Objective
We sought to identify predictors of preferences for cesarean among pregnant women, and estimate how different predictors influence preferences.
Study Design
This was a cross-sectional study based on the Norwegian Mother and Child Cohort Study (n = 58,881).
Results
Of the study population, 6% preferred cesarean over vaginal delivery. While 2.4% of nulliparous had a strong preference for cesarean, the proportion among multiparous was 5.1%. The probability that a woman, absent potential predictors, would have a cesarean preference was similar (<2%) for both nulliparous or multiparous. In the presence of concurrent predictors such as previous cesarean, negative delivery experience, and fear of birth, the predicted probability of a cesarean request ranged from 20–75%.
Conclusion
The proportion of women with a strong preference for cesarean was higher among multiparous than nulliparous women, but the difference was attributable to factors such as previous cesarean or fear of delivery and not to parity per se.
During the last decade there has been much interest in patient-demanded cesareans and the increasing cesarean delivery rates in industrialized countries. In the United States, approximately 5% of all deliveries were by cesarean in 1970, increasing to 20% in the 1980s and 32% in 2007. Although the United States has experienced a steeper increase over time, a similar pattern is seen in many European countries. In Norway, the cesarean delivery rate was approximately 2% in the 1970s, 12% in the 1980s, and 17% in 2008. This worldwide increase has caused great concern. One explanation that has been advanced for the increase is cesarean as the delivery option of choice. Increased attention to patient autonomy and shared decision making means that women who express a preference for cesarean delivery might obtain a surgical rather than vaginal delivery on the basis of a “weak” or even lacking medical indication. A consensus conference estimated that 4-18% of all cesareans were performed on maternal request. The reasons some women prefer cesarean section are therefore of interest to clinicians as well as policy makers.
Preferences for cesarean are often associated with factors such as anxiety and fear of birth. Previous cesarean delivery, previous negative birth experiences, maternal age, and socioeconomic factors are among other determinants. The effect of increased parity on delivery preferences has also been discussed, and previous studies have indicated that multiparous women more often prefer cesarean than nulliparous ones.
In this study we explore determinants of a cesarean preference in a large study sample, and predict the probability that different groups of pregnant women would prefer cesarean.
Materials and Methods
We used data from the Norwegian Mother and Child Cohort Study (MoBa), a study conducted by the Norwegian Institute of Public Health, and data from the Medical Birth Registry of Norway (MBRN).
MoBa is a cohort consisting of >100,000 pregnancies recruited into the study from 1999 through 2008. The target population was all women who gave birth in Norway and no exclusion criteria were applied. In total 50 of 52 maternity units participated. The total participation rate was approximately 44% of all the invited pregnancies. Women were recruited to the study through a postal invitation in connection with the routine ultrasound examination offered in Norway to all pregnant women at 17-19 weeks of gestation. Informed written consent was obtained from each participant. The study was approved by the Regional Committee for Medical Research Ethics and the Norwegian Data Inspectorate. The MoBa encompasses information on socioeconomic factors, physical and mental health, medication, and a variety of environment exposures and lifestyle habits before and during pregnancy.
Only women who responded to a question about their preferred choice of delivery method were included in the present study. From the MoBa data file (version IV) of 77,015 possible respondents, 1971 were excluded due to missing information on parity or delivery preference, 140 due to placenta previa, and 7330 were excluded due to neutral delivery preference. To ascertain independent observations, only data from the first time a woman participated in the MoBa were included and 8693 of repeat pregnancies were excluded. The final study sample encompassed 58,881 unique women, 29,373 nulliparous, and 29,508 multiparous.
Based on knowledge from previous research, we defined a priori a model with explanatory variables including socioeconomic information, obstetric and medical factors, emotional factors, and provider characteristics to study how these factors impact on the preference for cesarean ( Figure 1 ). The outcome variable, delivery preference, was measured in week 30 of pregnancy by response to the statement: “If I could choose, I would prefer to have a cesarean” captured on a 6-point response scale (“agree completely,” “agree,” “agree somewhat,” “disagree somewhat,” “disagree,” and “disagree completely”). As there were no neutral categories, the 2 middle response groups (“agree somewhat” and “disagree somewhat”) were excluded from the study analyses (n = 7330) because the direction of these preferences is unclear with respect to cesarean delivery on maternal request.
Emotional variables, captured in week 30, were measured by the respondents’ agreement with the following statements: “I am really dreading giving birth” (fear of birth), “I worry all the time that the baby will not be healthy or normal” and “On the whole, I am satisfied with the way I have been followed up by the health service.” Responses were captured on a 6-point response scale. If multiparous, previous delivery experience (“If you have given birth before, in general, how was the experience of giving birth?”) was scored on a 5-point scale, with answer categories of “very good,” “good,” “alright,” “bad,” or “very bad.” Information about a previous loss of a child (“Have you ever lost a child?” [“yes” or “no”]) was also elicited. Socioeconomic background characteristics were extracted from MoBa (survey at 17 weeks) and/or MBRN and included: age (<35 years, ≥35 years), marital status (married/cohabitant vs no partner), education (5 groups), work status (working, student/apprentice, not working), smoking habits, income (grouped), and county. Preexisting maternal comorbidity included diabetes (preexisting or gestational), chronic diseases (including hypertension, cardiac or renal disease, rheumatoid arthritis, or epilepsy), and anxiety/depression. These background variables were extracted from MBRN or MoBa survey at 17 weeks. We limited obstetric risk factors to those most likely to occur and with which women were likely to be familiar before stating delivery preferences in week 30 of pregnancy. The obstetric risk factors were bleeding before week 28, pregnancy due to in vitro fertilization, or multiple pregnancy, all variables extracted from the MBRN. Provider characteristics, extracted from MoBa survey at 30 weeks, included type of antenatal care [“hospital (outpatients) clinic” vs “other” (eg, captures public health center/midwife care, or general physician)]. If applicable, the sex of the obstetricians was also registered. We adjusted for the annual cesarean section rate at the hospital where the mother was to deliver (MBRN).
