Abstract
Objectives. To describe the cesarean rates in different child policy periods and assess the medical necessity of cesareans during the 2-child policy period.
Methods. We collected hospital-level aggregate data on 93 745 deliveries and individual-level data on 27 977 deliveries from 6 hospitals in the Hubei and Gansu provinces of China from 2013 to 2016. Experts in gynecology and obstetrics assessed the medical necessity of 1024 randomly selected cesareans in 2016.
Results. The overall cesarean rate decreased significantly from 45.1% in the 1-child policy period (January 2013–September 2014) to 40.4% in the selective 2-child policy period (October 2014–July 2016) and further to 38.9% in the universal 2-child policy period (August 2016–December 2016). The rate of cesarean delivery on maternal request decreased by 46.3%, whereas the rate of cesarean delivery indicated by a previous cesarean delivery increased by 118.8% (P < .001). The experts assessed 222 (21.6%) cesareans as lacking medical necessity.
Conclusions. The overall cesarean rate in Hubei and Gansu provinces decreased after the implementation of the 2-child policy, and one fifth of cesareans might be nonessential.
The rates of cesarean delivery have reached “epidemic” proportions in many nations. In China, the cesarean rate increased dramatically during the period of the 1-child policy, with the overall rate increasing from 3.4% in 1988 to 34.9% in 2014.1,2 Cesarean delivery on maternal request (CDMR), a subtype of cesarean delivery without medical indication, has become a substantial driver of the cesarean delivery rate, accounting for an estimated 40% of cesarean deliveries in China in 2010.3
During the 1-child policy period, couples tended to choose cesarean delivery, which they perceived as the safest delivery method. Because these prospective parents were allowed to have only 1 child, they did not need to be concerned about the potential adverse effects of cesarean delivery on future pregnancies.1,4 In response to the slowing population growth and noticeable problems in demographic structure, the selective and universal 2-child policies were successively introduced in China.5 In November 2013, the selective 2-child policy was announced and specified that a couple could have 2 children if 1 spouse was an only child.6 In October 2015, the universal 2-child policy was announced and allowed all couples to have 2 children.6 Previous studies reported that the implementation of the 2-child policy affected maternal characteristics, women’s delivery intentions, and their choice of delivery mode.7,8 Thus, we hypothesized that the rate of cesarean delivery and the distribution of cesarean indications would subsequently change.
We aimed to determine the trends in the overall and indication-specific cesarean rates across the 1-child policy, the selective 2-child policy, and the universal 2-child policy periods. We further assessed the medical necessity of cesarean deliveries in 2016, considering the statement by the World Health Organization (WHO) in 2015 that cesarean delivery should be undertaken when medically necessary.9
METHODS
We conducted a cross-sectional study in 6 hospitals in Gansu and Hubei provinces in China. On the basis of previously reported population-based data from the National Maternal and Child Health Statistics (NMCHS) database,2 we selected Hubei province (cesarean rate in 2013 = 48.3%) to represent an area with a high cesarean rate, and Gansu province (cesarean rate in 2013 = 15.1%) to represent an area with a low cesarean rate. From each province, we selected 2 secondary hospitals and 1 tertiary hospital, respectively. We collected the hospital-level aggregate data, including the number of all deliveries and cesareans in each month between January 2013 and December 2016, from each hospital. We extracted the individual-level data of each hospital from electronic or paper-based medical records in specific months determined a priori, depending on the annual delivery volume and the cooperation degree of the hospitals (Table A, available as a supplement to the online version of this article at http://www.ajph.org).
The individual-level data included maternal delivery mode, cesarean indications, and maternal and offspring characteristics. We categorized delivery mode as spontaneous vaginal delivery, assisted vaginal delivery, and cesarean delivery. For cesarean delivery, we considered the primary indication, including breech or transverse presentation, oligohydramnios, arrested labor, fetal distress, macrosomia, cord entanglement, premature rupture of membranes, multiple pregnancy, placenta previa, preeclampsia, antepartum hypertension, previous cesarean delivery, maternal request, advanced maternal age, and other. Maternal and offspring characteristics included maternal age, parity, gestational weeks, birth weight, and sex.
We included 93 745 deliveries in the hospital-level aggregate data. Of these, 31 275 were in the selected months, with 29 895 having available individual information. From the 29 895 deliveries, we excluded 28 because they lacked information on delivery mode; in addition, because the cesarean indications in 2 hospitals in 2013 were not available, we further excluded these cesarean deliveries (n = 1091) and the vaginal deliveries (n = 799) in these 2 hospitals in 2013. (During the 1-child policy period in the 2 hospitals, the maternal characteristics between the excluded women and the included women who delivered between January and September 2014 were comparable [Table B, available as a supplement to the online version of this article at http://www.ajph.org].) Finally, 27 977 deliveries with individual information remained in the analyses (Figure A, available as a supplement to the online version of this article at http://www.ajph.org).
