Editor's Choice

Uncomfortable findings

BMJ 2005; 331 doi: http://dx.doi.org/10.1136/bmj.331.7515.0-g (Published 01 September 2005) Cite this as: BMJ 2005;331:0-g
  1. Fiona Godlee, editor (fgodlee{at}bmj.com)

    What makes it worthwhile and possible to do a randomised controlled trial (RCT)? Here are some things: an important question without a definitive answer, scientific equipoise, feasibility, and funding. Not listed here, for obvious reasons, is whether the trial will come up with an acceptable answer. This week's BMJ presents several RCTs—including a well performed trial of flocculant-disinfectant for purifying water in rural Kenya (p 478). The trial is funded by the producer of the disinfectant but with the necessary safeguards against non-publication should the trial have failed to show that the product works. But should fears about the outcome of a trial ever prevent a trial from being done?

    Here, for example, is a question ripe for testing in an RCT: is elective caesarean section as safe as vaginal birth? Or to put it another way, should a woman be free to choose to have a caesarean section when she has no clinical need for one? It's a question being asked by increasing numbers of pregnant women, and one that in an age of patient choice should perhaps have only one answer. But we also live in an age of evidence based health care and limited resources, and at the moment midwives, GPs, obstetricians, and policy makers have no RCT evidence to turn to. What they do have, as Tina Lavender and colleagues have found (p 490), are strongly held and divided opinions that would make a trial hard if not impossible to do.

    The lack of consensus is clear from a look at national and international guidelines. The American College of Obstetricians and Gynaecologists says elective caesarean is ethical, and in Italy this is enshrined in law. But the International Federation of Gynaecology and Obstetrics says it's unethical, while the UK's National Institute for Health and Clinical Excellence—ever cautious—recommends seeking a second opinion.

    Lavender and colleagues are in no doubt that a trial comparing vaginal delivery and elective caesarean section would provide important evidence that could help to solve this controversy. And most of the nearly 900 midwives and obstetricians in England who responded to their questionnaire said that they wished they had the results of an RCT. But despite this only a minority of respondents were in favour of a trial taking place: about two thirds took the view that a trial would not be feasible, ethical, or desirable.

    Those who supported a trial, especially the midwives, did so largely because they believed it would show that vaginal delivery was better, while opposition to a trial was largely motivated by unease about elective caesarean section—because it interfered with nature, increased maternal morbidity, or had implications for resources or professional roles. The authors are right to identify the lack of individual professional equipoise as a major obstacle to such a trial. But should concerns about the cost of providing elective caesarean sections more widely be a reason for not seeking an answer to this important clinical question? I don't think so.

    Footnotes

    View Abstract