Abstract
Abstract
Objectives
Since 1985, the international healthcare community has recommended the ideal rate of caesarean section (CS) to be 10%–15% at the national level. The literature has reported that overused CS without necessary medical indications can be harmful to both maternal and child health. To generate evidence to support policy on CS, this study evaluated the trend over time of CS in Thailand during January 2016 to October 2021 (which included the COVID-19 pandemic period) and explored predictors of CS use.
Design and setting
This study was a retrospective secondary data analysis of de-identified hospitalisation data under the Universal Coverage Scheme (UCS) from the National Health Security Office’s e-Claims database. Descriptive analyses were conducted to explore the number and rate of CS over time and across different characteristics (ie, age, hospital type, COVID-19 status and delivery day) including a multivariable logistic analysis to explore predictors of CS. Interrupted time series analysis was adopted to investigate the effect of the COVID-19 pandemic on CS rate.
Participants
569 321 CS cases under UCS from 2016 to 2021.
Results
The results showed an increasing trend of CS rate, from 30% in January 2016 to 35% in October 2021. Both clinical (eg, medical indication and age) and non-clinical (eg, region and day of delivery) factors were significantly associated with CS. Furthermore, the COVID-19 pandemic had no significant effect on CS rate (level: −0.0016, 95% CI −0.0085 to 0.0053, p=0.66).
Conclusion
This study highlighted an increasing trend of CS in Thailand and could present supportive evidence that Thailand might have been facing an overuse of CS. More awareness and actions are warranted to ensure the movement towards reduction of unnecessary CS in Thailand.
Keywords: Cesarean Section, Retrospective Studies, Health Services
STRENGTHS AND LIMITATIONS OF THIS STUDY.
This study used recent national-level administrative data to evaluate the trend of caesarean section rate over time in Thailand during January 2016 to October 2021 (which included the COVID-19 pandemic period) and explored predictors of CS use, which can provide insights to support the movement to reduce unnecessary CS procedures.
The administrative database was designed for reimbursement purposes, and therefore, other potential predictors (eg, clinical factors and time of day of delivery) of CS use may not be included and could be areas for future research.
Only data from hospitals under the Universal Coverage Scheme were available, which may limit the generalisability of the findings.
Introduction
Since 1985, the international healthcare community has recommended the ideal rate of caesarean section (CS) to be between 10% and 15% at the population level.1 It was claimed that for around 10% of all births where it is clinically indicated, CS is an effective life-saving intervention for mothers and newborns. A higher rate of CS could be a signal to further explore for any potentially unnecessary procedures. Additionally, in women with CS, there is an increased risk of infections, haemorrhage, severe acute maternal morbidity and an increased risk of respiratory complications such as transient tachypnoea and respiratory distress and altered immune system in newborns, as well as long-term negative outcomes such as increased risk of placental accreta.2 3 Thus, CS can be considered as low-value care when performed without necessary medical indication since the potential harms outweigh possible benefits. Low-value care has recently been highlighted again as a public health concern in lower-income and middle-income countries (LMICs).4 A scoping review by Albarqouni et al pointed out that significant harm to health, wasted resources and financial burden were the major consequences of low-value care.4 As such, the WHO affirms that CS should be performed only when medically needed.1
The high CS rate is an ongoing public health concern, and many studies have been conducted to investigate the magnitude of CS. The study by Betrán et al showed that the global CS rate was about 20% in 2010 and was expected to reach 30% by 2030. It also found that there was, in general, more overuse of CS in high-income countries (HICs) than in LMICs, although there were exceptions where LMICs had higher CS rates than some HICs as well.5 In 2017, Brazil held the highest CS rate in the world at approximately 55%.5 The CS rates in HICs such as the USA, Canada, the UK and Australia were 32%, 29%, 28% and 35%, respectively.5 Moreover, Thailand had a relatively high CS rate of about 32% whereas the CS rate for its neighbouring countries ranged between 5% in Laos to 18% in Myanmar5 and Indonesia.6 Other LMICs have also published their CS information, which can be as low as 2.9% in Sierra Leone.7 Also, many initiatives have been created around the globe to reduce unnecessary CS8 apart from its ongoing upward trend.