Assessment of Medical Necessity for Cesarean
We invited 4 chief physicians and professors in gynecology and obstetrics from 4 of the top tertiary hospitals in China as experts to review the complete and detailed medical records of cesareans and assess their necessity (eMethod, available as a supplement to the online version of this article at http://www.ajph.org). According to uniform assessment standards, the experts classified these cesarean deliveries into 3 categories: absolutely necessary cesarean delivery (with obstetric or medical indications and a contraindication for vaginal delivery), relatively necessary cesarean delivery (with obstetric or medical indications and without a contraindication for vaginal delivery), and nonessential cesarean delivery (without an obstetric or medical indication and recommended for vaginal delivery). Two experts reviewed the same medical record. Disagreements were resolved by consensus. Additionally, the experts extracted the primary indications documented in the medical records.
We selected 1000 cesarean deliveries in 2016 from 5 hospitals (200 each) using the stratified random sampling method according to the month of delivery, and we included all cesareans (n = 42) in 2016 in 1 hospital. Of these selected cesareans (n = 1042), 18 medical records were not available. The experts assessed 1024 cesarean deliveries (Table A).
Child Policy Periods
According to the time of the selective and universal 2-child policy announcements and the duration of gestation, we divided the study phase into 3 periods: the 1-child policy period (January 2013–September 2014), the selective 2-child policy period (October 2014–July 2016), and the universal 2-child policy period (August 2016–December 2016). Our categorization was further validated by the increasing trend in the delivery number at the beginning of the selective and universal 2-child policy periods (Figure B, available as a supplement to the online version of this article at http://www.ajph.org).
Statistical Analyses
We used the hospital-level aggregate data to calculate the overall cesarean rates and the individual-level data to calculate the indication-specific cesarean rates and cesarean indication proportions. We defined the overall cesarean rate as the number of total cesareans divided by the number of all deliveries. We defined the indication-specific cesarean rate as the number of cesareans with a certain indication divided by the number of all deliveries. We defined the cesarean indication proportion as the number of cesareans with a certain indication divided by the total number of cesareans. In subgroup analyses, we calculated the overall cesarean rates stratified by province (Hubei and Gansu) and level of delivering hospital (tertiary and secondary hospitals) using hospital-level aggregate data, and we calculated the rates stratified by parity (primiparous and multiparous women) and maternal age (< 35 and ≥ 35 years) using the individual-level data.
We used analysis of variance to compare the means of continuous variables and the χ2 test to compare the proportions of categorical variables. We performed the χ2 test for trend to assess the trends in the cesarean rates across the 3 periods. We further used univariate and multivariable Poisson regression models to explore the associations between child policy periods and cesarean rates. We used univariate models to examine the crude relative risks (RRs) and 95% confidence intervals (CIs), taking into account the child policy periods, with the 1-child policy period as the reference group. We used multivariable models to examine the adjusted RRs and 95% CIs by further taking account of parity, maternal age, and level of delivering hospital. We performed all analyses using SPSS version 20.0 (IBM, Somers, NY). We considered a 2-tailed P value of less than .05 significant.
RESULTS
The overall cesarean rates derived from the hospital- and individual-level data by hospital and year were comparable (P > .05), indicating a reasonable representation of the individual-level data (Table A). Of the 27 977 women with individual data, 2663 (9.6%) were 35 years or older, 10 737 (38.6%) were multiparous, and 2453 (8.8%) delivered before 37 weeks. The proportion of women aged 35 years or older increased from 8.2% in the 1-child policy period to 9.6% in the selective 2-child policy period and further to 12.3% in the universal 2-child policy period (P < .001; Table 1). The proportion of multiparous women increased from 32.7% to 39.7% and further to 46.9% over the 3 periods (P < .001).
