Caesarean sections before and during the COVID-19 pandemic in western Sydney, Australia

Abstract Background To determine the changes in emergency and elective caesarean section (CS) rates since the COVID-19 pandemic, identify the groups most affected, and examine changes in the factors associated with CS rates, and reasons for CS. Methods We conducted a retrospective cohort study using routinely collected data of 22,346 births from before the pandemic (January 2018–February 2020) and 18,597 births during the pandemic (March 2020–December 2021). Data were analysed using multinominal logistic regression. Results The CS rate increased by 4.1% (from 30.1% to 34.2%), reflecting increases of 2.3% in emergency CS (from 11.5% to 13.8%) and 1.7% in elective CS (from 18.7% to 20.4%). Large groups with notable increases were women who were nulliparous (7.2% increase), from South Asia (6.0%), obese (5.2%) and giving birth at a small hospital (6.1%). Compared to pre-pandemic, the relative risk of an emergency CS versus a vaginal delivery increased 1.36 times (adjusted relative risk ratio (aRRR) = 1.36; 95% CI = 1.27, 1.45) and the risk of having an elective CS increased 1.11 times (aRRR = 1.11; 95% CI = 1.04, 1.20). Factors associated with both emergency and elective CS were age, region of birth, reproductive history, body mass index, hypertension, diabetes, mode of antenatal care and hospital. Socio-Economic Indexes for Areas and antenatal care were only associated with elective CS. Baby gender was only associated with emergency CS. Preterm gestation at delivery was associated with reduced emergency but increased elective CS. Foetal compromise was the most common indication for emergency CS (43.2%) and increased the most (8.0%). Previous CS was the most common indication for elective CS (61.5%) and reduced the most (1.9%). Conclusions Both emergency and elective CS rates increased significantly during the pandemic, with the former increasing at a higher rate. The persistent upward trend of CS rates, exacerbated by increasing proportions of nulliparous women undergoing CSs, is concerning. Plain Language Summary Australia has a very high caesarean section (CS) rate that varies greatly between groups of women with different socio-economic characteristics and reproductive histories. Information regarding changes in CS rate since the COVID-19 pandemic in Australia is limited. We conducted a study comparing CS rate before and during the pandemic, using routinely collected data. Both emergency and elective CS rates increased significantly during the pandemic with emergency CSs increased at a higher rate than elective CSs. Several groups of women experienced large increases in CS rate. Factors associated with and reasons for emergency CSs were different from those for elective CSs. Health services should be prepared to minimise effects of future pandemics on CS rate. To be most effective, interventions to reduce non-medically justified CSs should focus on women who are from South Asia, obese, admitted to a small hospital, and are nulliparous. Different approaches are needed to reduce emergency and elective CSs.


Introduction
Caesarean section (CS) rates are increasing globally, varying from 5% in developing countries to 43% in developed countries (Betran et al. 2021).Lack of access to CS can be dangerous for mothers and babies, but overuse of CS can be harmful because as with any surgery, it carries risks.Its overuse may also waste resources (WHO 2015).In 2019, Australia and New Zealand had a CS rate of 33.5%, one of the highest in the world (Betran et al. 2021).
The COVID-19 pandemic has affected all aspects of life including maternal care and childbirth.Since the pandemic, inconsistent changes in CS rates have been reported.Rates have decreased in Iceland (Einarsdottir et al. 2021), remained unchanged in China (Li et al. 2020) and increased slightly in England (Bhatia et al. 2021, Gurol-Urganci et al. 2022).
The pandemic in Australia began in March 2020.Compared to other countries, the number of COVID-19 cases in Australia has been relatively low (AIHW 2021a).Western Sydney is a metropolitan area of approximately one million people.The Local Health District (LHD) has three public hospitals, each with a different level of Maternity and Neonatal Service capability: the smallest, a level 3 hospital provides care for low risk women, a medium level 5 hospital provides care for medium risk women and a level 6 principal referral hospital cares for women with any level of risk.Around twothirds of pregnant women in the LHD were born overseas (Trinh et al. 2018), compared to one-third across the state of New South Wales (Trinh and Rubin 2006).The CS rate varies greatly by country of birth (Trinh et al. 2020).No COVID-19 infections among pregnant women were reported in 2020; approximately 100 were reported in 2021.
Information is limited regarding the effects of the pandemic on the overall CS rate and among different groups of women in the LHD and in Australia.The pandemic may have exacerbated the disparity in CS rates between these groups.The aims of this study were to determine the changes in emergency and elective CS rates since the COVID-19 pandemic, identify the groups most affected, and examine changes in the factors associated with CS rates, and reasons for CS.

