Few women wish to be delivered by caesarean section


*Dr I. Hildingsson, Gamla Karlebyvägen 22 E, S-87133 Härnösand, Sweden.


Objective To investigate how many women wish to have a caesarean section when asked in early pregnancy, and to identify background variables associated with such a wish.

Design National survey.

Setting Swedish antenatal clinics.

Population 3283 Swedish-speaking women booked for antenatal care, at approximately 600 Swedish antenatal clinics, during three weeks spread over one year (1999–2000).

Methods A questionnaire was mailed shortly after the first antenatal visit.

Main outcome measures Women's preferences for mode of delivery.

Results 3061 women completed the first questionnaire, corresponding to 94% of those who consented to participate after exclusion of reported miscarriages. The background characteristics of the study sample were very similar to a one-year cohort of women giving birth in Sweden during 1999. The result showed that 8.2% of the women would prefer to have a caesarean section. A wish for caesarean section was associated with parity, age, civil status, residential area and obstetric history. Women preferring caesarean section were more depressed and worried, not only about giving birth, but also about other things in life. A multivariate logistic regression model showed three factors being statistically associated with a wish for caesarean section: a previous caesarean section, fear of giving birth and a previous negative birth experience.

Conclusions Relatively few women wish to have a caesarean section when asked in early pregnancy, and these women seem to be a vulnerable group.


Increasing caesarean section rates are a matter of concern in many countries and possible explanations are being discussed1. In Sweden, with a low and stable rate around 10–12% for almost 20 years, the figure has continually risen during the last five years. The most recent figure (1999) available from the National Medical Birth Register was 14%, with several hospitals reaching 20%2,3. One of the reasons for the climbing rates has been attributed to the demand from women themselves. These women are often described as well educated, urban, in control, capable of making independent decisions and focussed on their professional careers.

Studies of women's preferences regarding mode of delivery are inconclusive. Two percent of Irish women wanted to have a caesarean section, all of whom had previously undergone such an operation3. Seven percent of British women4,5, and 4% of British midwives6 preferred a caesarean section. In contrast to these low figures, 28% of Australian women, with a previous caesarean section, had insisted on a caesarean section7, and 31% of female obstetricians in a Southern English city said they would prefer such a mode of delivery8.

The aim of the present study was to investigate how many women in Sweden wished to have a caesarean section when asked in early pregnancy, and to identify the characteristics of these women.


The study was conducted as a national survey of Swedish-speaking women who booked for antenatal care during three weeks evenly spread over one year. Antenatal care in Sweden is provided by midwives, in collaboration with medical staff when complications occur, in local antenatal clinics, and all pregnant women see a midwife at their first booking visit. The midwives were informed about the study via written information, and by an advertisement in the Swedish Journal of Midwifery. The midwives were asked to inform women about the study at their first antenatal visit in early pregnancy by handing out an information leaflet and asking them if they were interested in participating in the study. Women consented to participate by signing a form including their personal identity code and contact details. The recruitment took place in May and September 1999 and in January 2000. After each recruitment week, the list of names was sent to the research team and the questionnaire was then mailed to the women on the list. Two letters of reminder were sent to non-responders.

The questionnaire included questions about socio-demographic and obstetric background, attitudes to decision making, experience of previous birth and present pregnancy, the Cambridge Worry Scale measuring common concerns during pregnancy9 and the Edinburgh Postnatal Depression Scale (EPDS)10.

The question about mode of delivery was worded “If you have the possibility to choose, how would you like to give birth?” The response alternatives were: ‘Vaginal birth’ and ‘Caesarean section’.

Background characteristics of the sample were compared with data from a one-year cohort of women giving birth in Sweden in 1999, extracted from the Swedish Medical Birth Register, which includes information on socio-demographic variables, care procedures and health outcomes.

Statistical analyses were conducted using SPSS for Windows. The Cambridge Worry Scale, with responses expressed on a six-point scale, was dichotomised into ‘no worries’ (0–3) and ‘major worries’ (4–5). The cutoff point for the EPDS was set at >14, as suggested when used during pregnancy11. Differences between women who preferred a caesarean and a vaginal birth, respectively, were calculated by χ2 test and risk ratios (RR) with 95% confidence intervals were estimated (Mantel–Haenszel's method). The association between various demographic and obstetric factors and women's preferences regarding mode of delivery was tested by means of binary logistic regression. Before being entered into the logistic regression model, questions about previous birth experience and feelings about being pregnant were dichotomised into two categories (1 = very negative + rather negative + mixed feelings; 2 = very positive + rather positive).