TABLE 1—
Maternal and Offspring Characteristics According to Child Policy Period: Hubei and Gansu Provinces, China, 2013–2016
| Characteristic | All (n = 27 977), Mean ±SD or No. (%) | 1-Child Policy Period (n = 8664), Mean ±SD or No. (%) | Selective 2-Child Policy Period (n = 15 178), Mean ±SD or No. (%) | Universal 2-Child Policy Period (n = 4135), Mean ±SD or No. (%) | Pa |
|
Mother |
|||||
| Age, y | 27.8 ±4.8 | 27.4 ±4.8 | 27.9 ±4.7 | 28.8 ±4.6 | < .001 |
| < 35 | 25 210 (90.4) | 7 944 (91.8) | 13 657 (90.4) | 3 609 (87.7) | < .001 |
| ≥ 35 | 2 663 (9.6) | 711 (8.2) | 1 444 (9.6) | 508 (12.3) | |
| Parity | |||||
| Primiparous | 17 080 (61.4) | 5 800 (67.3) | 9 099 (60.3) | 2 181 (53.1) | < .001 |
| Multiparous | 10 737 (38.6) | 2 823 (32.7) | 5 987 (39.7) | 1 927 (46.9) | |
| Gestational wk | 38.6 ±2.3 | 38.5 ±2.4 | 38.6 ±2.3 | 38.6 ±1.9 | .048 |
| < 37 | 2 453 (8.8) | 786 (9.1) | 1 269 (8.4) | 398 (9.7) | .76 |
| ≥ 37 | 25 361 (91.2) | 7 839 (90.9) | 13 803 (91.6) | 3 719 (90.3) | |
| Level of delivering hospital | |||||
| Secondary | 15 615 (55.8) | 5 529 (63.8) | 8 372 (55.2) | 1 714 (41.5) | < .001 |
| Tertiary | 12 362 (44.2) | 3 135 (36.2) | 6 806 (44.8) | 2 421 (58.6) | |
|
Offspring |
|||||
| Birth weight, g | 3246.3 ±523.9 | 3 241.9 ±524.6 | 3 251.2 ±524.1 | 3 237.1 ±521.6 | .99 |
| < 4000 | 25 601 (93.6) | 7 912 (93.5) | 13 887 (93.4) | 3 802 (94.5) | .10 |
| ≥ 4000 | 1 752 (6.4) | 550 (6.5) | 980 (6.6) | 222 (5.5) | |
| Sex | .51 | ||||
| Male | 14 695 (53.4) | 4 583 (53.5) | 8 020 (53.5) | 2 092 (52.7) | |
| Female | 12 839 (46.6) | 3 984 (46.5) | 6 980 (46.5) | 1 875 (47.3) | |
Note. The percentages of cases with missing data on maternal age, parity, gestational wk, birth weight, and sex were 0.4%, 0.6%, 0.6%, 2.2%, and 1.6%, respectively.
P value was for the trend across the 3 periods.
Trends of Cesarean Delivery Rates
Of the 93 745 deliveries included in the hospital-level aggregate data, 39 215 (41.8%) were cesarean deliveries. The overall cesarean rate decreased from 45.1% during the 1-child policy period to 40.4% during the selective 2-child policy period and further to 38.9% during the universal 2-child policy period (P < .001; Table 2). After adjustments for parity, maternal age, and level of delivering hospital, the selective (adjusted RR = 0.91; 95% CI = 0.88, 0.94) and universal 2-child policy (adjusted RR = 0.87; 95% CI = 0.82, 0.91) periods had decreasing rates of cesarean delivery compared with the 1-child policy period. This decreasing trend was consistently observed across subgroups stratified by province and level of delivering hospital as well as among primiparous women and those younger than 35 years (P < .001).
TABLE 2—
Cesarean Rates During Different Child Policy Periods: Hubei and Gansu Provinces, China, 2013–2016
| 1-Child Policy Period |
Selective 2-Child Policy Period |
Universal 2-Child Policy Period |
||||||||
| Rate | Crude RR (95% CI) | Adjusted RR (95% CI) | Rate | Crude RR (95% CI) | Adjusted RR (95% CI) | Rate | Crude RR (95% CI) | Adjusted RR (95% CI) | Pa | |
| All | 45.1 (15 020/33 271) | 1 (Ref) | 1 (Ref) | 40.4 (18 468/45 741) | 0.90 (0.87, 0.94) | 0.91 (0.88, 0.94) | 38.9 (5 727/14 733) | 0.85 (0.81, 0.90) | 0.87 (0.82, 0.91) | < .001 |
| Province | ||||||||||
| Hubei | 60.8 (6 798/11 180) | 1 (Ref) | 1 (Ref) | 55.8 (7 656/13 718) | 0.89 (0.85, 0.93) | 0.88 (0.84, 0.92) | 54.7 (2 023/3 702) | 0.86 (0.80, 0.92) | 0.83 (0.77, 0.89) | < .001 |
| Gansu | 37.2 (8 222/22 091) | 1 (Ref) | 1 (Ref) | 33.8 (10 812/32 023) | 0.92 (0.86, 0.99) | 0.89 (0.83, 0.95) | 33.6 (3 704/11 031) | 0.95 (0.87, 1.04) | 0.88 (0.81, 0.97) | < .001 |
| Level of delivering hospital | ||||||||||
| Tertiary | 41.4 (8 924/21 567) | 1 (Ref) | 1 (Ref) | 37.0 (11 737/31 741) | 0.95 (0.89, 1.01) | 0.93 (0.87, 1.00) | 37.0 (3 905/10 565) | 0.95 (0.87, 1.03) | 0.91 (0.83, 0.99) | < .001 |
| Secondary | 52.1 (6 096/11 704) | 1 (Ref) | 1 (Ref) | 48.1 (6 731/14 000) | 0.92 (0.88, 0.96) | 0.90 (0.86, 0.95) | 43.7 (1 822/4 168) | 0.87 (0.81, 0.94) | 0.84 (0.78, 0.91) | < .001 |
| Parity | ||||||||||
| 1 | 49.2 (2 857/5 800) | 1 (Ref) | 1 (Ref) | 40.7 (3 706/9 099) | 0.83 (0.79, 0.87) | 0.86 (0.82, 0.90) | 35.8 (780/2 181) | 0.71 (0.65, 0.76) | 0.76 (0.70, 0.82) | < .001 |
| ≥ 2 | 55.8 (1 576/2 823) | 1 (Ref) | 1 (Ref) | 55.3 (3 313/5 987) | 0.99 (0.93, 1.05) | 1.00 (0.94, 1.06) | 54.0 (1 041/1 927) | 0.97 (0.90, 1.05) | 1.00 (0.92, 1.08) | .24 |
| Maternal age, y | ||||||||||
| < 35 | 50.4 (4 006/7 944) | 1 (Ref) | 1 (Ref) | 45.2 (6 173/13 657) | 0.89 (0.86, 0.93) | 0.90 (0.87, 0.94) | 42.5 (1 535/3 609) | 0.84 (0.79, 0.89) | 0.87 (0.82, 0.92) | < .001 |
| ≥ 35 | 64.3 (457/711) | 1 (Ref) | 1 (Ref) | 61.6 (890/1 444) | 0.95 (0.85, 1.07) | 0.96 (0.85, 1.07) | 60.0 (305/508) | 0.83 (0.72, 0.96) | 0.84 (0.73, 0.98) | .12 |
Note. CI = confidence interval; RR = relative risk.
