The past decade has seen a continued rise in the cesarean delivery rate, especially in the developed world. In the United Kingdom, the cesarean delivery rate was reported at 22% for the period 2000–2001.1 In the United States, 23% of births were by cesarean delivery in 2003.2 Similarly, in Australia 28% of babies were born by cesarean delivery in 20033 compared with 19.4% in 1994. Some of this increase may be attributable to the improved safety of abdominal delivery with regional anesthetic, routine administration of antibiotics, and reliable blood-product supply. In the face of a safe and accessible alternative, the fetal risks associated with more hazardous vaginal deliveries have become less acceptable, resulting in a reduction in vaginal breech delivery,4 midcavity instrumental delivery,3 vaginal twin delivery, trials of labor, and a reduced tolerance of suspected fetal compromise in labor. The identified maternal hazards of an emergency, rather than planned elective, cesarean delivery (Yoles I, Maschiach S. Increased maternal mortality in cesarean section as compared to vaginal delivery? Time for re-evaluation [abstract]. Am J Obstet Gynecol 1998;178 suppl:S78) have influenced decisions on a trial of labor where the risk of recourse to emergency cesarean delivery is felt to be high. The potential pelvic floor damage associated with labor and birth,5 and its impact on urinary6,7 and fecal incontinence8 has been more recently highlighted although there remains conjecture as to how much protection is conferred by cesarean delivery. Increasing maternal age and decreasing family size have made the potential fetal9 and maternal risks10,11 of repeat cesarean delivery less relevant to some patients. Finally, respect for patient autonomy regarding mode of birth has also played a substantial role.12
Despite these considerations, the rate of intervention is widely regarded as too high. The World Health Organization has recommended a cesarean delivery rate of no greater than 15% in the developed world (WHO Consensus Conference on Appropriate Technology for Birth, Fortaleza, Brazil, April 22–26, 1985), despite a paucity of scientific data validating this figure. An attempted vaginal birth involves a small fetal risk of death or serious morbidity that is almost completely avoided with cesarean delivery. This includes the risks of stillbirth in late pregnancy, intrapartum asphyxia resulting in stillbirth or neonatal encephalopathy, intrapartum trauma, chorioamnionitis, and meconium aspiration. Consistent with this, a study examining hospital stillbirth rates among 65 maternity units in all Thames Regions observed that birth weight–adjusted stillbirth rates were negatively correlated with rates of obstetric intervention.13 Likewise, after comparing the relationship between cesarean delivery rates and birth weight–specific perinatal mortality in the three largest maternity hospitals in Dublin, Matthews observed “…there was no cesarean delivery rate above which no further falls (in perinatal mortality) were seen.”14 It is therefore relevant to consider what level of fetal risk—of death or serious disability—patients and clinicians consider acceptable to avoid a cesarean delivery and deliver vaginally. This issue was examined in a study of 107 pregnant and nonpregnant women in 1989 where the median level of indifference between vaginal birth and cesarean delivery occurred at a fetal risk of stillbirth of 2.5 per 10,000.15 Pregnant women were less tolerant of risk than the group overall (median level of indifference, 0.5 per 10,000) whereas those choosing home birth were more risk tolerant (median level of indifference, 100 per 10,000). The last two decades has seen a substantial change in obstetric practice and patient expectation, so we decided to re-examine the level of fetal risk that pregnant women and their caregivers now consider acceptable to achieve a vaginal birth.
MATERIALS AND METHODS
Six hundred primigravid and multigravid pregnant women were recruited, using convenience sampling, from the Mercy Hospital for Women during 2002–2004. Ethics approval for the study was obtained from the Research Ethics Committee of the Health Care Network, Mercy Health and Aged Care.
The Mercy Hospital for Women is a tertiary teaching obstetric facility in Melbourne, Australia, delivering approximately 5,500 women per annum. Models of care offered to women during their pregnancy include traditional care in a combined obstetric and midwifery antenatal clinic, shared care (between general practitioner and antenatal clinic), team midwifery and the Family Birthing Centre.
