By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
Please cite this paper as: Waldenström U, Gottvall K, Rasmussen S. Caesarean section in nulliparous women of advanced maternal age has been reduced in Sweden and Norway since the 1970s: a register-based study. BJOG 2012;119:1591–1596.
Objective To investigate rates of caesarean delivery in Sweden and Norway from 1973 to 2008 in relation to advanced and very advanced maternal age.
Design Register study.
Setting Sweden and Norway.
Sample All nulliparous women aged over 30 years with a singleton pregnancy, with the fetus in a cephalic presentation, and delivering at term between 1973 and 2008 were evaluated. The study population comprised 329 824 women in Sweden and 127 810 women in Norway.
Methods Data from the national Medical Birth Registers were used to describe caesarean section rates in three age groups: 30–34 years (reference group); 35–39 years (advanced age group); and ≥40 years (very advanced age group). Logistic regression analyses estimated the risk in each age group over four decades, in each of the two national samples.
Results Caesarean delivery decreased from 1973–1979 to 2000–2008 in the two oldest age groups in Sweden (35–39 years, OR = 0.53, 95% CI = 0.50–0.58; ≥40 years, OR = 0.36, 95% CI = 0.30–0.43) and Norway (35–39 years, OR = 0.61, 95% CI = 0.54–0.68; ≥40 years, OR = 0.45, 95% CI = 0.34–0.58), but increased in women aged 30–34 years. The caesarean delivery rate in the two oldest groups peaked in the second half of the 1970s. Regardless of time point, the caesarean delivery rate was always highest in women aged ≥40 years, followed by women aged 35–39 years and lowest in women aged 30–34 years.
Conclusions Caesarean delivery in nulliparous women of advanced and very advanced age peaked by end of the 1970s in Sweden and Norway. The subsequent reduction was contemporaneous with the introduction of electronic fetal monitoring and a more consistent use of the partogram, suggesting that more effective surveillance of labour increased the chance of a vaginal birth in these high-risk women.
The risk of caesarean delivery increases with maternal age,1 and delaying childbirth until the late 30s, and beyond, has become a more frequent phenomenon over the past decades in the developed world. Delaying childbirth may therefore contribute to the rising rates of caesarean section.2 According to the national Medical Birth Registers, introduced in Sweden in 1973 and in Norway a few years earlier, maternal age at first birth increased from means of 23.9 and 23.0 years in 1973 to 28.4 and 27.5 years in 2008 in Sweden and Norway, respectively.3,4 During the same time period the national rates of caesarean section increased from 5.3 to 17.2% in Sweden,3 and from 3.0 to 17.2% in Norway.5 We are not aware of any contemporary population-based studies that have been performed in societies where the postponement of childbirth has become increasingly more common that have investigated the development of caesarean section rates in women of advanced age, over a longer period of time.
It has not yet been established to what extent biological factors explain the positive association between maternal age and caesarean section. The higher prevalence of dystocia,6–9 malpresentation,8–10 and preterm birth,11–13 diagnoses associated with caesarean delivery, in older nulliparous women are probably related to the ageing processes in the uterus and placenta. Main et al.7 found a continuously increasing risk of uterine dysfunction related to maternal age in nulliparous women. Smith et al.14 also suggested that there is a true biological effect of age based on their findings of a linear relationship between maternal age across the age span and the risk of caesarean delivery, operative vaginal delivery and long duration of labour. Different theories have been presented regarding possible mechanisms in the ageing myometrium and decidua, such as hormonal effects on the uterus,15 and decreased uterine contractility and number of oxytocin receptors.8,16 Biological mechanisms, such as low uteroplacental perfusion, caused by poor uteroplacental vasculature, have also been discussed in relation to the increased risks of intrauterine growth retardation and fetal death in older women.17
The aim of our study was to investigate the development of nulliparous term singleton cephalic caesarean delivery rates in Sweden and Norway from 1973 to 2008, in relation to advanced (35–39 years) and very advanced (≥40 years) maternal age.