P value was for the trend in cesarean rate across the 3 periods.
The cesarean indication proportions and the indication-specific cesarean rates during the 3 periods are shown in Table 3. Throughout the study period, maternal request (33.6%) and previous cesarean delivery (19.7%) were the 2 most common indications. Interestingly, their indication-specific cesarean rates showed opposing trends across the 3 periods. A decreasing trend was observed for the rates of CDMR, which were 18.8%, 16.3%, and 10.1% in the 3 periods, respectively (P < .001). By contrast, we observed an increasing trend for the rate of cesarean delivery indicated by a previous cesarean delivery, which rose from 6.4% to 9.9% and further to 14.0% (P < .001).
TABLE 3—
The Proportion and Rate of Cesarean Indications Across the 3-Child Policy Periods: Hubei and Gansu Provinces, China, 2013–2016
| Rateb |
||||||
| Indication | Total Proportiona (n = 14 511) | Total (n = 27 977), % | 1-Child Policy Period (n = 8664), % (No.) | Selective 2-Child Policy Period (n = 15 178), % (No.) | Universal 2-Child Policy Period (n = 4135), % (No.) | Pc |
| Maternal request | 33.6 (4 515) | 16.1 | 18.8 (1 631) | 16.3 (2 466) | 10.1 (418) | < .001 |
| Previous cesarean delivery | 19.7 (2 640) | 9.4 | 6.4 (557) | 9.9 (1 504) | 14.0 (579) | < .001 |
| Breech or transverse presentation | 4.9 (662) | 2.4 | 2.6 (225) | 2.3 (343) | 2.3 (94) | .15 |
| Cord entanglement | 3.5 (475) | 1.7 | 3.0 (256) | 1.0 (156) | 1.5 (63) | < .001 |
| Oligohydramnios | 3.4 (454) | 1.6 | 2.1 (178) | 1.4 (217) | 1.4 (59) | .001 |
| Premature rupture of membranes | 3.3 (441) | 1.6 | 1.9 (162) | 1.4 (205) | 1.8 (74) | .23 |
| Macrosomia | 3.1 (417) | 1.5 | 1.5 (131) | 1.6 (243) | 1.0 (43) | .12 |
| Antepartum hypertension | 2.5 (329) | 1.2 | 1.3 (109) | 1.1 (166) | 1.3 (54) | .88 |
| Multiple pregnancy | 2.2 (298) | 1.1 | 1.1 (99) | 1.1 (171) | 0.7 (28) | .04 |
| Placenta previa | 2.0 (267) | 1.0 | 1.0 (87) | 0.9 (136) | 1.1 (44) | .98 |
| Fetal distress | 2.0 (263) | 0.9 | 0.9 (80) | 0.7 (106) | 1.9 (77) | < .001 |
| Other | 17.5 (2 345) | 8.4 | 9.7 (842) | 8.1 (1 235) | 6.5 (268) | < .001 |
The cesarean indication proportion was calculated as the number of cesareans with a certain indication divided by the total number of cesareans.
The indication-specific cesarean rate was calculated as the number of cesareans with a certain indication divided by the number of total deliveries.
P value was for the trend in indication-specific cesarean rate across the 3 periods.