Women were recruited after 36 weeks of gestation, the study explained, and written information provided. This information included a paragraph about risks in everyday life and specific fetal risks associated with their pregnancy that had already passed, including risk of miscarriage (25%), risk of preterm birth (7%), and risk of major abnormality being discovered at the time of their second trimester ultrasound examination (2%). Although the focus of this study was on fetal risk, it was acknowledged that other risks (predominantly maternal) of cesarean delivery would need to be considered. Women were informed that elective cesarean delivery was associated with more abdominal pain, longer recovery, increased risk of infection, increased risk of repeat cesarean birth, and the risk that it was “less satisfying” for many women when compared with vaginal birth.
To more effectively communicate the risks involved, a visual aid for probabilities was created. This was a folio consisting of 11 A1 posters, each with 10,000 faces represented. Yellow faces represented “healthy births” and red faces represented “poor outcome.” To orient the patients to the concept, the first three charts visually represented the risk of the previously described obstetric outcomes: miscarriage (2,500 red faces per 10,000 faces), preterm birth (700 red faces per 10,000 faces), and major abnormality on midtrimester ultrasound examination (200 red faces per 10,000 faces). The fourth poster had just three of the 10,000 faces shaded purple (Fig. 1). It was explained to patients that this represented the unavoidable small risk of stillbirth at term (3 per 10,000). Patients were then shown a series of seven further posters, with an additional 1 per 10,000, 3 per 10,000, 6 per 10,000, 10 per 10,000, 20 per 10,000, 100 per 10,000, and 200 per 10,000 faces shaded red. These were shown to the patients in a consistent order: the lowest risk (1 per 10,000) followed by the highest risk (200 per 10,000), then the second lowest risk followed by the second highest risk and so on through the series of posters until they reached equipoise, and were able to respond to the statement “if the increased risk of serious harm to the baby of vaginal delivery was 1 in…, I would prefer a cesarean birth to a vaginal delivery.” It was explained to patients that “serious harm” would include either death or serious (lifelong) disability. Participants could also generate their own figure, whether below 1 per 10,000, above 200 per 10,000, or between two risk levels depicted by the posters. Demographic information, such as age, parity, occupation, level of education, complications in this or a previous pregnancy, chosen model of care for this pregnancy, and intended mode of delivery was recorded. The information was de-identified and stored, along with each patient’s stated level of risk.
Consultant obstetricians (n=16), obstetricians in training (n=16), hospital midwives (n=179), and a group of final year medical students (n=83) were similarly recruited to the study. The introduction to the study was similar, although the risks of cesarean delivery, in particular the risks of abnormal placentation and accreta in subsequent pregnancies, were highlighted. Using the same visual aid for probabilities, participants were asked to respond to the statement “if the level of fetal risk associated with vaginal delivery was greater than 1 in…, I would recommend a cesarean birth.” They were then asked to consider the following two scenarios: 1) a 41-year-old primigravida with 3 years of primary infertility and in vitro fertilization pregnancy, 2) a 24-year-old planning a large family.
The participants were asked whether, under the above circumstances, they would change their previously stated acceptable level of fetal risk before recommending a cesarean birth.
The core hypothesis was that a clear majority of patients would perceive the level of acceptable risk to be less than 10 per 10,000. A 10% difference from equipoise was considered to be clinically important, that is, 11 per 10,000. We considered a liberal estimate of the standard deviation to also be 10 per 10,000. Using a one-sample t test, 600 women would be sufficient with 80% power and α=0.05.
The risk thresholds were summarized, and demographic and obstetric variables examined as modifiers of patients’ acceptable risk thresholds. Because the data were not normally distributed (on qualitative assessment by the statistician), a binary outcome was created for the purposes of multivariable analysis (less than 20 per 10,000 births compared with 20 or more per 10,000 births). This threshold was chosen because it was where the data were approximately split (thereby having sufficient numbers above and below this threshold for meaningful analysis) but was also considered to be an important clinical threshold. The obstetric and demographic predictors were examined using univariable and multivariable logistic regression. Statistical analysis was performed with the statistical package Stata 6.0 (StataCorp, College Station, TX).