In Thailand, a few studies have examined the CS trend at both the hospital level and the national level. At the hospital level, the CS rate ranged from 48% to 56%—a tertiary hospital in Bangkok at 49% in 2017,9 a tertiary hospital in southern Thailand at 55.5% from 2014 to 2016,10 a tertiary hospital in central Thailand at 48% in 2016,11 and another tertiary hospital in southern Thailand at 50% during 2011 to 2018.12 At the national level, a survey by the National Statistical Office in 2019 indicated that the CS rate among mothers aged between 15 years and 49 years old in Thailand was around 35%.13 Another study by Liabsuetrakul et al showed that Thailand’s CS rate significantly increased from 23% in 2009 to 33% in 2017 and was projected to reach 59% in 2030 if the same rate of escalation continues.14
One of WHO’s recommendations is ongoing monitoring of the CS rate to reduce medically unnecessary use of CS for the sake of mother and child health.15 Therefore, this study examined CS rate in Thailand between 2016 and 2021 and explored potential predictors associated with higher likelihood of CS. Additionally, the period of this study covers the COVID-19 pandemic (2020) in which healthcare facilities in Thailand were overwhelmed and all elective surgeries were supposed to be halted (including unnecessary CS). As such, this study also assesses the impact of the pandemic response to the CS rate in Thailand. This information may further shed light on whether the trend of CS continues in Thailand and some CS cases may be considered as a possible low-value care.
Methods
Study population and setting
This study covered all mothers who gave birth in the inpatient department (IPD) under the Universal Coverage Scheme (UCS) in Thailand from January 2016 to October 2021. UCS is one of the three major public health insurance schemes in Thailand that covers approximately 72% of Thailand’s population who are not covered by the other two schemes (Social Security Scheme (SSS) for private sector employees and Civil Servant Medical Benefits Scheme (CSMBS) for government officers).16 The analyses employed de-identified individual-level data from the IPD e-Claim database of UCS provided by the National Health Security Office, the public healthcare payer.
Variables
Dependent variable
All women who underwent CS were identified using the International Classification of Diseases, 10th revision (ICD-10) code O82 (single delivery) or O84.2 (multiple delivery) by CS17 and the ICD, 9th revision, clinical modification code 74—classical CS (74.0), low cervical CS (74.1), extraperitoneal CS (74.2), CS of other specified type (74.4) and other CS of unspecified type (74.99).18 The number of mothers giving birth in hospitals was obtained from ICD-10 codes O80–O84. The CS rate was calculated as follows:
Explanatory variables
The study explored variables including demographic (maternal age), health status (COVID-19 infection status from one of these ICD-10 codes: B34.2, B97.2, U07.1, U07.2) and other characteristics of the delivery (namely, hospital type, region and day of the week that a CS was performed). Age was categorised into three groups: less than 20 years old; 20–35 years old; and more than 35 years old. Hospital types include centre hospital (>500 beds and capable of tertiary care); general hospital (200–500 beds and capable of secondary care); community hospital (10–150 beds and capable of primary care); private hospital and clinic under UCS; and hospital outside the oversight of the Ministry of Public Health (MOPH).19 Region was classified into four groups: central, northern, north-eastern and southern (table 1). The list of primary indications and secondary indications for CS is provided in online supplemental appendix 1 and appendix 2. These lists of indications are results from consultations with nationally recognised clinical experts with international experience and accreditation from Royal Colleges as a clinical indication required for CS. For days of the week, we considered weekdays to be Monday to Friday, and the weekend to be Saturday and Sunday. The COVID-19 pandemic in Thailand occurred from 2020 with the lockdown period between March and May 2020. All these variables could support a better understanding of factors contributing to CS in Thailand.
Table 1. Number of CS cases by each characteristic group.