Changes in obstetric problems and practice over the last 35 years may contribute to differences in caesarean deliveries when women of advanced maternal age are compared with younger women. Rates of caesarean delivery in older women may be increased compared with younger women because of well-recognised age-related medical problems, such as diabetes,8–10,13,18–20 and obesity,13,21 which have become more prevalent in both Sweden and Norway.22,23 Also, the patient’s own preference is more often taken into account than some decades ago.24 A Swedish population-based study from the year 2000 showed that older women more often expressed a wish to be delivered by caesarean section than younger women.25 On the other hand, caesarean deliveries in older women may be reduced since the 1970s as a consequence of the natural childbirth movement, which peaked in Sweden around 1985–1995, and in Norway some years later. The influence of obstetricians and midwives management preferences on the relationship between maternal age and caesarean delivery rates may also be an important factor.
We obtained data sets from the Swedish and Norwegian Medical Birth Registers for all nulliparous women aged 30 years and older who had given birth in the period from 1973, when the Swedish Register commenced, to the year 2008. The Swedish Medical Birth Register covers 98–99% of all births, and is based on the medical records used in antenatal, intrapartum and postpartum care. Starting with the first antenatal visit, information is collected prospectively, including medical and obstetric interventions, pain-relieving methods, and maternal and perinatal complications during labour and delivery.26 The Norwegian register started in 1967, and was one of the first national programmes that was not just based on the registration of birth defects but, similar to the Swedish register, included the compulsory notification of all births in the country, collected prospectively from the first booking visit in early pregnancy to the postnatal period.27
By including information from two independent national samples we hoped to get valid data about the effect of advanced maternal age on caesarean delivery practices. Sweden and Norway have much in common, such as high socio-economic standards, health-care systems with great similarities, including obstetricians and midwives being trained and practicing in a similar way, and comparable health outcomes, with a female life expectancy at birth of 83 years in both countries. The rates of caesarean section (17%) and perinatal mortality (4.4/1000 estimated from 28 weeks of gestation) were the same in Sweden and Norway in 2008, and are relatively low compared with many other countries. One difference that could possibly affect caesarean delivery practices is related to the organisation of maternity care, which was more decentralised in Norway than in Sweden, with a larger proportion of smaller, midwife-led units operating in remote areas. However, more recently maternity care in Norway has become more centralised, but this development started around 10 years later than in Sweden, in the 1980s.28 Also, Norwegian midwives commenced providing antenatal care only recently, whereas Swedish midwives have a long history as primary caregivers during normal pregnancies.
In the Norwegian data set maternal age was categorized into three age groups, 30–34, 35–39 and ≥40 years, whereas the Swedish data set included age for each individual. We chose to present most of our findings by the three age categories, and defined advanced maternal age as 35–39 years and very advanced age as ≥40 years. We used the age group 30–34 years as the reference group.
We included women at term (≥37 weeks of gestation) with one fetus in cephalic position, which gave 329 824 women in the Swedish sample and 127 810 women in the Norwegian sample. We present the annual caesarean section rate in each age group from 1973 to 2008. Logistic regression analysis was used to calculate crude odds ratios (not shown) with 95% confidence intervals for having a caesarean section in each age group in each of four decades, 1973–1979, 1980–1989, 1990–1999 and 2000–2008, and adjusted odds ratios that took into account possible changes in the population over time. Besides the four time periods, the following variables were included in the multivariate analyses: marital status; chronic diseases, reported at the first antenatal visit during pregnancy (diabetes, chronic hypertension, asthma, kidney disease and epilepsy); infant birthweight; and epidural analgesia. Swedish data were also adjusted for country of birth. Lack of consistency or lack of data over the entire observation period in the respective data set precluded adjustment for body mass index and smoking.