The clinical characteristics of the patients are shown in Table 1. In response to the statement “if the increased risk of serious harm to the baby of vaginal delivery was 1 in…, I would prefer a cesarean birth to a vaginal delivery,” the median level of acceptable risk was 10 per 1,000 births (95% confidence interval [CI] 10–13 per 10,000). The patient responses to each of the risks portrayed by the visual probability aid are summarized, alongside the staff responses, in Table 2.
The demographic and obstetric variables examined for their impact on risk tolerance are shown in Table 3, using a threshold of less than 20 per 10,000 compared with 20 or more per 10,000 births for the comparison. Participation in a lower intervention model of care (Family Birthing Centre and team midwifery) was significantly associated with choosing a higher level of acceptable fetal risk to achieve a vaginal birth, when compared with conventional antenatal care or shared care with a general practitioner. Women with a complicated pregnancy were less tolerant of fetal risk, as were those already planning to deliver by cesarean birth.
In response to the statement “if the level of fetal risk associated with vaginal birth was greater than 1 in…, I would recommend a cesarean birth,” the median level of acceptable risk for staff overall was 10 per 10,000 births (95% CI 10–20 per 10,000). The risk thresholds for the four occupational groups did not significantly differ (data not shown).
In response to the scenario “In a 41-year-old primigravida with 3 years of infertility and an in vitro fertilization pregnancy…,” 67% (95% CI 61–72%) reported that they would accept a lower level of fetal risk than previously stated before recommending a cesarean birth, whereas 28% (95% CI 23–33%) reported their prior level of acceptable risk would be unchanged. In response to the scenario “In a 24-year-old planning a large family…,” 42% (95% CI 36–48%) reported that they would accept a higher level of fetal risk than previously stated before recommending a cesarean birth, whereas 53% (95% CI 47–59%) reported that their prior level of acceptable risk would be unchanged.
Despite widely publicized concerns about intervention rates in obstetrics, these findings suggest that patients and their caregivers have a low tolerance for fetal risk associated with vaginal birth. Many patients would not wish to expose their fetuses to risks that continue to be part of contemporary obstetric practice. Risks of death or serious harm cited with “high risk” vaginal deliveries, such as twins and vaginal birth after cesarean birth, are approximately 30 per 10,00017 and 5 per 10,000,18,19 respectively. Risk estimates derived from retrospective cohorts are, of course, imprecise, and individual risk will be determined by clinicians, patients, and the individual place and circumstance of birth.20 Nevertheless, these “best estimates” of fetal risk are higher than the majority of patients in this study would consider acceptable.
Although patients reported a wide variation in their level of “acceptable fetal risk,” several predictors of risk threshold were identified. Women in “lower intervention” models of care were more tolerant of fetal risk; this finding parallels the original findings of Thornton and Lilford.15 This group of women may regard the concept of medical risk with some suspicion. In her essay arguing that the role of risk has become a “defining concept in medicalized child birth,” Bernice Hausman21 contends that the “notion of risk [is] used as a threat to compel compliance with medicalized norms…,” ignoring the “core feminist belief…that pregnancy is a normal physiological process…and thus not, in most circumstances, deserving of intense medical scrutiny.” Such views are more likely to shift respondents toward minimizing intervention rather than minimizing risk. In contrast, women in a more medicalized model of care, those in whom complications had developed or those for whom cesarean delivery was already intended were less tolerant of fetal risk associated with vaginal birth. These women may have lowered their expectations for vaginal birth because of a heightened awareness of the potential risks faced in their pregnancy.
Although potentially compromised by a smaller sample size, this study also found obstetric caregivers, across all disciplines, to be very “risk averse.” Most staff members were able to identify their level of acceptable “fetal risk.” Two thirds, however, were prepared to lower this threshold for a perceived “precious baby,” (in the setting of infertility or advanced maternal age), whereas half were prepared to raise it where the consequences of scarring the uterus (in a young woman contemplating a large family) were considered unacceptable.