| Group | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 |
|---|---|---|---|---|---|---|
| Age group | ||||||
| <20 years old | 16 726 (16.4%) |
15 951 (15.1%) |
13 532 (13.5%) |
11 841 (12.5%) |
10 971 (11.8%) |
8200 (11.0%) |
| 20–35 years old | 72 272 (70.7%) |
75 867 (71.7%) |
72 612 (72.8%) |
69 297 (73.3%) |
68 224 (73.6%) |
55 246 (74.4%) |
| >35 years old | 13 222 (12.9%) |
13 933 (13.2%) |
13 624 (13.7%) |
13 376 (14.2%) |
13 568 (14.6%) |
10 859 (14.6%) |
| Total | 102 220 (100%) |
105 751 (100%) |
99 768 (100%) |
94 514 (100%) |
92 763 (100%) |
74 305 (100%) |
| Hospital type | ||||||
| Centre hospital | 34 810 (34.1%) |
35 726 (33.8%) |
33 476 (33.4%) |
31 731 (33.6%) |
30 406 (32.8%) |
22 755 (30.5%) |
| General hospital | 44 040 (43.1%) |
45 834 (43.3%) |
43 571 (43.7%) |
41 171 (43.6%) |
39 973 (43.1%) |
32 158 (43.3%) |
| Private hospital and clinic | 15 130 (14.8%) |
16 319 (15.4%) |
16 010 (16.1%) |
15 790 (16.7%) |
16 772 (18.1%) |
14 842 (20.0%) |
| Community hospital | 3996 (3.8%) |
3478 (3.3%) |
2558 (2.6%) |
2068 (2.1%) |
1779 (1.9%) |
1086 (1.5%) |
| Outside MOPH supervision | 4244 (4.2%) |
4394 (4.2%) |
4153 (4.2%) |
3754 (4.0%) |
3833 (4.1%) |
3464 (4.7%) |
| Total | 102 220 (100%) |
105 751 (100%) |
99 768 (100%) |
94 514 (100%) |
92 763 (100%) |
74 305 (100%) |
| Region | ||||||
| Central region | 28 209 (27.5%) |
28 518 (27.0%) |
26 526 (26.6%) |
24 989 (26.4%) |
24 241 (26.1%) |
18 939 (25.5%) |
| Northern region | 15 795 (15.5%) |
16 125 (15.3%) |
15 287 (15.3%) |
14 682 (15.6%) |
14 026 (15.1%) |
11 225 (15.0%) |
| North-eastern region | 34 439 (33.7%) |
36 152 (34.2%) |
34 505 (34.6%) |
31 941 (33.8%) |
31 605 (34.1%) |
26 194 (35.3%) |
| Southern region | 23 777 (23.3%) |
24 956 (23.5%) |
23 450 (23.5%) |
22 902 (24.2%) |
22 891 (24.7%) |
17 947 (24.2%) |
| Total | 102 220 (100%) |
105 751 (100%) |
99 768 (100%) |
94 514 (100%) |
92 763 (100%) |
74 305 (100%) |
| COVID-19 status | ||||||
| Positive | – | – | – | – | – | 1083 (1.46%) |
| Negative | – | – | – | – | – | 73 222 (98.54%) |
| Total | 74 305 (100%) |
CS, caesarean section; MOPH, Ministry of Public Health.
Data analysis
Descriptive analyses were conducted where the trend over time of the numbers and rates of CS across different characteristics (ie, age, hospital type, COVID-19 status and day of the week) were investigated. Interrupted time series analysis was adopted to investigate the impact of the COVID-19 pandemic on the trend and level of CS rate as the lockdown policy did not prohibit emergency care including birthgiving procedures. Furthermore, multivariable logistic analysis was used to find the potential predictors of CS in Thailand. Additionally, another regression model focusing mainly only on those with CS was conducted where the outcome was either a CS with clinical indications or CS without any clinical indications. All statistical analyses were performed by using Stata statistical software (StataCorp. 2021. Stata Statistical Software: Release 17. College Station, Texas, USA)20
Patient and public involvement
None.
Results
Trends over time
Thailand showed a decreasing trend in the total number of births from 2016 to 2021 as well as the number of women who underwent vaginal delivery (VD) while the number of women who underwent CS seemed stable. In addition, there was an obvious seasonality trend in birth—highest in September and lowest in February (figure 1A). When converted to a rate, we found that Thailand had an increasing trend of CS rate—from 30% in January 2016 to 35% in October 2021 (figure 1B).
Figure 1. Overall trend of CS from 2016 to 2021 1A: total number of newborns from 2016 to 2021, 1B: CS rate from 2016 to 2021, 1C: CS rate by age group from 2016 to 2021 and 1D: CS rate by hospital type from 2016 to 2021. CS, caesarean section; VD, vaginal delivery.
The results showed that women under 20 years of age (green line) had the lowest CS rate around 20% to 25% between 2016 and 2021, whereas women older than 35 years held the highest rate around 40% to 50%. Although each age group exhibited a different level of CS rate, all groups portrayed a similar increasing trend of CS (figure 1C).