Study design
This was a retrospective cohort study using routinely recorded information in the eMaternity database on all women who gave birth in the LHD's public hospitals.Data on all births of babies �20 weeks gestation or �400 g birthweight were included.In the case of multiple births, only the record of the first baby was included.Data examined included demographic and health behaviour characteristics (e.g.age, region of birth, level of socio-economic disadvantage, and smoking status), medical history (e.g.height, weight, hypertension, and diabetes), reproductive history (previous pregnancy), care during pregnancy (first antenatal visit, mode of antenatal care), labour and birth outcomes (gestational age, baby birthweight, and gender) (Table 1).The coding and capture of data were consistent between before and during the pandemic.
Data of women who gave birth between January 2018 and February 2020 (the 'before' group, 26 months) were compared with data of women who gave birth between March 2020 and December 2021 (the 'after' group, 22 months).The number of women (22,346 before and 18,597 after) was large enough to allow detection of any substantial differences.
Countries of birth (COB) were grouped into regions based on similarities in socio-economic characteristics and the geographic closeness of the countries.Socio-Economic Indexes for Areas (SEIFA) quintiles were determined using Local Government Area (LGA) codes, which represent local government boundaries.A CS was defined as emergency if the woman had started labour, and as elective if the CS occurred without labour.There were no missing data for the outcome variables (mode of birth).Instances of missing data accounted for less than 1% for all variables except SEIFA (3.3%), which was still very low.Missing values were excluded from univariate analyses, but the observations were still included in the multiple analyses.
Changes in CS rates per 100 births from pre-pandemic to during the pandemic were calculated.Descriptive analyses were used to describe proportions and means.Significant differences in proportion were identified using the Chi-squared test, and differences in mean were identified using one-way analysis of variance (ANOVA).
Multivariable multinominal logistic regression was used to identify factors associated with emergency and elective CSs, using vaginal delivery as the base outcome.Factors potentially associated with CSs were the pandemic, demographic characteristics, medical histories, reproductive history, care during pregnancy and birth outcomes.Reduced models were fitted using backward stepwise regression technique.All variables with a p value �.10 were included in the final models.Data were analysed using Stata 14 (StataCorp 2015).Ethics approval was obtained from the Local Health District Human Research Ethics Committee (2107-13 QA APPROVAL).

Maternal characteristics
There were 41,557 babies with gestational age �20 weeks gestation or �400 g birthweight (22,681 before and 18,876 after) from 40,938 births (22,346 before and 18,592 after) (Figure 1).Nearly, half of the births occurred at the principal referral hospital (49.1% after).One-third of women were aged between 30 and 34 years (37.3%) or were nulliparous (39.0%).Two-thirds of women were immigrants (64.7%), half originating from South Asia (30.0%).The largest change since the pandemic was the increase in the proportion of women who had their first antenatal comprehensive assessment within 10 weeks (15.5%, p < .001)(Table 1).

CS rate
There were 6737 (30.1%) women who had a CS before the pandemic and 6367 (34.2%) had a CS after, a significant increase of 4.1%, p < .001.The increase was observed in both emergency CS (from 11.5% to 13.8%, an increase of 2.3%), and elective CS (from 18.7% to 20.4%, an increase of 1.7%).There was a shift towards elective CS among the public hospital high-risk group and towards emergency CS among team midwifery and shared care groups.The rate was highest among multiparous women with a prior CS (85.7%) and lowest among multiparous women without a prior CS (11.4%) (Figure 2).