Communicating numbers meaningfully is difficult. That risks were communicated effectively is crucial to the validity of these findings. Mindful of this, the following were taken into account during study design. The use of different denominators is known to generate confusion as to which is the higher risk,22 and so all risks to patients were presented using a consistent denominator. Absolute numbers (1 per 10,000 versus 3 per 10,000) were used for comparisons rather than relative risks,23 and statements were framed both positively and negatively, ie, “the outcome will be positive 9,999 times per 10,000 births” in addition to “there is a risk of death or serious injury in 1 per 10,000 births.” We used a visual aid for probabilities,24 accepting that literacy and numeracy skills may be variable. In providing information about cesarean birth, we highlighted that obstetric intervention is a balance of risks and benefits.24 Women were informed about immediate and future risks associated with elective caesarean birth, and the specific risks of abnormal placentation, particularly placenta accreta, in subsequent pregnancies were highlighted to staff participants. We also informed patients that there was an unavoidable risk of stillbirth, irrespective of the mode of birth because we felt that if patients perceived that the “only fetal risk,” as opposed to the “added fetal risk,” was associated with vaginal birth, it might excessively bias them toward cesarean birth. In choosing the risk of stillbirth that was “unavoidable,” we used the background risk of stillbirth/week between 38 and 41 weeks of gestation. We reasoned that, for a patient at 38 weeks of gestation weighing the options of vaginal birth versus cesarean birth, the risk of stillbirth before her elective cesarean birth the following week was 3 per 10,000.25 As such, this was the “lowest risk possible” and one to which she could not help but expose her baby. In this study, patients were being asked specifically to consider what “added risk,” over and above this background risk, was acceptable to achieve vaginal birth.
This study has limitations. Despite best efforts to communicate numbers meaningfully, it is more difficult to control for emotional responses to fetal risk. Some women may have found it confronting to accept that the fetus was exposed to any risk at all. Psychological and social factors are known to significantly influence the way in which patients respond to risk information and may be relevant in this study. People are more sensitive to risks of low probability and high consequence, for example, death from a plane crash, compared with the much greater risk of death associated with obesity or smoking.26 This is particularly so for risks involving children; parents are likely to be more sensitive to a tenuous link between childhood vaccination and autism than more common risks to child health such as road safety.27 It is possible that death or damage as a consequence of vaginal birth may be perceived similarly, resulting in heightened patient sensitivity to these risks. In addition, a clear majority of patients who agreed to participate in our study were tertiary educated, which is not usually representative of a public antenatal population. Such participants are more likely to be older and plan smaller families, perhaps making them more “risk averse.” Caution is therefore necessary before generalizing these findings to other obstetric settings and populations.
Nevertheless, this study has demonstrated that most patients have a low tolerance for fetal risk to achieve vaginal birth. It further highlights that “one figure” does not “fit-all,” given the complexity of informed decision making in obstetrics, where multiple factors contribute to both a patient’s and clinician’s view of what constitutes an acceptable risk. This study underscores the problem of high patient expectations in the emotive area of childbearing and fetal outcome and the need for continuing community education that the risks of childbirth can be minimized, but not eliminated.
1. Mayor S. Caesarean section rate in England reaches 22%. BMJ 2002;324:1118.
2. Centers for Disease Control and Prevention. Rates of cesarean delivery–United States, 1993. MMWR Morb Mortal Wkly Rep 1995;44:303–7.
3. Australian Institute of Health and Welfare. National Perinatal Statistics Unit. Australia’s mothers and babies, 2003. Available at: http://www.npsu.unsw.edu.au/Stats.htm
. Retrieved October 27, 2006.
4. Rietberg CC, Elferink-Stinkens PM, Visser GH. The effect of the Term Breech Trial on medical intervention behaviour and neonatal outcome in The Netherlands: an analysis of 35,453 term breech infants. BJOG 2005;112:205–9.