The study was approved by the Regional Research and Ethical Committee at Karolinska Institutet, Sweden (Dnr 98-358).


All the 608 antenatal clinics in Sweden were approached and 593 chose to participate in the study. Seven antenatal clinics in a region consisting of both a middle sized city and a rural area withdrew because of other ongoing studies. The remaining eight non-participating clinics reported heavy workload. The total number of women booked for antenatal care during the three recruitment weeks was approximately 5400. This figure was an estimation based on data from the antenatal care midwives and from the National Birth Register. Approximately 550 women were not invited because of language difficulties, 172 had an early miscarriage and 75 were booked at the non-participation clinics. Of the remaining 4603 women eligible for the study, 3283 (71%) were approached and consented to participate and 3061(94%) of these women completed the first questionnaire. The women were on average in gestational week 15.6 (SD 3.4, range 7–40 weeks) when responding to the questionnaire.

The background characteristics of the study sample were very similar to those of a one-year cohort of women giving birth in Sweden in 1999, as demonstrated in Table 1. The major difference was country of birth, which was expected considering the inclusion criteria of the study.

Table 1.  Background characteristics.
 Study population (n= 3061*), n (%)A cohort of all women who gave birth in Sweden in 1999 (n= 83,059*), n (%)
  1. *Women with missing information regarding a specific variable were excluded.

  2. **Data not available from the National Birth Register.

Primiparas1302 (43)36,505 (44)
Multiparas1759 (57)46,554 (56)
Number of children
<21590 (91)40,493 (87)
>2154 (9)5174 (11)
Mean age29.429.5
Age groups
<25 years626 (20)13,484 (16)
25–35 years2119 (69)59,728 (72)
>35316 (10)9753 (12)
Civil status
Married/cohabiting2888 (95)75,583 (95)
Other family situation119 (4)2725 (3)
Single40 (1)1445 (2)
Country of birth
Sweden2760 (90)67,558 (83)
Other Nordic countries70 (2)1894 (2)
Other European countries103 (3)3431 (4)
Other countries127 (4)8847 (11)
Elementary school (years 1–9)224 (7) 
High school1658 (55) 
College or university1154 (38) 
Residential area**
Large city828 (27) 
Middle sized city579 (19) 
Small city654 (22) 
Rural area964 (32) 
Smoking333 (11)9979 (13)
Obstetric history
Previous miscarriage663 (22)16,558 (20)
Previous abortion**651 (21) 
Infertility >1 year**320 (10) 
Assisted conception68 (2)992 (1)
Previous stillbirth31 (1)715 (1)
Present pregnancy**
Planned2181 (71) 
Unplanned but welcome688 (22) 
Timing could have been better125 (4) 
Abortion considered56 (2) 

Of the 3013 women who answered the question about preferred mode of delivery, 8.2% preferred to have a caesarean section. Of the socio-demographic and obstetric variables listed in Table 2, the following were associated with a wish for caesarean section: age >35 years, not being married or cohabiting, living in a small city, smoking, previous abortion, previous elective and emergency caesarean section. A previous stillbirth was associated with a doubled risk of wanting a caesarean section, but this increase was not statistically significant.

Table 2.  Preference regarding mode of delivery in relation to socio-demographic and obstetric background.
 Caesarean section (246), n (%)Vaginal birth (2767), n (%)RR95% CI
  1. *Reference (1.0): Women not exposed for the variable studied.

