatly by level of facility and region and was much higher in the secondary level hospitals than in the tertiary level hospitals. This suggested that clinical practice patterns affect the number of caesarean sections performed. In China in 2011, there was not a gatekeeper system or a pre-operative view required for caesarean section [16] so if a woman requested a caesarean section, some hospitals were more likely to comply.

Prior studies in China have reported that fear of pain and perceived better health for the child and mother are the main reasons women request caesarean section [17‐19]. In China in 2011, most public hospitals were unable to routinely offer pain relief or epidurals in labour making the fear of pain a real concern. Emotional and nursing support for women in labour requires adequate nurse to patient staffing ratios, which may be lacking with the current system.

CPD diagnosed prior to labour was the second most common coded reason for non-indicted caesarean section, accounting for 20% of the non-indicated caesarean section or 4.2% of total deliveries. Even in the tertiary hospitals, CPD prior to labour accounted for 18.9% of the non-indicated caesarean section or 3.6% of total deliveries. In most hospitals in China routine clinical pelvimitry is performed antenatally. Previous Chinese studies classifed this as a medical indication for caesarean section. Despite abandonment of this practice in the USA and UK due to its poor predictive value for failure to progress in labour, in China it is a common physician documented reason for caesarean section.

In China, tertiary level hospitals often provide leadership in clinical teaching, research, and development of clinical guidelines. Physicians from tertiary hospitals may be considered more informed than those at secondary level hospitals, but this study suggests a wide gap between the tertiary level physician practice and international guidelines. The development of standardized evidence-based guidelines in obstetrics may help obstetricians decrease the number of non-indicated caesarean section.

In China, high risk pregnant women delivered more often in the tertiary hospitals. The higher overall caesarean section rate in tertiary level hospitals compared to secondary level hospitals may be partially explained by the higher risk patient mix at the tertiary level hospitals.

In this study 33.5% of deliveries were caesarean section with documented indications, with 29.3% of primary caesarean section with indications. Among the indicated caesarean sections, the most common diagnosis was NRFHRT or foetal distress (12.8% of all caesarean section, 7.0% of all deliveries). This is consistent with a previous study of teaching hospitals in China [20] and is higher than that in the United States (4.0-5.9%) [21, 22]. The rate for this indication was much higher in the tertiary hospitals than in the secondary hospitals in this study. NRFHT can be a relatively subjective indication for caesarean section [22, 23]. In this study NRFHT was diagnosed before labour 36.0% (2824/7840) of the time. This finding may be partially explained by the common overuse of technology in low risk pregnancies in China. For example, in many hospitals, the standard of care is to perform weekly non-stress tests (NSTs) in all healthy pregnant women in the third trimester. The positive predictive value of these tests is poor when used in low risk populations. Isolated abnormal umbilical cord dopplers or nonspecific electronic foetal monitor findings may result in unnecessary caesarean section for “foetal distress”.

The non-indicated caesarean section rate would be even higher if some diagnoses that we included in the “indicated” category such as preeclampsia/eclampsia/HELLP, oligohydramnios, third trimester bleeding (previa/accrete/abruption placenta), multiple gestation, and suspected macrosomia were further analyzed and re-categorized since they are not globally accepted and evidence based indications for caesarean section. In this analysis we were unable, for example, to discern which pregnancies had placenta previa (an accepted caesarean section indication) versus an abruptio placenta without intrauterine foetal death (not a caesarean section indication) or which multiple gestation cases had a non-vertex presenting twin, higher order multiples or non-concordant breech second twins (all accepted caesarean section indications). Thus with rigorous auditing of the “indicated” category of cases, many would be reclassified as “non-indicated”, raising the overall non-indicated caesarean section rate.

Isolated oligohydramnios is also not an evidence based indication for caesarean section.