The annual caesarean delivery rates in the three age groups are presented in Figure 1, together with the total nulliparous, term, cephalic, caesarean delivery rates. In both countries caesarean delivery was consistently more prevalent in women of very advanced age (≥40 years), followed by women of advanced age (35–39 years) and then by women aged 30–34 years. The general development was similar in the two countries, with the highest rates in the 1970s in women of very advanced and advanced age. In both of these age groups, the caesarean delivery rate increased steeply from 1973, but peaked slightly earlier in Sweden than in Norway: in 1977 (80.4%) in Swedish women of very advanced age and in 1979 (75%) in Norwegian women of very advanced age; and in women of advanced age in 1976 (45.9%) and 1979 (38%), respectively. In women aged 30–34 years, caesarean delivery increased slowly over the observation period in both countries, and this development followed the same pattern as the total rate, including all age groups from teenagers to the oldest, but at a slightly higher level. The differences between age groups decreased from the 1970s to the 1990s, but were then fairly stable. During the last two decades the difference between the very advanced and the advanced age groups ranged from 12 to 13.4 mean percentage points in Sweden and from 12.6 to 17.2 mean percentage points in Norway, and between the advanced age group and women aged 30–34 years from 7.5 to 8.1 mean percentage points in Sweden and from 7.2 to 8.5 mean percentage points in Norway (for the exact caesarean section rates in the respective decades, see Table 1).
Table 1. Term singleton cephalic caesarean sections during four time periods ranging from 1973 to 2008 in Sweden and Norway in nulliparous women aged 30–34, 35–39 and ≥40 years
Sweden (n = 329 824)
Norway (n = 127 810)
CS n (%)
CS n (%)
Crude and adjusted odd ratios (OR) with 95% confidence intervals (95% CI).
*Civil status (living with baby’s father versus not living with baby’s father), country of birth (Sweden versus not Sweden), chronic disease reported at first antenatal visit (asthma, diabetes, chronic hypertension, kidney disease or epilepsy), epidural analgesia and infant birthweight.
**Civil status (married or cohabiting versus not married or cohabiting), chronic disease reported at first antenatal visit (asthma, diabetes, chronic hypertension, kidney disease or epilepsy), epidural analgesia and infant birthweight.
Maternal age 30–34 years
16 678 (16.5)
Maternal age 35–39 years
Maternal age ≥40 years
The risk of having a caesarean delivery was strikingly reduced from the 1970s to 2000–2008 in women of very advanced age, particularly in Sweden, and the same pattern was found in women of advanced age (Table 1). In contrast, the risk of having a caesarean birth in women aged 30–34 years increased over time. The odds ratios for all three age groups were lowest in the 1990s in both Sweden and Norway, except in women of very advanced age in Norway during the period 2000–2008.
Figure 2, with maternal age on the x-axis, shows that caesarean deliveries in Sweden in the 1970s slowly increased from 30 to 34 years of age, and then more dramatically at the age of 35 years, whereas during the following decades a sharp deviation was reached at age 40 years and older.
We found that the caesarean delivery rates in women of advanced and very advanced age were highest during the first years of our observation period: in Sweden around the years 1974–1985, and in Norway during the period 1977–1990. In contrast, caesarean delivery in women aged 30–34 years increased more or less continuously from 1973 to 2008 in both countries, in the same way as the total national rates for all nulliparas (term, singleton, cephalic).
The years with the highest caesarean delivery rates in the two oldest age groups were contemporaneous with a time when electronic fetal monitoring (CTG) was still not introduced in all delivery units in the respective countries. In Sweden, for example, CTG was mostly used by researchers prior to 1975, and was introduced in all university clinics around 1980 and in all delivery wards around 1985. Also, the partogram was still not used consistently in all delivery units. Discussion of our findings with senior obstetricians with clinical experience from the 1960s and beyond supports our interpretation that better tools to monitor the progress of labour, such as CTG monitors and the partogram, partly explains the reduction of caesarean delivery rates in the oldest women after the peak in the second half of the 1970s. Prior to the introduction of these methods pregnancies were treated more similarly by group of risk category, such as ‘nulliparous 35 years or older’, which was an unquestioned risk category specified by guidelines from the National Board of Health and Welfare in Sweden,29 and by the Society of Obstetrics and Gynaecology in Norway.30 When caesarean section became a more safe procedure and was used more frequently in the beginning of the 1970s, it almost became the standard mode of delivery for nulliparous women of very advanced age. A trial of labour was regarded as more risky, especially when considering that it might be the last chance to have a child when assisted reproduction techniques were still not available. This interpretation is supported by Figure 2, which shows that the caesarean delivery rate in Sweden during the 1970s increased sharply at the age of 35 years, which then was the cut-off for high-risk pregnancy, whereas the development during the last two decades shows a more continuous increase by maternal age.