Additionally, these 2 rates significantly differed by parity (Figure C [available as a supplement to the online version of this article at http://www.ajph.org]) and level of delivering hospital (Figure D [available as a supplement to the online version of this article at http://www.ajph.org]). The CDMR rate among primiparous women decreased from 20.7% to 16.8% and to 10.5% across the 3 periods, and the corresponding values among multiparous women were 15.1%, 15.7%, and 9.8%, respectively (P < .001). Among multiparous women, the rate of cesarean indicated by a previous cesarean delivery increased from 18.4% to 23.8% and to 29.2% across the 3 periods (P < .001). Throughout the study phase, the CDMR rate in the secondary hospitals (27.8%) was much higher than that in the tertiary hospitals (1.4%; P < .001). In the secondary hospitals, the CDMR rates were 28.1%, 28.5%, and 23.8% during the 3 periods (P = .02), whereas the rates in the tertiary hospitals were 2.5%, 1.2%, and 0.4% (P < .001). The rates of cesarean indicated by a previous cesarean delivery in the secondary hospitals were 6.9%, 10.1%, and 15.7% across the 3 periods (P < .001), and the corresponding values for the tertiary hospitals were 5.6%, 9.7%, and 12.8% (P < .001).
Medical Necessity of Cesarean Delivery
Of the 1024 reviewed cesarean deliveries in 2016, 357 (34.9%) were considered absolutely necessary, 445 (43.5%) were relatively necessary, and 222 (21.6%) were nonessential. The proportion of absolutely necessary cesarean deliveries in the tertiary hospitals compared with the secondary hospitals (48.4% vs 26.2%) was higher, and the proportion of relatively necessary cesarean deliveries was lower (30.3% vs 51.8%; P < .001). Moreover, the proportions of nonessential cesarean deliveries were comparable between the tertiary (21.3%) and secondary (22.0%) hospitals (P > .05).
The proportions of primary indications documented in medical records significantly differed according to medical necessity (P < .001; Table 4). For absolutely necessary cesarean deliveries, the proportion of breech or transverse presentation, fetal distress, placenta previa, and preeclampsia was 45.2%. For relatively necessary cesarean deliveries, the indication of previous cesarean delivery accounted for the largest proportion (73.3%); maternal request accounted for the most substantial proportion (45.2%) for nonessential cesarean deliveries.
TABLE 4—
Proportions of the Top 10 Indications Documented in the Medical Records of the Cesarean Deliveries According to Necessity: Hubei and Gansu Provinces, China, 2013–2016
| Absolutely Necessary |
Relatively Necessary |
Nonessential |
||||
| Order | Indication | No. (%) | Indication | No. (%) | Indication | No. (%) |
| 1 | Breech or transverse presentation | 73 (20.9) | Previous cesarean delivery | 310 (73.3) | Maternal request | 89 (45.2) |
| 2 | Fetal distress | 35 (10.0) | Macrosomia | 23 (5.4) | Premature rupture of membranes | 24 (12.2) |
| 3 | Multiple pregnancy | 33 (9.4) | Oligohydramnios | 19 (4.5) | Fetal distress | 11 (5.6) |
| 4 | Previous cesarean delivery | 32 (9.1) | Preeclampsia | 14 (3.3) | Cephalopelvic disproportion | 11 (5.6) |
| 5 | Placenta previa | 29 (8.3) | Cord entanglement | 7 (1.7) | Oligohydramnios | 10 (5.1) |
| 6 | Arrested labor | 21 (6.0) | Fetal distress | 7 (1.7) | Previous cesarean delivery | 9 (4.6) |
| 7 | Preeclampsia | 21 (6.0) | Premature rupture of membranes | 6 (1.4) | Cord entanglement | 7 (3.6) |
| 8 | Hepatic disease | 12 (3.4) | Hepatic disease | 4 (1.0) | Advanced maternal age | 4 (2.0) |
| 9 | Macrosomia | 12 (3.4) | Advanced maternal age | 4 (1.0) | Antepartum hypertension | 4 (2.0) |
| 10 | Oligohydramnios | 10 (2.9) | Occipitoposterior position | 4 (1.0) | In vitro fertilization | 3 (1.5) |
DISCUSSION
After the implementation of the selective and universal 2-child policies, we observed decreasing trends in the overall cesarean rates in Hubei and Gansu provinces, with the CDMR rates decreasing and the rates of cesarean indicated by a previous cesarean delivery increasing. Regarding the medical necessity of cesarean delivery during the 2-child policy periods, the nonessential cesarean deliveries accounted for less than one quarter (21.6%).
In our study, the overall cesarean rate of Hubei and Gansu provinces first decreased from 45.1% during the 1-child policy period to 40.4% during the selective 2-child policy period, which was comparable with that reported by a previous Chinese study (rate declined from 45.8% to 40.8%).10 We additionally observed that the overall cesarean rate further decreased to 38.9% during the universal 2-child policy period. Maternal (e.g., maternal age, parity) and institutional (e.g., level of delivering hospital) characteristics might influence the cesarean rate.11 After adjustment for these characteristics, the overall cesarean rate persistently declined by 13% from the 1-child policy period to the universal 2-child policy period, which was approximate to that reported by hospital-based data from China’s National Maternal Near Miss Surveillance System (NMNMSS; 18%).12 In parallel, the cesarean rate in the United States had a 2.4% decrease (32.7% to 31.9%) from 2013 to 2016,13,14 which was much less than that in our study, suggesting a remarkable decrease in the cesarean rate in the studied areas.