Primiparas92 (7.2)1192 (92.8)1.0(ref)
Multiparas154 (8.9)1575 (91.1)1.21.0–1.6
Number of children
<2140 (8.9)1427 (91.1)1.0(ref)
>211 (7.4)138 (92.6)0.80.5–1.6
Age groups
<25 years48 (7.8)571 (92.2)1.00.7–1.4
25–35 years163 (7.8)1925 (92.2)1.0(ref)
>35 Years35 (11.4)271 (88.6)1.51.0–2.1
Civil status
Married or cohabiting225 (7.9)2621 (92.1)1.0(ref)
Other family situation19 (12.4)134 (87.6)1.61.0–2.4
Country of birth
Sweden220 (8.1)2502 (91.9)1.0(ref)
Other countries26 (9.0)264 (91.0)1.10.8–1.6
Elementary school (years 1–9)22 (10.2)193 (89.8)1.20.8–1.8
High school142 (8.7)1495 (91.3)1.0(ref)
College or University80 (7.0)1058 (93.0)0.80.6–1.1
Residential area
Large city72 (8.8)745 (91.2)1.51.0–2.2
Middle sized city34 (6.0)533 (94.0)1.0(ref)
Small city61 (9.4)586 (90.6)1.61.0–2.4
Rural area78 (8.2)870 (91.8)1.40.9–2.0
Smoking habits
Not smoking202 (7.6)2461 (92.4)1.0(ref)
Smoking43 (13.1)286 (86.9)1.71.3–2.3
Obstetric history*
Previous miscarriage57 (8.7)595 (91.3)1.10.8–1.4
Previous abortion66 (10.2)578 (89.8)1.31.0–1.8
Infertility >1 year26 (8.3)286 (91.7)1.00.7–1.5
Assisted conception4 (6.2)61 (93.8)0.80.3–1.9
Previous stillbirth5 (17.2)24 (82.8)2.10.9–4.7
Previous mode of delivery
Vaginal delivery80 (5.3)1443 (94.7)1.0(ref)
Elective caesarean section41 (49.4)42 (50.6)9.46.9–12.8
Emergency caesarean section61 (32.4)127 (67.6)6.24.6–8.3

Of the attitudinal and psychological variables listed in Table 3, the following were associated with a wish for a caesarean section: unfortunate timing of pregnancy or abortion considered, negative or mixed feelings about this pregnancy, negative or mixed feeling about previous birth, major worries during pregnancy, particularly about the approaching birth, the relationship with family and friends, the ability to care for the newborn baby and antenatal depression. Women who preferred a caesarean section were also more worried about internal examinations, going to the hospital, something being wrong with the baby and about the risk of having a miscarriage. Making their own decisions, after having received information, was not a characteristic specific for women preferring a caesarean section.

Table 3.  Preference regarding mode of delivery in relation to attitudes and psychosocial variables.
 Caesarean section (246), n (%)Vaginal birth (2767), n (%)RR95% CI
  1. *Reference (1.0): Women not exposed for the variable studied.

Attitudes towards decision making
Would like to decide myself after information77 (7.1)1010 (92.9)0.80.6–1.0
Would like the doctor or midwife to decide after information and consultation with me19 (7.3)240 (92.7)0.80.5–1.3
Would like joint decision between doctor or midwife and myself148 (9.0)1504 (91.0)1.0(ref)
Present pregnancy
Planned156 (7.3)1992 (92.7)1.0(ref)
Unplanned but welcome62 (9.1)616 (90.9)1.31.0–1.7
Timing could have been better16 (12.9)108 (87.1)1.81.1–2.9
Abortion considered10 (19.2)42 (80.8)2.61.5–4.7
Experience of present pregnancy
Very positive152 (6.5)2174 (93.5)1.0(ref)
Rather positive47 (11.6)359 (88.4)1.81.3–2.4
Mixed feelings42 (16.2)217 (83.8)2.51.8–3.4
Rather negative3 (33.3)6 (66.7)5.12.0–13.0
Previous birth experience
Very positive24 (4.0)575 (96.0)1.0(ref)
Rather positive22 (5.2)400 (94.8)1.30.7–2.3
Mixed feelings38 (8.9)387 (91.1)2.21.4–3.7
Rather negative23 (14.8)132 (85.2)3.72.1–6.4
Very negative43 (38.0)70 (62.0)9.56.0–15.0
Major worries*
Housing18 (9.3)175 (90.7)1.10.7–1.8
Money problems38 (12.5)266 (87.5)1.61.2–2.2
Problems with law1 (4.8)20 (95.2)0.60.1–3.9
Relationship with partner13 (12.4)92 (87.6)1.50.9–2.6
Relationship with family and friends15 (22.1)53 (77.9)2.81.8–4.4
Own health23 (12.4)162 (87.6)1.61.1–2.3
The health of someone close34 (9.3)333 (90.7)1.10.8–1.6
Employment problems30 (8.8)310 (91.2)1.10.7–1.6
Something being wrong with the baby91 (12.1)661 (87.9)1.81.4–2.2
Going to hospital25 (14.3)150 (85.7)1.81.2–2.7
Internal examination16 (15.2)89 (84.8)1.91.2–3.1
Giving birth138 (22.7)469 (77.3)5.04.0–6.4
Coping with the newborn baby21 (17.8)97 (82.2)2.31.5–3.4
Giving up work9 (11.8)67 (88.2)1.50.8–2.6
Partners presence at birth8 (12.5)56 (87.5)1.50.9–3.0
Risk of msicarriage74 (11.0)600 (89.0)1.51.1–1.9
EPDS >14*35 (15.2)195 (84.8)2.01.4–2.8