Groups most affected
Large groups of women with a notable increase in CS rate were nulliparous (2653 or 39.4% of all CSs after; rate increased 7.2-36.6%),from South Asia (2381 CSs or 35.5% of all CSs after; rate increased 6.0-42.7%),obese (1548 CSs or 23.0% after; rate increased 5.2-43.0%)and gave birth at the smallest hospital (726 CSs or 10.8% after; rate increased 6.1-30.9%)(Table 1).
All factors except smoking were associated with either emergency or elective CS (Table 2).Higher SEIFA and early entry to antenatal care were associated with only increased elective CS.Having a male baby was only associated with emergency CS.The proportion of macrosomia among male babies was 1.47% compared to 0.88% among female babies (p < .001).Preterm gestation at delivery was associated with reduced emergency (aRRR ¼ 0.46; 95% CI ¼ 0.38, 0.57) but increased elective CS (aRRR ¼ 2.39; 95% CI ¼ 2.04, 2.80).Previous CS had a much higher impact on elective (aRRR ¼ 62.93) than on emergency CS (aRRR ¼ 11.98).

Reasons for CS
Previous CS was the most common reason for a CS both before and during the pandemic and reduced the most during the pandemic (from 41.4% to 38.1%, a reduction of 3.4%) and for elective CS (from 63.4% to 61.5%, a reduction of 1.9%).Foetal compromise, the second most common reason overall, increased the most (by 4.4-23.4%)and was the most common reason for emergency CSs (an increase of 8.0-43.2%).These two causes accounted for nearly two-thirds of all CSs (61.5%) (Table 3).Previous CS was the main reason for CS among multiparous women with a previous CS (83.2% after), while foetal compromise was the main reason for CS among nulliparous women (40.1% after) and multiparous women without a CS (29.7% after).

Discussion
We examined the incidence of CSs before and during the COVID-19 pandemic in all hospitals in a LHD.The CS rate increased by 4.1%, reflecting significant increases in both emergency and elective CSs.Large groups with significant increases were nulliparous women, women born in South Asia, obese women and women who gave birth in the smallest hospital.
While most of the women's characteristics changed little during the pandemic, the proportion of women who had early comprehensive assessment increased by 15.5%.This could be the direct result of the 'Baby Steps' project that was implemented at the small hospital to increase the effectiveness of maternal care services.

High and increasing CS rate during the pandemic
One in three women gave birth by CS.Several potential issues, such as older maternal age (Stastna et al. 2022), obesity (RANZCOG 2017) and clinicians' skill and experience (Panda et al. 2018) could be contributing factors.
Despite the low prevalence of COVID-19 infection, the CS rate increased during the pandemic.Many factors may have contributed to this.Restrictions, including a curfew, that affected freedom of movement were imposed during the pandemic.Reduced physical exercise (Nielsen et al. 2017), increased obesity (Al-Kubaisy et al. 2014, RANZCOG 2017) and extra stress caused by the pandemic (Kotlar et al. 2021) have been linked with an increased CS rate (Zochowski et al. 2021).
Hospital guidelines and practices in response to the pandemic might have also contribute to the increased CS rate.Although the hospitals quickly implemented measures to prevent infection such as having separate wards to treat infected patients and providing personal protective equipment (e.g.masks, face shields and goggles), some staff thought further measures could have improved safety such as establishing a direct access to COVID-19 wards and having sufficient and regular supply and training of personal protective equipment (Trinh et al. 2022).The potential effects of other factors, namely staff and public anxiety and staff furloughing, on decisions about minimising risk and duration of the birth process through CS should not be overlooked.Staff might feel anxious and lower threshold for a CS, to shorten the time the pregnant women spent in the hospital, in the hope of minimising the risk of COVID-19 infection.During the pandemic, no support person was allowed during labour and birth.This could cause women wanting to bring forward the birth process.These reasons could also explain the increase in rates of induction and CS among inductions.In addition, the 'Safer Baby Bundle' project introduced in October 2019 to reduce stillbirth (CEC. 2019) may also have contributed to CS rates by encouraging women to focus on and report decreased foetal movements, thereby increasing consideration of induction and CS.
Care throughout pregnancy might also affect the CS rate.During the pandemic, the use of telehealth consultations in the LHD increased but did not compensate for the reduction in face-to-face visits, resulting in a 7.1% overall reduction of consultations (Trinh et al. 2022).In the case of suspected foetal compromise, which demands expertise in decision making and physical skills (Luesley and Kilby 2016), the identification of risk factors and subsequent use of electronic foetal monitoring might not have been to their usual capacity amidst pandemic demands.Furthermore, women might have had fewer opportunities to visit the birthing unit and foster a patient-clinician relationship.Such factors might have fuelled anxiety and CS rates.