Because of the major geographic variation in the cesarean rates and trends—possibly attributable to the marked diversity of geography, economy, and life circumstances throughout China, with varied factors such as health system and local patient preferences2—we decided to choose 1 typical province with a high-baseline rate and another with a low-baseline rate to explore their trends of cesarean rates. Although controlling the cesarean rate has been a national priority since 2002,2,15 national data from the NMNMSS showed no decrease in cesarean rate before 2013,12 and data from the NMCHS showed that a decreasing trend was only observed in areas with the highest baseline rates.2 We observed significant decreases in cesarean rates in the province with high-baseline rate and in the province with low-baseline rate after the implementation of the 2-child policy, indicating that besides the strategies of controlling cesarean rates,16–18 the child policy adjustments might have contributed to the decreasing cesarean rates. In addition, we noted the decreasing trends at different levels of delivering hospital, among primiparous women and those younger than 35 years, suggesting universal decreases in cesarean rates.
The decline in the cesarean rate was most pronounced in primiparous women, with a reduction of 24% after multivariable adjustments. The primary contributor to this decreasing trend was the reduction in the CDMR rate, with a nearly 50% decrease across the 3 periods. When people were permitted to have only 1 child, prospective parents tended to request a cesarean delivery because they hoped to have a “perfect baby” and sought to avoid issues related to vaginal delivery (e.g., pain and risk of pelvic floor collapse),4,11 leading to a high CDMR rate in the 1-child policy period.3,19 After implementation of the 2-child policy, more than 50% of primiparous women planned for a second child and were less likely to opt for cesarean delivery after considering the adverse consequences of this procedure in future pregnancies.8,20Additionally, obstetricians were also restricted from providing nonessential cesareans. For example, cesareans performed for nonmedical indications need to be reported to the hospital director for review because this rate is an indicator for hospital quality assessment.21 The decreasing trend in CDMR rate was consistently shown among multiparous women, possibly because of the rate-controlling policies.16–18
The overall cesarean rate among multiparous women remained as high as 54% in the universal 2-child policy period, mainly because of the more than 1-fold increase in the repeat cesarean rate. Although this is not necessarily medically indicated in women at otherwise low obstetrical risk, among deliveries of women with a previous cesarean delivery in China, more than 97% were repeat cesarean delivery.22 Previous studies showed that the trial of labor after cesarean delivery might be a preferred option under optimal circumstances,23,24 whereas the trial of labor after cesarean delivery is still uncommon in China because of concerns such as uterine rupture and uneasy physician–patient relationships.25
In 2015, the WHO stated that it is necessary to provide cesarean delivery to women in need.9 Whether a woman is in need ought to be ascertained on a case-by-case basis.26 Using indications to determine cesarean delivery necessity has always been problematic because of the lack of uniform definitions for many indications27 and the complexity of obstetric cases. Therefore, we invited experts to assess the medical necessity of cesarean deliveries 1 by 1. In our study, more than three quarters of cesareans were assessed as absolutely or relatively necessary. The absolutely necessary cesareans were commonly indicated by severe maternal and neonatal complications, such as breech or transverse presentation, placenta previa, and preeclampsia. The proportion of absolutely necessary cesareans was higher in the tertiary hospitals. Many complicated deliveries were transferred to the tertiary hospitals because of their more comprehensive and specialized medical care.28 Relatively necessary cesareans appeared to be indicated by less severe complications, with more than 70% indicated by previous cesarean delivery, a well-known relative indication.29 In China, whether a repeat cesarean is necessary depends on not only obstetric factors such as uterine thickness but also medical factors such as a hospital’s capability in dealing with complicated deliveries.
One fifth of cesareans were assessed as nonessential cesarean deliveries, with nearly half of these cesareans performed because of maternal request in the absence of medical indications, and the other half performed for indications probably with subjective minimum criteria. Although the proportions of nonessential cesareans were comparable in the secondary and tertiary hospitals (22.0% vs 21.3%) during the 2-child policy period, the CDMR rate in the secondary hospitals (> 20.0%) was much higher than that in the tertiary hospitals (< 1.5%). One reason for this disparity might be the stricter rules for performing CDMR in the tertiary hospitals. However, we could not exclude the possibility that the CDMR rate in the tertiary hospitals might be underestimated if a maternal request was grouped with other indications, considering that tertiary versus secondary hospitals were more concerned about the failure to reach a target of reducing nonmedical cesarean rates.16,21 In addition, the high CDMR rate in the secondary hospitals suggested that their obstetricians were more likely to comply with maternal requests, possibly because of financial incentives and medico-legal concerns.4
Strengths and Limitations
One of the major strengths of our study was that we comprehensively explored the trends in cesarean rates over the periods of child policy adjustments in both a province with high and one with low baseline cesarean rates. Another strength was that we further investigated the distribution of medical necessity of cesarean deliveries during the 2-child policy periods on the basis of case-by-case review by authoritative experts in gynecology and obstetrics.