Based on a set of personal, medical, and emotional factors we defined a “reference woman,”–ie, a subset of women with expected low risk of interventions and adverse outcome–inspired by the standard primipara method originally suggested by Paterson et al. The reference nulliparous woman is without any of the usual risk factors associated with a preference for cesarean. She is <35 years of age, married or cohabiting, with a median level education and no comorbidities. She has an average level of emotional stress (including low levels of fear of birth and low worries about not having a healthy child).
The reference multiparous resembles the nulliparous in the relevant factors, and she also has a good previous delivery experience with no prior cesarean.
From the regression coefficients we estimated the predicted probabilities that the reference woman would have a preference for cesarean. Subsequently, we estimated the probabilities for women with different combinations of risk factors.
We used SPSS software (SPSS Inc, Chicago, IL) for descriptive statistics and χ 2 tests and STATA (Stata Corp, College Station, TX) to conduct logistic regression analysis (Logit) to determine the predictors of preference for cesarean. The dependent variable were dichotomized such that response categories “agree completely” and “agree” were classified as a cesarean preference, whereas “disagree completely” and “disagree” were classified as vaginal preference. Because parity has been identified as an important factor associated with many aspects of pregnancy and delivery, nulliparous (para 0) and multiparous (para 1+) women were analyzed separately. A P value < .05 was considered statistically significant. The goodness of fit was estimated by means of log likelihood ratio test (McFadden). Observations with missing values for any of the variables were excluded from the analyses (n = 2557 for nulliparous and n = 5984 for multiparous).
Results
The mean age of the study sample was 28 years for nulliparous and 32 years for multiparous women. There was a higher proportion age <35 years among nulliparous compared to multiparous (92% vs 76%). The majority in both parity groups was married or cohabitant. In both groups, nearly 40% had completed up to 4 years of university, while 26% of nulliparous and 21% of multiparous had ≥4 years of university. In all, 85% of the respondents were employed. Approximately 2% of the total population was pregnant with >1 fetus. There was a similar percentage of comorbidity in both parity groups, while the percentage of in vitro fertilization was 3% among nulliparous compared to 1% among multiparous ( Table 1 ). Among multiparous respondents 12% had a previous cesarean. The majority of the respondents were satisfied with their follow-up during pregnancy and with their previous delivery experience ( Table 2 ).
| Characteristic | Para 0 (n = 29,373) | Para 1+ (n = 29,508) | ||
|---|---|---|---|---|
| n | (%) | n | (%) | |
| Age, y | ||||
| <35 | 27,057 | (92.1) | 22,291 | (75.5) |
| ≥35 | 2316 | (7.9) | 7217 | (24.5) |
| Marital status | ||||
| Married/cohabitant | 28,100 | (95.7) | 28,908 | (98.3) |
| Single | 1250 | (4.3) | 498 | (1.7) |
| Native language | ||||
| Norwegian (including Sami) | 27,674 | (94.2) | 28,147 | (95.4) |
| Urdu | 35 | (0.1) | 35 | (0.1) |
| English | 158 | (0.5) | 163 | (0.6) |
| Other | 1506 | (5.1) | 1163 | (4.0) |
| Education | ||||
| Compulsory (secondary) school | 461 | (1.6) | 760 | (2.6) |
| High school | 8835 | (30.4) | 10,419 | (35.9) |
| Higher education <4 y | 11,611 | (39.9) | 11,259 | (38.8) |
| Higher education >4 y | 7535 | (25.9) | 5989 | (20.6) |
| Other | 631 | (2.2) | 623 | (2.1) |
| Work status | ||||
| Student/apprentice | 3092 | (10.7) | 1381 | (4.8) |
| Working | 24,770 | (85.6) | 24,478 | (84.8) |
| Not working | 1083 | (3.7) | 3016 | (10.4) |
| Previous cesarean | NA | NA | 3522 | (11.9) |
| Plurality–twins | 481 | (1.6) | 429 | (1.5) |
| Pregnancy due to in vitro fertilization | 974 | (3.3) | 374 | (1.3) |
| Vaginal bleeding up to wk 28 | 1242 | (4.2) | 1237 | (4.2) |
| Diabetes mellitus | ||||
| No | 28,989 | (98.8) | 29,107 | (98.7) |
| Preexisting diabetes | 153 | (0.5) | 154 | (0.5) |
| Gestational diabetes | 208 | (0.7) | 240 | (0.8) |
| Preexisting anxiety/depression | 2582 | (8.8) | 2304 | (7.8) |
| Preexisting chronic diseases a | 737 | (2.5) | 768 | (2.6) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