Groups of women most affected
Women most at risk of having a CS were nulliparous, from Southern Asia, diabetic or attending the small hospital.A high CS rate among nulliparous women, corroborating previous findings (Trinh et al. 2020), increased by 7.2% during the pandemic.Pandemic-related changes that may have negatively affected women include the substitution of some faceto-face antenatal care (Trinh et al. 2022) and classes with online sessions, the discontinuation of tours of the birth unit, and the temporary ban of a support person during birth.These factors might have increased anxiety and stress surrounding childbirth, especially among nulliparous women (Smith et al. 2019).In addition to efforts aimed at reducing CSs among multiparous women with a prior CS, measures to reduce CSs among nulliparous women would be particularly effective in controlling the persistent upward CS trend.Women from South Asia accounted for one-third of all women and one-third of all CSs.They experienced a particularly high proportion of CSs (42.7%) and increase in CSs (6.0%).The increase could be a mirroring effect of that experienced in their COB, such as India, where increased maternal literacy, age and wealth have coincided with a higher CS rate (Roy et al. 2021).The association between obesity and shoulder dystocia that may necessitate CS is more profound among South Asian than Australian-born women, possibly due to a smaller pelvic among South Asian women (Davies-Tuck M et al. 2016).Tailored interventions in appropriate languages that consider traditional beliefs and practices during  pregnancy and childbirth of women from South Asia could be useful to reduce CS rate among this large group of immigrants.
The increase in both the proportion of women who were obese and obese women who had a CS is concerning.Obesity is associated with many complications such as gestational hypertension, preeclampsia, gestational diabetes and heavy foetal birth weight.Obese women might have been more anxious about the diagnosis and treatment of these complications during the pandemic, increasing their stress and the likelihood of a CS (Zochowski et al. 2021).
Despite the increase in CS rate at the small hospital (6.1%), the CS rate was still lower than in other hospitals.The increase in early antenatal care at the small hospital due to the Baby Steps project (ACI 2021) may have largely contributed to the CS increase.Early entry to antenatal care has been reported to be associated with a higher CS rate (Trinh et al. 2018, Fabbro et al. 2022).Women with higher risk status might seek early entry to antenatal care.It could also be that the earlier the antenatal care, the more risk factors are diagnosed, leading to more investigations and interventions.Other influential factors could be changes in policies, such as the Safer Baby Bundle, and staffing, i.e. more COVID-related absenteeism.
The fact that women with pre-existing diabetes and women having public hospital high-risk care had fewer emergency CS and more elective CS might suggest improvement in the screening for elective CS among women with risk factors.The shift towards emergency CS among women with team midwifery care could reflect disruption to normal care during the pandemic.

Differences in factors associated with emergency and elective CS versus vaginal birth
In addition to the above large groups, other groups of women also had increased CS rate.Similar to findings from other studies, factors associated with CS were younger or older age, region of birth, multiparous women with previous CS (Trinh et al. 2020, AIHW 2021b), higher BMI (Berendzen and Howard 2013), hypertension, having private antenatal care, living in high socio-economic area (Trinh et al. 2018, Roy et al. 2021), having a low-birthweight baby (Temerinac et al. 2015, Trinh et al. 2018, Roy et al. 2021) and having a male baby (Lieberman et al. 1997).
In our study, having a male baby was associated with emergency CS.As previously hypothesised (Lieberman et al. 1997), macrosomia was more common among male babies and might partly explain the observed gender difference.In addition, failure to progress, as opposed to foetal distress, can result in a caesarean delivery as may be relevant to the size of the male foetus.
Planned preterm delivery is often the result of complications, which may increase the need for an elective CS.As expected, we found that preterm birth was associated with higher risk of elective CS but lower risk of emergency CS.Fewer preterm pregnancies go into labour and result in an emergency CS.