Some major limitations of our study should be considered. The cesarean rate estimates in our study might not be nationally representative, considering that survey estimates are likely to be sensitive to the sampling areas.2 Nevertheless, our trends in cesarean rates corresponded very closely to the national data from the NMNMSS,12 providing assurance of validity to our survey estimates. We did not extract individual information of deliveries in all months because of staff and resource constraints. However, the individual-level data subset is likely representative of the hospital-level data because of the comparable overall cesarean rates.
Conclusions
After the implementation of the 2-child policy, the overall cesarean rate in Hubei and Gansu provinces consistently decreased, from 45.1% during the 1-child policy period to 40.4% during the selective 2-child policy period and further to 38.9% during the universal 2-child policy period. The CDMR rate declined markedly because low-risk primiparous women tended to choose vaginal delivery over cesarean delivery when they planned to have a second child. By contrast, the rate of cesarean indicated by a previous cesarean delivery rose substantially because the high primary cesarean rate during the 1-child policy period translated into a high repeat cesarean rate during the 2-child policy periods. There is a need to address how obstetric services cope with the stress from an increasing number of multiparous women, especially those with a previous cesarean delivery. In addition, strategies for controlling nonessential cesareans need to be continually reinforced, especially in lower level hospitals.
ACKNOWLEDGMENTS
This study was supported by the National Key Research & Development Program of China (grant 2016YFC1000401), the National Natural Science Foundation of China (grant 81571517), and a Joint Program of the Department of Maternal and Child Health, National Health Commission of China and the United Nations Children’s Fund, China Office.
We express our appreciation to the experts in gynecology and obstetrics for their help in assessing the medical necessity of cesarean deliveries: Xiaotian Li, Obstetrics and Gynecology Hospital of Fudan University, Shanghai; Hongbo Qi, First Affiliated Hospital of Chongqing Medical University, Chongqing; and Xietong Wang, Provincial Hospital Affiliated With Shandong University, Jinan. We thank Keyi Si and Lijun Zhao, Peking University, and staff from the 6 hospitals for their contribution to data collection.
Note. The views expressed here do not necessarily reflect the views of the funders. The funders had no role in the design or conduct of the study; the collection, management, analysis, or interpretation of the data; the preparation, review, or approval of the article; or the decision to submit the article for publication.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.
HUMAN PARTICIPANT PROTECTION
The register-based data did not contain any personal information, and the analysis was exempted from review by the Peking University Health Science Center institutional review board.
REFERENCES
- 1.Feng XL, Xu L, Guo Y, Ronsmans C. Factors influencing rising caesarean section rates in China between 1988 and 2008. Bull World Health Organ. 2012;90(1):30–39, 39A. doi: 10.2471/BLT.11.090399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Li HT, Luo S, Trasande L et al. Geographic variations and temporal trends in cesarean delivery rates in China, 2008–2014. JAMA. 2017;317(1):69–76. doi: 10.1001/jama.2016.18663. [DOI] [PubMed] [Google Scholar]
- 3.Zhu YB, Li HT, Zhang YL, Li ZW, Zhang L, Liu JM. Secular trends of cesarean delivery and cesarean delivery on maternal request among primiparous women with singleton pregnancy in Southern and Northern China during 1993–2010. Zhonghua Yi Xue Za Zhi. 2012;92(25):1734–1737. [PubMed] [Google Scholar]
- 4.Hellerstein S, Feldman S, Duan T. China’s 50% caesarean delivery rate: is it too high? BJOG. 2015;122(2):160–164. doi: 10.1111/1471-0528.12971. [DOI] [PubMed] [Google Scholar]
- 5.National Health and Family Planning Commission of the PRC. Official: two-child policy will be implemented after revised laws come into force. 2015. Available at: http://www.chinadaily.com.cn/m/chinahealth/2015-11/05/content_22381024.htm. Accessed April 26, 2018.