5. Dietz HP, Bennett MJ. The effect of childbirth on pelvic organ mobility. Obstet Gynecol 2003;102:223–8.
6. Rortveit G, Daltveit AK, Hannestad YS, Hunskaar S. Urinary incontinence after vaginal delivery or cesarean section. N Engl J Med 2003;348:900–7.
7. Glazener C, Herbison G, Macarthur C, Lancashire R, McGee M, Grant A, et al. New postnatal urinary incontinence: obstetric and other risk factors in primiparae. BJOG 2006;113:208–17.
8. Faridi A, Willis S, Schelzig P, Siggelkow W, Schumpelick V, Rath W. Anal sphincter injury during vaginal delivery: an argument for cesarean section on request? J Perinat Med 2002;30:379–87.
9. Smith GC, Pell JP, Dobbie R. Caesarean section and risk of unexplained stillbirth in subsequent pregnancy. Lancet 2003;362:1779–84.
10. Gilliam M, Rosenberg D, Davis F. The likelihood of placenta previa with greater number of cesarean deliveries and higher parity. Obstet Gynecol 2002;99:976–80.
11. Hemminki E, Merilainen J. Long-term effects of cesarean sections: ectopic pregnancies and placental problems. Am J Obstet Gynecol 1996;174:1569–74.
12. Minkoff H, Chervenak FA. Elective primary cesarean delivery. N Engl J Med 2003;348:946–50.
13. Joyce R, Webb R, Peacock JL. Associations between perinatal interventions and hospital stillbirth rates and neonatal mortality. Arch Dis Child Fetal Neonatal Ed 2004;89:F51–6.
14. Matthews TG, Crowley P, Chong A, McKenna P, McGarvey C, O’Regan M. Rising caesarean section rates: a cause for concern? BJOG 2003;110:346–9.
15. Thornton JG, Lilford RJ. The caesarean section decision: patients’ choices are not determined by immediate emotional reactions. J Obstet Gynaecol 1989;9:283–8.
16. Edwards A, Elwyn G, Mulley A. Explaining risks: turning numerical data into meaningful pictures. BMJ 2002;324:827–30.
17. Smith GC, Shah I, White IR, Pell JP, Dobbie R. Mode of delivery and the risk of delivery-related perinatal death among twins at term: a retrospective cohort study of 8073 births. BJOG 2005;112:1139–44.
18. Landon MB, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW, et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med 2004;351:2581–9.
19. Appleton B, Targett C, Rasmussen M, Readman E, Sale F, Permezel M. Vaginal birth after Caesarean section: an Australian multicentre study. VBAC Study Group. Aust N Z J Obstet Gynaecol 2000;40:87–91.
20. Smith GC, Pell JP, Pasupathy D, Dobbie R. Factors predisposing to perinatal death related to uterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ 2004;329:375–7.
21. Hausman BL. Risky business: framing childbirth in hospital settings. J Med Humanit 2005;26:23–38.
22. Grimes DA, Snively GR. Patients’ understanding of medical risks: implications for genetic counseling. Obstet Gynecol 1999;93:910–4.
23. Gigerenzer G, Edwards A. Simple tools for understanding risks: from innumeracy to insight. BMJ 2003;327:741–4.
24. Paling J. Strategies to help patients understand risks. BMJ 2003;327:745–8.
25. Hilder L, Costeloe K, Thilaganathan B. Prolonged pregnancy: evaluating gestation-specific risks of fetal and infant mortality. Br J Obstet Gynaecol 1998;105:169–73.
26. Alaszewski A, Horlick-Jones T. How can doctors communicate information about risk more effectively? BMJ 2003;327:728–31.
© 2007 The American College of Obstetricians and Gynecologists
27. Bellaby P. Communication and miscommunication of risk: understanding UK parents’ attitudes to combined MMR vaccination. BMJ 2003;327:725–8.