In a multivariate logistic regression model, all variables that were statistically significant in the univariate analysis were included, and Table 4 shows that only three then contributed to the explanation of preferred mode of delivery. In multiparous women (n= 1707), a wish to be delivered by caesarean section was associated with a previous caesarean delivery, particularly an elective, fear of giving birth and a negative experience of the previous birth. When primiparas (n= 1302) were analysed separately, only fear of childbirth (major worry about giving birth: OR 5.4; 3.5–8.3, P < 0.001) remained as a statistically significant predictor for wishing a caesarean delivery.

Table 4.  Factors associated with a wish for a caesarean delivery analysed by logistic regression. n= 1707.
 POdds Ratio95% CI
Previous elective caesarean section<0.00119.311.0–33.7
Major worries about giving birth<0.0015.43.3–7.7
Previous emergency caesarean section<0.0015.23.3–8.1
Previous negative birth experience<–5.0


This study of a national sample of Swedish-speaking women showed that 8.2% would prefer to be delivered by caesarean section, when asked in early pregnancy. This figure is probably very close to the national figure, considering the high response rate to the questionnaire, and the similarities in background characteristics between the study sample and the 1999 birth cohort. Smokers were slightly under-represented in the sample. Since smoking was associated with a wish for caesarean section, we standardised for smoking habits (based on data from the Swedish Medical Birth Register), the estimated proportion of women wishing a caesarean section then increased to 8.3%.

The relatively small proportion of women who would prefer a caesarean delivery when asked in early pregnancy, and the finding that a previous caesarean section was the strongest predictor of preferred mode of delivery, suggests that women's own wishes are of limited help in explaining the rising caesarean section rates. Similarly to our results, Gamble and Creedy12 reported that few women requested a caesarean section in the absence of current or previous obstetric complications. They believed that a focus on women's requests might divert attention away from physician-led influences on the continuing high caesarean rates. Leitch and Walker13 showed that indications for caesarean sections had not changed, but the thresholds for applying them were lower than before, thereby suggesting that patient demand, had little explanatory value. Murray14 showed that the higher caesarean section rates in private patients compared with public patients could not be explained by patient demand.

A previous caesarean section, major worries about giving birth, and in multiparous women, a previous negative birth experience were the most important factors contributing to a wish for a caesarean delivery in this study. Consequently, the main reason for a woman's request for a caesarean section on non-medical grounds was fear of childbirth, a finding that is supported by other studies15–17.

We do not know if women are more worried about giving birth today, or if they express their concerns more openly, or if fear of childbirth contributes in any way to the increasing caesarean section rates. However, this study showed that a wish for caesarean section was not associated with well educated urban women, keen on making their own decisions, scheduling childbirth into their well controlled agendas and career planning. Contrary to this stereotype, women who wished a caesarean section were more often older than 35 years, single, smokers, tended to be less well educated, lived in small cities, had more often considered abortion, were more depressed and worried during pregnancy, not only about the birth but about many other things as well, and they were not more interested in making their own decisions than women who preferred a normal delivery. Psychological support during pregnancy may offer an alternative to a caesarean section for women who fear a vaginal birth, and/or have a previous negative birth experience.


We thank all the women who participated in the survey, and all the midwives who helped us with the information about the study and the recruitment of the participants.

This study was supported by grants from the County Council of Västernorrland; Mälardalens University; Vårdalstiftelsen and Karolinska Institutet, Stockholm.