Changes in reported reasons for CS during the pandemic
Previous CS was documented as a reason for a third of all CSs and was a main cause of CSs among multiparous women with a previous CS.A previous CS does not necessitate subsequent CSs, as evidenced by the increase in vaginal birth after a caesarean section (VBAC) in the United States in the mid-1980and -1990s (Habak and Kole 2022).The recommendation in New South Wales is for 60% of women to undergo VBAC (NSW Health 2010).Clinicians may be hesitant to encourage a VBAC, aware of the potential for serious complications, particularly in the presence of previously identified CS risk factors such as obesity, diabetes and hypertensive disorders (Wu et al. 2019).Interestingly, the CS rate among multiparous women with a previous CS reduced by 3.4% during the pandemic.It is possible that the fear of contracting COVID-19 during a CS hospital stay might have deterred some women.
Interventions and education for pregnant women and clinicians are needed as the preponderance of CS further supports its use by clinicians, who in turn are often relied upon to make decisions in the best interest of their patients (Loke et al. 2019).Trial of labour after caesarean (TOLAC) (ACOG 2019), a planned attempt to labour for women opting for a VBAC undertaken by an interdisciplinary team, has had reported success in the United States (Scott 2011).
The increase in 'foetal compromise' might relate to the physical (Nielsen et al. 2017) or mental (Kotlar et al. 2021) decline experienced by some women during the pandemic, physicians' pandemic-related lower threshold for CS, or may reflect decisions in line with the World Health Organization for the conduct of CSs when medically warranted (WHO 2015).

Strengths and limitations
A strength of this study was its most up-to-date data over of a four-year period, including 22 months during the pandemic, enabling assessment of change during the short and medium-term periods of the pandemic.Data of all women residing in the LHD and giving birth at one of its three public hospitals were included.Public patients account for 90% of all women giving birth in the LHD (Trinh et al. 2018), attesting to the representativeness of the results.Data on emergency and elective CS were analysed separately, enabling meaningful interpretation of the results.A comprehensive range of associated factors was included in the analyses to allow adjustment for confounders.
Data for comorbidities such as heart disease, HIV and genital herpes were not available.The presence of these medical conditions might shed further light on factors associated with the observed high and increasing rates of CSs.However, we were able to confirm that conditions such as obesity, hypertension and diabetes were strongly associated with increased CSs.The administrative data utilised were not primarily collected for research purposes.However, validation studies have confirmed the high quality of the maternal hospital data (NSW Health 2000).
Changes in hospital guidelines, training, staffing and supervisions since the pandemic would have affected healthcare services and could have impacted on aspects of care during pregnancy and childbirth.However, the effects of these changes were not measured.This study was conducted in one metropolitan LHD in western Sydney and therefore results can only be generalised to other hospitals servicing areas with similar demographics.

Conclusions
Both emergency and elective CS rates increased during the pandemic.The persistent upward trend, enhanced by increasing proportions of nulliparous women undergoing CSs, is concerning.Among the large groups, women from South Asia, obese women and women attending a small hospital experienced high increases in CS rates.Future maternal programs such as the 'Safer Baby Bundle' and 'Baby Steps' should consider their potentially inadvertent effect of increases in elective CSs.Interventions to promote CSs only when medically justified should focus on identified vulnerable groups, especially nulliparous women, to prevent subsequent CSs.Future research should include comorbidities whenever possible.

Figure 1 .
Figure 1.Records included in the study.

Figure 2 .
Figure 2. Caesarean section rate before and during the COVID-19 pandemic: categories and women's characteristics.

Table 1 .
Characteristics and clinical factors of women who gave birth and of women who had caesarean section, 2018-2021.

Table 2 .
Continued.Model adjusted for all factors except SEIFA Index, antenatal care and smoking.b Model adjusted for all factors except smoking and baby's gender.c Different effects on emergency and elective CS. a

Table 3 .
Top 10 reported reasons for caesarean section.