Figure 1D illustrates that all types of hospitals showed a rising trend in CS rate, especially in community hospitals that used to keep the CS rate below the ideal rate (from 10% in January 2016 to 20% in October 2021). Centre hospitals and general hospitals were the front-runners in performing CS before the COVID-19 period; the CS rates in both hospital types were moving from 40% to 50% in the past 6 years, followed by the hospital outside the oversight of MOPH, which held CS rates around 35%–40%. However, the trend in 2021 has now changed, and private hospitals and clinics throughout the country under UCS became new leaders after the lockdown period with the CS rate at a remarkable 57% in the second quarter.
Mothers also tend to give birth to their children on weekdays rather than on weekends. Figure 2 shows the comparison of delivery patterns by days of the week between modes of delivery (ie, VD and CS) from 2019 to 2021. In both VD and CS, most of the labour occurred during weekdays and less on Fridays and weekends. However, the difference between weekdays and weekends was higher for the CS procedure (a significant drop in CS performed at around 7% from Thursday to Friday and the numbers plunged even more on Saturday). These patterns were consistent throughout 2019–2021.
Figure 2. Delivery rate by day of the week.
Impact of COVID-19 on CS rate
The interrupted time series analysis indicated that there was no significant change in the level and trend of CS rate and VD rate due to the COVID-19 pandemic (figure 3) (online supplemental appendix 3). When using the lockdown period (week 13 of 2020) as an intervention, there was a slightly insignificant drop on the level and rise on the trend of CS rate after the intervention. On the other hand, the level of VD rate increased, while the trend dropped insignificantly. Moreover, we examined the CS rate among COVID-19-positive mothers and found that 51% of mothers in the COVID-19-positive group underwent CS whereas only 33% in the COVID-19-negative group underwent CS during the same period.
Figure 3. Interrupted time series analysis of CS and VD rates before and during COVID-19. CS, caesarean section; VD, vaginal delivery.
Predictor of CS
A multivariable logistic regression model was used to examine the potential factors associated with CS (table 2). The findings suggest that medical indication, age, hospital type and region were significantly associated with the likelihood of undergoing CS in mothers. All of the estimated ORs from the model were statistically significant at the 95% confidence level, although the relative contribution of each variable to the model varied.
Table 2. Multivariable regression model for factors associated with caesarean section.
| Caesarean section | OR | P value > |z| | 95% CI |
|---|---|---|---|
|
Indication group (vs No indications) |
|||
| Secondary indications | 10.3651 | 0.000 | 9.7368 to 11.0339 |
| Primary indications | 580.0511 | 0.000 | 552.3362 to 609.1567 |
|
Hospital type (vs Centre hospital) |
|||
| General hospital | 1.1538 | 0.000 | 1.1062 to 1.2034 |
| Community hospital | 0.2186 | 0.000 | 0.2075 to 0.2303 |
| Private hospital | 1.2803 | 0.000 | 1.1324 to 1.4475 |
| Outside MOPH supervision | 0.5987 | 0.000 | 0.5400 to 0.6637 |
|
Indication group * Hospital type (vs No indications with Centre hospital) |
|||
| Secondary indications * General hospital | 1.6999 | 0.000 | 1.5684 to 1.8424 |
| Secondary indications * Community hospital | 2.4132 | 0.000 | 2.1860 to 2.6640 |
| Secondary indications * Private hospital | 2.9562 | 0.000 | 2.1436 to 4.0768 |
| Secondary indications * Outside MOPH hospital | 1.1478 | 0.133 | 0.9587 to 1.3743 |
| Primary indications * General hospital | 1.3322 | 0.000 | 1.2453 to 1.4252 |
| Primary indications * Community hospital | 3.3057 | 0.000 | 3.0621 to 3.5687 |
| Primary indications * Private hospital | 1.7363 | 0.000 | 1.3618 to 2.2137 |
| Primary indications * Outside MOPH hospital | 0.8967 | 0.000 | 0.7887 to 1.0196 |
|
Age group (vs <20 years old) |
|||
| 20–35 years old | 1.7046 | 0.000 | 1.6458 to 1.7656 |
| >35 years old | 1.8536 | 0.000 | 1.7663 to 1.9453 |
|
Region (vs Central region) |
|||
| Northern region | 1.3112 | 0.000 | 1.2569 to 1.3678 |
| North-eastern region | 1.1353 | 0.000 | 1.0973 to 1.1746 |
| Southern region | 1.0506 | 0.000 | 1.0131 to 1.0896 |
interaction term
MOPH, Ministry of Public Health.