Consistent with previous research in China, our study showed that isolated oligohydramnios was one of the major indications for caesarean section in China, accounting for 3.9% of all caesarean section. A prior study involving 2326 women of rural eastern China showed that, during antenatal care, 46.8% women received at least 3 ultrasound scans and the maximum number reached 11 [8]. It showed a statistically significant association between the number of antenatal ultrasound scans and caesarean section rates. A growing body of evidence demonstrates that isolated oligohydramnios in the absence of other maternal or foetal risk factors is not associated with adverse perinatal outcomes [24]. Caesarean section for preeclampsia could also be reassigned to non-indicated caesarean section if we were able to analyze the cases with more clinical information.

In this study, the overall operative vaginal delivery rate was very low at 1.2% and even lower in the secondary hospitals. This is consistent with Wang’s report, which surveyed 887 health facilities in the year 2002 and reported a national caesarean section rate of 38.0% and an extremely low operative vaginal delivery rate (5.5%) [25]. Low operative vaginal delivery rates may be due to pressure on the caregivers to practice “defensive medicine” and lack of training and experience with the techniques. Caesarean section for malpresentation accounted for 3.2% of all deliveries. Perhaps due to the same pressures few hospitals offer external cephalic versions.

Numerous studies of caesarean section have shown a strong association between increased maternal education, older maternal age and low parity and an increased risk of caesarean section [6, 15, 16]. In China, tertiary hospitals are located in big cities, treating mainly urban patients or rural patients referred from secondary and primary hospitals. Secondary and primary hospitals are usually in moderate or small cities, treating patients from towns or rural areas. In this study, the tertiary level hospital parturients were older, more educated, more likely to be having their first child later, pregnancy complications and indicated caesarean section.

The major strength of this study is that it is one of the largest-scale surveys conducted in China to date, covering three geographical regions, and focusing on non-indicated caesarean section in China. The biggest limitation of this study is that the facilities were not randomly selected, which may have introduced selection bias. Thus the caesarean section rates should not be regarded as representative for the entire country, nor for all the regions. Furthermore, our survey included only the secondary and tertiary hospitals, so our results cannot be generalized to smaller facilities. Second, prior studies show an association between insurance coverage and caesarean section but this cannot be analyzed since insurance data was not recorded in our survey. Third, we have no data on labour management including: induction of labour, augmentation of labour, or the availability of anesthesia. Furthermore, In the paper based on the same database, which we published in Chinese in 2014 [26], we put the indictions, such as cephalopelvic disproportion prior to labor, Age >35,etc. into indicated caesarean section as we used to do, it showed that the non-indicated caesarean section rate was 24.553%, was lower than that of this article, we have corrected in this article. Even so, there is some misclassification bias given that many of the caesarean section classified as “indicated” would likely be “non-indicated” if we had more clinical data on the cases. This suggests that we have underestimated the number of non-indicated caesarean sections.

The ideal caesarean section rate is not known. One ecological study using longitudinal data in 159 countries showed caesarean section rates higher than around 10% at the population level are not associated with decreases in maternal and neonatal mortality rate [27]. Another suggested that rates up to approximately 19% were associated with lower maternal and neonatal mortality rates in WHO member states [28]. Regardless, the caesarean section rate in China in this study of 54.5% is well above accepted target rates with at least 38% that were non-indicated. The motivations for providers are to perform caesarean sections for non-medical, vague or subjective indications is a complex issue. A broader perception of caesarean section as safe has been raised as a possibility. The use of globally recognized indications that are evidence- based to create clinical obstetric guidelines could standardize diagnosis and treatment in pregnancy and result in fewer unnecessary caesarean section at tertiary and secondary level hospitals. Additionally, medical malpractice reform and focus on the birth experience may be a way to reduce caesarean section rates. Analysis of maternal and perinatal outcomes associated with mode of delivery in China needs further study. With the end of the One Child policy, as of January 2016, the risks associated with non-indicated caesarean section will likely increase as more women are allowed and choose to have a second child. This makes it imperative to reduce the number of unnecessary primary caesarean sections.