- 6.Zeng Y, Hesketh T. The effects of China’s universal two-child policy. Lancet. 2016;388(10054):1930–1938. doi: 10.1016/S0140-6736(16)31405-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Li PC, Liao ZJ, Wang C et al. Maternal characteristics and birth outcomes under different family planning policy. Zhongguo Sheng Yu Jian Kang Za Zhi. 2017;28(6):36–40. [Google Scholar]
- 8.Zhao J, Shan N, Yang X et al. Effect of second child intent on delivery mode after Chinese two child policy implementation: a cross sectional and prospective observational study of nulliparous women in Chongqing. BMJ Open. 2017;7(12):e018823. doi: 10.1136/bmjopen-2017-018823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.World Health Organization Human Reproduction Programme. WHO statement on caesarean section rates. Reprod Health Matters. 2015;23(45):149–150. doi: 10.1016/j.rhm.2015.07.007. [DOI] [PubMed] [Google Scholar]
- 10.Kang CY, Gao YQ, Song L, Pang RY, Wang Y. Delivery modes before and after the new fertility policy: a comparative study. Zhongguo Fuyou Baojian. 2016;31(2):362–365. [Google Scholar]
- 11.Liu Y, Li G, Chen Y et al. A descriptive analysis of the indications for caesarean section in mainland China. BMC Pregnancy Childbirth. 2014;14:410. doi: 10.1186/s12884-014-0410-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liang J, Mu Y, Li X et al. Relaxation of the 1 child policy and trends in caesarean section rates and birth outcomes in China between 2012 and 2016: observational study of nearly seven million health facility births. BMJ. 2018;360:k817. doi: 10.1136/bmj.k817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Martin JA, Hamilton BE, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for 2013. Natl Vital Stat Rep. 2015;64(1):1–65. [PubMed] [Google Scholar]
- 14.Martin JA, Hamilton BE, Osterman MJK. Births in the United States, 2016. NCHS Data Brief. 2017;(287):1–8. [PubMed] [Google Scholar]
- 15.National Health and Family Planning Commission of the PRC. General guidelines for the implementation of the national 10-year programs for the development of women and children (2001–2010). Available at: http://www.moh.gov.cn/zhuzhan/zcjd/201304. Accessed April 26, 2018.
- 16.National Health and Family Planning Commission of the PRC. The answer to the proposal of the third session of the twelfth National Committee of the CPPCC National Committee. 2016. Available at: http://www.nhfpc.gov.cn/zwgkzt/taxx/201601/182a8f2ff4b14056b6ff3446e0ab891a.shtml. Accessed April 26, 2018.
- 17.National Health and Family Planning Commission of the PRC. Notice of the ministry of health on printing and distributing the implementation plan of the program for the development of Chinese women and children from 2011 to 2020. 2012. Available at: http://www.moh.gov.cn/fys/s7900/201202/cffcb29c192d4e529c6290c280507df9.shtml. Accessed April 26, 2018.
- 18.Beijing Municipal Commission of Health and Family Planning. Beijing launched the program for advancing vaginal delivery and promoting maternal and infant health. 2014. Available at: http://www.bjchfp.gov.cn/xwzx/wnxw/201411/t20141106_102063.htm. Accessed April 26, 2018.
- 19.Blustein J, Liu J. Time to consider the risks of caesarean delivery for long term child health. BMJ. 2015;350:h2410. doi: 10.1136/bmj.h2410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ecker J. Elective cesarean delivery on maternal request. JAMA. 2013;309(18):1930–1936. doi: 10.1001/jama.2013.3982. [DOI] [PubMed] [Google Scholar]
- 21.National Health and Family Planning Commission of the PRC. Notice on reassessment of baby-friendly hospitals. 2014. Available at: http://www.moh.gov.cn/fys/s3585/201406/556c0b7673e8470f9641c28d119a9f31.shtml. Accessed April 26, 2018.
- 22.Cui HS, Li HT, Zhu LP, Li ZW, Zhou YB, Liu JM. Secular trends in cesarean delivery and cesarean delivery on maternal request among multiparous women who delivered a full-term singleton in Southern China during 1993–2005. Beijing Da Xue Xue Bao. 2013;45(3):422–426. [PubMed] [Google Scholar]
- 23.Ma RM, Duan T, Lao TT. VBAC should be encouraged as a means to reduce the caesarean section rate in China: FOR: VBAC reduces not only the caesarean section rate but also other associated issues. BJOG. 2016;123(suppl 3):10. doi: 10.1111/1471-0528.14291. [DOI] [PubMed] [Google Scholar]
- 24.Black M, Bhattacharya S, Philip S, Norman JE, McLernon DJ. Planned repeat cesarean section at term and adverse childhood health outcomes: a record-linkage study. PLoS Med. 2016;13(3):e1001973. doi: 10.1371/journal.pmed.1001973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Silver RM. VBAC should be encouraged as a means to reduce the caesarean section rate in China: AGAINST: This strategy won’t work! BJOG. 2016;123(suppl 3):11. doi: 10.1111/1471-0528.14292. [DOI] [PubMed] [Google Scholar]
- 26.Betran AP, Torloni MR, Zhang JJ, Gülmezoglu AM. WHO statement on caesarean section rates. BJOG. 2016;123(5):667–670. doi: 10.1111/1471-0528.13526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Torloni MR, Betran AP, Souza JP et al. Classifications for cesarean section: a systematic review. PLoS One. 2011;6(1):e14566. doi: 10.1371/journal.pone.0014566. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Li X, Huang J, Zhang H. An analysis of hospital preparedness capacity for public health emergency in four regions of China: Beijing, Shandong, Guangxi, and Hainan. BMC Public Health. 2008;8:319. doi: 10.1186/1471-2458-8-319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mylonas I, Friese K. Indications for and risks of elective cesarean section. Dtsch Arztebl Int. 2015;112(29–30):489–495. doi: 10.3238/arztebl.2015.0489. [DOI] [PMC free article] [PubMed] [Google Scholar]