International concerns about rising rates of caesarean section are counterbalanced by arguments that planned caesarean section without specific clinical indication (such as breech presentation or HIV infection) falls within legitimate maternal choice.1 Professional opinion is divided. To perform a caesarean section without clinical reason is seen as ethical, in response to maternal request, by the American College of Obstetricians and Gynecologists; is enshrined in law in Italy; but is viewed as unethical by the International Federation of Gynecology and Obstetrics. The National Institute for Health and Clinical Excellence recommends that a second opinion should be offered.2 A well designed, randomised controlled trial of planned caesarean section compared with planned vaginal birth could provide important evidence.3
Participants, methods, and results
We aimed to survey all consultant obstetricians and heads of midwifery (gatekeepers to such a trial) practising in England between January 2003 and May 2004. We explored their views of women's requests for caesarean section without clinical indication and of a possible randomised controlled trial in a postal survey. We used semistructured questionnaires with closed questions followed by free text spaces to provide supporting rationale. Comparisons were made between professionals and according to parental status, sex, and type of unit where they worked. We used χ2 tests to compare the proportion of respondents saying “yes” to each question. Two of the authors (TL, CK) manually analysed qualitative responses.
Table 1.
Findings of survey. Values are numbers (with rounded percentages) of participants unless otherwise indicated
|
|
|
Consultant obstetricians
|
|
|
|||
|---|---|---|---|---|---|---|---|
| Survey question | Answer | All (n=660) | Male (n=468) | Female (n=188)* | χ2 test comparing sexes† | Heads of midwifery (n=123) | χ2 test comparing professional groups‡ |
| Do you believe that primigravid women (in the absence of clinical indications) should choose their method of delivery? | Yes
|
321 (49)
|
254 (54)
|
67 (36)
|
P<0.001 | 33 (27)
|
P<0.001 |
| No
|
325 (49)
|
207 (44)
|
117 (62)
|
85 (69)
|
|||
| Missing | 14 (2) | 7 (1) | 4 (2) | 5 (4) | |||
| Is a randomised controlled trial of elective caesarean section v vaginal birth ethical? | Yes
|
246 (37)
|
189 (40)
|
57 (30)
|
P=0.013 | 26 (21)
|
P<0.001 |
| No
|
367 (56)
|
252 (54)
|
114 (61)
|
86 (70)
|
|||
| Missing | 47 (7) | 27 (6) | 17 (9) | 11 (9) | |||
| Is a randomised controlled trial of elective caesarean section v vaginal birth feasible? | Yes
|
159 (24)
|
128 (27)
|
31 (17)
|
P=0.001 | 35 (29)
|
P=0.32 |
| No
|
434 (66)
|
297 (64)
|
136 (73)
|
72 (59)
|
|||
| Missing | 67 (10) | 43 (9) | 20 (11) | 16 (13) | |||
| Is a randomised controlled trial of elective caesarean section v vaginal birth desirable? | Yes
|
294 (45)
|
218 (47)
|
76 (41)
|
P=0.15 | 39 (32)
|
P=0.005 |
| No
|
318 (48)
|
223 (48)
|
94 (50)
|
73 (59)
|
|||
| Missing | 48 (7) | 27 (6) | 17 (9) | 11 (9) | |||
| Would you recruit women to a randomised controlled trial of caesarean section v vaginal delivery? | Yes
|
247 (37)
|
186 (40)
|
61 (32)
|
P=0.073 | 21 (17)
|
P<0.001 |
| No
|
376 (57)
|
259 (55)
|
116 (62)
|
89 (72)
|
|||
| Missing | 37 (6) | 23 (5) | 11 (6) | 13 (11) | |||
| Would you consider an elective caesarean section for yourself (or partner)? | Yes
|
216 (33)
|
180 (39)
|
35 (19)
|
(P<0.001)§ | 9 (7)
|
(P<0.001)§ |
| No
|
316 (48)
|
189 (40)
|
127 (68)
|
100 (81)
|
|||
| N/A
|
98 (15)
|
80 (17)
|
18 (10)
|
0
|
|||
| Missing | 30 (5) | 19 (4) | 8 (4) | 14 (11) | |||
Sex was unknown for four.