This interpretation of our findings suggests that the introduction of CTG, where the evidence for use in low-risk women is uncertain,31 together with the partogram, increases the chance for women pregnant with their first baby at an advanced and very advanced age having a vaginal birth. Still, we cannot exclude that the growing critique of what was perceived as medicalisation of childbirth during the 1980s, in combination with women becoming more vocal and aware of their rights, also had an effect by reducing on the very high caesarean delivery rates in older women during the first decade of our observation period.
In both Sweden and Norway the caesarean delivery rates in older nulliparas stabilised from the early 1990s, and then followed approximately the same development as in women aged 30–34 years, but at a higher level. This stability suggests that biological aging could be an important explanation of the observed differences between the age groups. The magnitude of these differences would then give a hint about the size of the biological effect. However, the indications for caesarean delivery are probably interpreted more widely in older nulliparas, as advanced maternal age is still a risk factor and the pregnancy is more often thought to be the last in the woman’s reproductive life. The biological effect should therefore be less than the observed differences of around 7–8 percentage points between the age groups of 30–34 and 35–39 years, and 12–17 percentage points between the age groups of 35–39 and 40 years and older.
In all age groups caesarean delivery was least common during the 1990s, and this was most obvious in Sweden. The late 1980s was a period when the medicalisation of childbirth was a hot topic in the media in Sweden, and alternative models of maternity care, such as in-hospital birth centres, were most popular, and this trend could possibly have reduced the caesarean delivery rates in general. In Norway, the natural childbirth movement appeared a few years later. In both countries and in all three age groups caesarean deliveries increased during the latest observation period of 2000–2008. This development has not been related to a general decline in health status, but rather to changes in the attitudes of caregivers,32 and to some extent on patient request,33 although more than 90% of Swedish women would prefer a vaginal birth.34 Our data do not support the hypothesis that caesarean delivery rates in women of advanced and very advanced age decrease when the caregivers get more used to this patient group.
Both the strength and limitations of this study are related to the characteristics of the two data sources. Important information, such as the distinction between elective and emergency caesarean section, and maternal background characteristics, such as body mass index and smoking habits, were not available for the complete observation period. The strength of the study was that almost all nulliparous women aged 30 years and older who gave birth at term with one fetus in cephalic position in Sweden and Norway from 1973 to 2008 were included in the study. By including data from two countries, where the collection of national data on all births commenced earlier than in other countries, we could provide comprehensive and valid data about clinical practices regarding caesarean delivery rates in older nulliparous women.
Caesarean delivery in nulliparous women of advanced and very advanced age peaked by the end of the 1970s in Sweden and Norway. The subsequent reduction was contemporaneous with the introduction of electronic fetal monitoring and a more consistent use of the partogram, suggesting that more effective surveillance of labour increased the chance of a vaginal birth in these high-risk women.
The fact that caesarean delivery rates have not increased as steeply in Sweden and Norway as in many other countries, and that caesarean delivery in older women is now less common than 20 years ago, suggests that there is scope for reducing the very high rates observed in many countries.
Disclosure of interests
We are not aware of any conflict of interest in relation to this article.
Contribution to authorship
Conceived and planned by UW and SR. Performed by UW, SR and KG. Analysed by UW, SR and KG. Written by UW. Both SR and KG contributed to writing the article with major comments and editing.
Ethical approval was obtained from the Swedish and Norwegian Medical Birth Registers when approving our application for the respective data sets.
This study was funded by the Swedish Research Council, Karolinska Institutet, Stockholm, Sweden, and by Bergen University College, Norway.
We are grateful to all the women, midwives and obstetricians who have provided data to our National Medical Birth Registers.