From the model, mothers who underwent CS were likely to be: (1) Medically indicated (OR=580.05, 95% CI 552.34 to 609.16); (2) Admitted to a private hospital compared with the centre hospital (OR=1.28, 95% CI 1.1 to 1.4); (3) Older (OR=1.85, 95% CI 1.77 to 1.95); and (4) Admitted to the hospital in the northern region compared with central region (OR=1.31, 95% CI 1.26 to 1.37). Furthermore, the interaction term in the model showed that mothers who had primary indications and were admitted to the community hospital were more likely to undergo CS (OR=3.30, 95% CI 3.06 to 3.57) compared with those without indications and admitted to the centre hospital. This model has been diagnosed and passed the goodness-of-fit tests, specifically the link test and the Hosmer and Lemeshow’s goodness-of-fit test to detect the specification error.21
In another regression model focusing only on those with CS (online supplemental appendix 4), the findings showed that private hospitals were more likely to perform CS without clinical indications compared with central hospitals (OR=1.3959, 95% CI 1.23 to 1.58). Furthermore, younger women were more likely to have CS without clinical indications compared with older women (OR=0.8240, 95% CI 0.78 to 0.86). The north-east region was less likely to perform CS without clinical indications compared with the central region (OR=0.83, 95% CI 0.78 to 0.87).
Discussion
Our trend over time provided evidence to support the contention that Thailand is facing an increasing trend of CS and has evidently surpassed the rate that the international healthcare community recommends, and further highlighted by WHO.1 The findings highlighted that this topic is important for maternal and child healthcare in Thailand as overuse of CS without necessary medical indication can be harmful to both mother and child health.2 3
The CS rate was positively associated with maternal age.22,25 The results were consistent with the two underlying causes—older mothers had more likelihood of having a comorbidity (eg, hypertension, diabetes) which was risky for VD,24 and a higher chance that she had already undergone CS in her previous delivery.25 However, high maternal age was not the only factor to determine the appropriateness of CS, as some literature still recommends that VD should be encouraged in women above 35 years old.26 27 Furthermore, we observed that CS occurred more in centre hospitals, general hospitals and private hospitals (overall CS rates around 35%–55%). This tendency might be the result of a big hospital having ample resources to perform CS (whether it is necessary or not) and possibly being in charge of more complicated pregnancy cases. However, private hospitals in this study only captured those under UCS, which might not be a good representation of all private hospitals in Thailand. The analysis of all private hospitals may show different findings with a possibility of an even higher CS rate,28 an area for future research to explore. Another intriguing finding was the significant increasing rate of CS in community hospitals, where the rate doubled in the previous 6 years (from 10% in January 2016 to 20% in October 2021), possibly as a result of the escalation of the capacity of medical services over the time. Nevertheless, this trend shows that even in the smallest size of the hospital where resources may be limited, the CS rate is growing.
The analysis of CS and VD by day of the week reported some special patterns. In VD, the proportion of babies born throughout the week from Monday to Sunday was nearly the same, which makes sense since we cannot predict which day the amniotic sac will rupture. Conversely, CS was performed mostly during weekdays and significantly dropped on Fridays and weekends. This pattern raises the doubt of whether there were any underlying non-medical factors that CS was scheduled and performed on a particular weekday (eg, family beliefs and convenience, highlighting the potential demand for CS which is based on non-medical reasons).
Another reason for the overuse of CS in Thailand may be attributed to the study population, which consists solely of participants under UCS—the scheme with the fewest covered services among all three health schemes. The UCS adopted the retrospective payment method, in that the reimbursement amount was based on case mix (diagnosis-related groups and the relative weight), whereas the other two schemes were based on fee for the service system.29 Therefore, CS incurred a higher reimbursement amount than VD since it uses more resources. This rate difference highlights that there could be other potential factors contributing to overutilisation of CS apart from financial incentive (higher reimbursement rate for CS), better time management by obstetricians (structured scheduling rather than having to be prepared 24/7), fear of medical lawsuit, personal beliefs, all of which might be worthwhile to explore further.30,33 Specifically, regarding the reimbursement rate, an additional analysis showed that if the reimbursement for CS and VD were standardised, hospitals with a CS rate below 40% would experience financial gains, as VD incurs lower costs than CS.34 Although this analysis is currently hypothetical, we are actively engaging with relevant stakeholders to implement this change in reimbursement rate. This initiative exemplifies a value-based payment approach in obstetric care, particularly for LMICs dealing with the challenge of increasing CS rates.