P values for χ2 test comparing proportions of “yes” replies between sexes for obstetricians.
P values for χ2 test comparing proportions of “yes” replies between professional groups.
Not applicable. Values have been omitted from the analysis.
Altogether 660/924 (71%) eligible obstetricians and 123/169 (73%) midwives responded (table). Almost half of the obstetricians and a quarter of midwives believed that a woman should choose her method of delivery. A minority thought a trial was feasible, ethical, or desirable. Female obstetricians were less likely to support a trial than male ones. Whether or not the obstetrician and midwife had children did not influence their responses; nor did the type of unit in which the professionals worked.
A full description of qualitative findings will be presented elsewhere. Most respondents providing qualitative commentary wished they had the results of a randomised controlled trial. Obstetricians and midwives who were opposed to a trial offered similar reasons, motivated by unease with routine caesarean section (interference with nature, maternal morbidity, and impact on organisational resources and professional roles). Marked differences occurred in the responses of health professionals who supported a trial: obstetricians mainly believed that lack of evidence prevented women making informed choices, whereas midwives were confident that a trial would prove that vaginal birth was superior.
What is already known on this topic
Caesarean section, in the absence of clear clinical indication, is one of the most contentious issues in modern obstetrics, fuelling debates about the possible need for a randomised trial of delivery methods
Evidence about the extent to which obstetricians support women's requests for caesarean section is conflicting, and views on the need for a trial of planned caesarean section versus planned vaginal birth have not been reported
What this study adds
A minority of consultant obstetricians and heads of midwifery would support a randomised trial of planned caesarean section compared with planned vaginal birth
Comment
Our quantitative findings indicate that a minority of professionals would recruit to a trial comparing planned caesarean section with planned vaginal birth. However, the qualitative finding—that midwives who favoured a trial did so because of their confidence in the benefits of vaginal birth—adds complexity because it negates the necessary individual professional equipoise.4 We explored the opinions of senior obstetricians and midwives simultaneously and nationally, and we identified views about a possible randomised trial. To gain unbiased views, we deliberately did not present participants with a protocol, rather than give the impression that a trial was planned. We also believe that evidence about benefits and risks is insufficient to develop a protocol.
If caesarean birth were shown to be as safe as normal birth in a non-inferiority trial, the NHS would have to consider whether it would be willing to offer such a choice, given the huge resource implications.2 If the cost makes offering choice to all women unfeasible then carrying out a trial would be unethical. The ethical, moral, and practical challenges to a trial are considerable and would require involvement of women and society at large.
This article was first posted on bmj.com on 22 August 2005: http://bmj.com/cgi/doi/10.1136/bmj.38560.572639.3A
The authors acknowledge the support of Ruth Cattrell, research midwife, all of the heads of midwifery, and consultant obstetricians who participated.
Contributors: TL, CK, GG, MG, and JPN designed the study. TL, CK, and AH analysed the data. TL, CK, and AH wrote the paper. TL, CK, AH, GG, MG, and JPN reviewed and amended drafts of the paper. All authors contributed critical comments to the paper. TL is the principal investigator of the survey and principal guarantor of the paper.
Funding: University of Central Lancashire, University of Liverpool and Liverpool Women's Hospital NHS Trust.
Competing interests: None declared.
Ethical approval: Liverpool Research Ethics Committee (twice; reference 01/008-08/03/2001, reference 01/008-11/03/2003).
References
- 1.Minkoff H, Chervenak F. Elective primary cesarean delivery. N Engl J Med 2003;348: 946-50. [DOI] [PubMed] [Google Scholar]
- 2.National Collaborating Centre for Women's and Children's Health. Caesarean section clinical guideline. London: RCOG Press, 2004. [DOI] [PubMed]
- 3.Ecker JL. Once a pregnancy always a cesarean? Rationale and feasibility of a randomized controlled trial. Am J Obstet Gynecol 2004;190: 314-8. [DOI] [PubMed] [Google Scholar]
- 4.Weijer C, Shapiro SH, Glass KC, Enkin MW. Clinical equipoise and not the uncertainty principle is the moral underpinning of the randomised controlled trial. BMJ 2000;321: 756-8. [DOI] [PMC free article] [PubMed] [Google Scholar]