In addition, we investigated the impact of COVID-19 on CS rate using interrupted time series analysis and found that the lockdown policy due to COVID-19 did not affect the CS rate, despite the prohibition on unnecessary surgery during this period. We also discovered that half of the mothers with COVID-19 received CS, which was higher than the group that was not infected with COVID-19. This finding was expected since management of resources and virus contamination was more effective in CS, especially during the time of an outbreak. However, the finding did not align with the clinical practice guidelines of the management of COVID-19 infection in pregnancy3 which indicates that CS should only be performed in mothers with COVID-19 who experience severe illness. Otherwise, the choice of VD or CS should be based on medical indication.
In addition, the multivariable logistic regression showed that clinical indications were the most heavily influential factors to determine whether CS would be performed in that mother or not, and which aligned with WHO’s recommendations.1 We found that mothers who were admitted in the northern region were more likely to undergo CS, which is unexpected. Our initial hypothesis was that the hospital in the central region might comprise the highest CS rates relative to other regions, as this is the most developed area of the country. However, this finding could be due to the fact that more hospitals and patients in the central region are covered by other health insurance schemes (ie, SSS, CSMBS). As such, the exploration of another health service scheme might be worthwhile to get a comprehensive picture of the CS rate in Thailand.
Lastly, the regression model focusing only on those with CS supported the ideas that CS without clinical indications might be more common in private hospitals, among younger women and in the central part of the country (urban region). Given that all CS are financed through public resources, these findings can help inform more targeted and equitable interventions, in a way strengthening the case for future policy actions aimed at reducing inequities in resource allocation.
Strengths and limitations
To the best of our knowledge, this study is the most recent study to examine the CS trend in Thailand which also covered the COVID-19 pandemic period. This study explored the CS trend across various maternal characteristics (ie, age, hospital type, COVID-19 status and day of the week), where the findings may be informative to other similarly situated countries. The results highlight a concerningly high CS rate in Thailand.
However, there are some limitations to this study. Data availability was the major constraint since only e-Claim data from hospitals under UCS were available for this study. Consequently, certain information which may influence the CS rate was not included, such as patients’ special arrangements for the delivery,35 time of the day of delivery and maternal preference. Also, the number of private hospitals under UCS might not represent all private hospitals. Future research should explore databases from SSS, CSMBS and the private sector to provide a more comprehensive view of the current situation in Thailand. Furthermore, all data were from the existing real-world administrative e-Claim database where data were meant to be used for reimbursement purposes only. The data (especially on the clinical aspect) might not have been thoroughly audited, and this might influence the accuracy of the diagnoses. Lastly, this analysis focused only on the national rate which may not capture the inequality in different income levels. For instance, the aggregate rate could reflect a very low rate (underuse) in the lower income group and a high rate (overuse) in the higher income group.
Conclusion
This study confirmed an increasing trend of CS in Thailand from January 2016 to October 2021 based on the analysis of real-world administrative data. The COVID-19 pandemic did not have any significant impact on CS rate. Both clinical and non-clinical factors contributed to CS in various degrees. As such, a better understanding of the factors that affect the decision to perform CS on the service provider side and the decision to undergo CS on the patient side is necessary to support the movement towards reduction in CS. The findings also affirmed that Thailand might be facing the problem of overuse of CS. With this piece of supportive evidence, more actions should be supported to promote the awareness to the general public, health professionals and policy makers to ensure that CS is used appropriately in Thailand. Research on both mother and provider preferences for CS in Thailand, and effective behavioural interventions, would be worthwhile. Overuse of CS is a long-standing issue in Thailand, reflecting entrenched attitudes and processes in the country. The overuse of CS suggests that many Thai mothers are being exposed to the risk of harm for no good reason, and further investigation and action are required.
Supplementary material
Acknowledgements
The authors thank Ms Jutatip Thungthong and Mr Poonchana Wareechi from the National Health Security Office for data support. The authors also thank Mr Pisanu Kantipong for sharing their wisdom and insights during the course of this research. The authors also thank their colleagues at HITAP, especially the CIDHealth team, for support and encouragement in pursuing this project.
Footnotes
Funding: This work was supported by the Health Systems Research Institute (HSRI), Thailand (grant number 64281002RM002L0). The study’s design, data collection, analysis and interpretation, as well as the report’s composition, were all done independently from the sponsors. The conclusions, interpretations and findings presented in this article may not necessarily represent the funding agencies’ point of view.
Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-096080).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study received ethical approval from the Institute for the Development of Human Research Protections (IHRP) in Thailand (Certificate of Approval No. IHRP2020114). Informed consent was not required due to the retrospective nature of the secondary data analysis.
Data availability free text: Based on the confidentiality agreement to access the data with the data custodian, the data sets created and/or analysed during the current study are not publicly available. However, general information on the database is available from the corresponding author on reasonable request.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.World Health Organization WHO statement on caesarean section rates. 2015
- 2.Sandall J, Tribe RM, Avery L, et al. Short-term and long-term effects of caesarean section on the health of women and children. The Lancet. 2018;392:1349–57. doi: 10.1016/S0140-6736(18)31930-5. [DOI] [PubMed] [Google Scholar]
- 3.Gregory KD, Jackson S, Korst L, et al. Cesarean versus vaginal delivery: whose risks? Whose benefits? Am J Perinatol. 2012;29:7–18. doi: 10.1055/s-0031-1285829. [DOI] [PubMed] [Google Scholar]
- 4.Albarqouni L, Arab-Zozani M, Abukmail E, et al. Overdiagnosis and overuse of diagnostic and screening tests in low-income and middle-income countries: a scoping review. BMJ Glob Health. 2022;7:e008696. doi: 10.1136/bmjgh-2022-008696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Betran AP, Ye J, Moller A-B, et al. Trends and projections of caesarean section rates: global and regional estimates. BMJ Glob Health. 2021;6:e005671. doi: 10.1136/bmjgh-2021-005671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zahroh RI, Disney G, Betrán AP, et al. Trends and sociodemographic inequalities in the use of caesarean section in Indonesia, 1987-2017. BMJ Glob Health. 2020;5:e003844. doi: 10.1136/bmjgh-2020-003844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Holmer H, Kamara MM, Bolkan HA, et al. The rate and perioperative mortality of caesarean section in Sierra Leone. BMJ Glob Health. 2019;4:e001605. doi: 10.1136/bmjgh-2019-001605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Torloni MR, Brizuela V, Betran AP. Mass media campaigns to reduce unnecessary caesarean sections: a systematic review. BMJ Glob Health. 2020;5:e001935. doi: 10.1136/bmjgh-2019-001935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Anekpornwattana S, Yangnoi J, Jareemit N, et al. Cesarean Section Rate in Siriraj Hospital According to the Robson Classification. Thai Journal of Obstetrics and Gynaecology. 2020;28:6–15. doi: 10.14456/tjog.2020.2. [DOI] [Google Scholar]
- 10.Sukmanee J, Liabsuetrakul T, Peeyananjarassri K. Rates and Indications of Cesarean Section Using the Robson Classification in a University Hospital in Southern Thailand 2014-2016. Journal of Health Science and Medical Research. 2020;38:307–19. doi: 10.31584/jhsmr.2020750. [DOI] [Google Scholar]
- 11.Yukaew N. Cesarean Section Rate According to Robson’s Classification. Journal of Preventive Medicine Association of Thailand. 2018;7:262–71. [Google Scholar]
- 12.Charoonwatana T, Suwanbamrung C, Saengow U. Cesarean Section According to Robson Classification in a Tertiary Hospital, Southern Thailand. J Obstet Gynecol Cancer Res . 2022;7:213–20. doi: 10.30699/jogcr.7.3.213. [DOI] [Google Scholar]
- 13.National Statistical Office of Thailand . Survey Findings Report: Thailand Multiple Indicator Cluster Survey 2019. Bangkok, Thailand: National Statistical Office of Thailand; 2020. [Google Scholar]
- 14.Liabsuetrakul T, Sukmanee J, Thungthong J, et al. Trend of Cesarean Section Rates and Correlations with Adverse Maternal and Neonatal Outcomes: A Secondary Analysis of Thai Universal Coverage Scheme Data. AJP Rep. 2019;9:e328–36. doi: 10.1055/s-0039-1697656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.World Health Organization WHO recommendations non-clinical interventions to reduce unnecessary caesarean sections. 2018 [PubMed]
- 16.Isaranuwatchai W, Chinnacom D, Poonchai A, et al. Development of The Benefits Package for The Universal Coverage Scheme. 2020.
- 17.World Health Organization . The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research. Genève, Switzerland: World Health Organization; 1993. [Google Scholar]
- 18.World Health Organization ICD - ICD-9-CM - international classification of diseases, ninth revision, clinical modification. 2021
- 19.Makornsarn C, Sungruam S, Unburi S. Thailand Medical Services Profile 2011. Department of Medical Services, Ministry of Public Health; 2014. [Google Scholar]
- 20.College Station, TX: StataCorp LLC; 2021. Stata statistical software: release 17. [Google Scholar]
- 21.University of California Los Angeles . OARC Stats; [3-Jan-2023]. Logistic regression diagnostics.https://stats.oarc.ucla.edu/stata/webbooks/logistic/chapter3/lesson-3-logistic-regression-diagnostics/ Available. Accessed. [Google Scholar]
- 22.Rydahl E, Declercq E, Juhl M, et al. Cesarean section on a rise-Does advanced maternal age explain the increase? A population register-based study. PLoS ONE. 2019;14:e0210655. doi: 10.1371/journal.pone.0210655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bergholt T, Skjeldestad FE, Pyykönen A, et al. Maternal age and risk of cesarean section in women with induced labor at term—A Nordic register‐based study. Acta Obstet Gynecol Scand. 2020;99:283–9. doi: 10.1111/aogs.13743. [DOI] [PubMed] [Google Scholar]
- 24.Luke B, Brown MB. Elevated risks of pregnancy complications and adverse outcomes with increasing maternal age. Hum Reprod . 2007;22:1264–72. doi: 10.1093/humrep/del522. [DOI] [PubMed] [Google Scholar]
- 25.Dhakal-Rai S, van Teijlingen E, Regmi PR, et al. Factors contributing to rising cesarean section rates in South Asian countries: A systematic review. Asian J Med Sci. 2022;13:143–74. doi: 10.3126/ajms.v13i2.40904. [DOI] [Google Scholar]
- 26.Aasheim V, Waldenström U, Rasmussen S, et al. Experience of childbirth in first-time mothers of advanced age - a Norwegian population-based study. BMC Pregnancy Childbirth. 2013;13:53. doi: 10.1186/1471-2393-13-53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.IC R, DE D, D M, et al. Vaginal Delivery in a Cohort of Pregnant Women Aged 40 or Older. Austin J Obstet Gynecol. 2014;1 [Google Scholar]
- 28.Einarsdóttir K, Haggar F, Pereira G, et al. Role of public and private funding in the rising caesarean section rate: a cohort study. BMJ Open. 2013;3:e002789. doi: 10.1136/bmjopen-2013-002789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Saengruang N, Witthayapipopsakul W, Rajatanavin N. The management of provider payments in the universal coverage scheme (UCS) in Thailand: national health security office. 2020
- 30.Panda S, Daly D, Begley C, et al. Factors influencing decision-making for caesarean section in Sweden - a qualitative study. BMC Pregnancy Childbirth. 2018;18:377. doi: 10.1186/s12884-018-2007-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tienmontree K, Chunuan S, Kala S. Factors Influencing Decision-Making for Cesarean Section among Primigravidae. Nursing Journal of The Ministry of Public Health. 2021;31:96–109. [Google Scholar]
- 32.Mulchandani R, Power HS, Cavallaro FL. The influence of individual provider characteristics and attitudes on caesarean section decision-making: a global review. J Obstet Gynaecol. 2020;40:1–9. doi: 10.1080/01443615.2019.1587603. [DOI] [PubMed] [Google Scholar]
- 33.Hoxha I, Lama A, Bunjaku G, et al. Office hours and caesarean section: systematic review and Meta-analysis. Res Health Serv Reg . 2022;1:4. doi: 10.1007/s43999-022-00002-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Teerawattananon Y, Isaranuwatchai W, Karunayawong P, et al. Proposal for adjusting reimbursement rate for childbirth in hospitals to reduce unnecessary cesarean sections and promote childbirth in Thailand. 2024. https://www.hitap.net/documents/190361 Available.
- 35.Hanvoravongchai P, Letiendumrong J, Teerawattananon Y, et al. Implications of Private Practice in Public Hospitals on the Cesarean Section Rate in Thailand. Hum Resour Develop. 2000;4 [Google Scholar]