From MIDWIFERY MATTERS, Issue No.91, Winter 2001
Margaret Jowitt
THE RCT (randomised controlled trial) presents all sorts of problems for midwifery.
First and foremost the individual woman is taken right of the picture straight away because, by definition, RCTs deal with statistics, averages and probability. They are designed to annihilate the individual. They try to answer the question as to what is best for the average patient/case/subject. A midwife basing her practice only on RCTs is no longer a midwife because she is no longer 'with woman' but with some sort of average woman who does not actually exist.
Carl Jung's account of the problem of statistics seems particularly pertinent to midwifery:
The statistical method shows the facts in the light of the ideal average but does not give us a picture of their empirical reality, it can falsify the actual truth in a most misleading way. This is particularly true of theories which are based on statistics. The distinctive thing about real facts, however, is their individuality. Not to put too fine a point on it, one could say that the real picture consists of nothing but exceptions to the rule, and that, in consequence, absolute reality has predominantly the character of irregularity
However, I certainly do not want to discount the value of statistics in midwifery. In Safer Childbirth? Marjorie Tew provides probably the best statistical evidence of the superiority of midwifery care. Delivered at Home by Julia Allison is another powerful use of statistics. The statistical evidence pertinent to midwifery care tends to be based on large proportions of whole populations and to compare midwives' figures with hospital figures, which is really a comparison of midwife-led care with obstetric care. The drawback is that most of this large scale evidence is historical and so intrinsically less 'reliable' than the prospectivity of the RCT.
Second, RCTs are designed to test one treatment against another treatment or indeed no treatment. What 'treatments' are there to test in midwifery?
I believe that midwifery is essentially a 'hands off' profession dealing with a population of healthy women performing their natural function where the aim is to prevent the need for any 'treatment' at all but instead to empower women to give birth to their own children. For midwives RCTs of obstetric treatments are useful for proving scientifically that, on the whole, no treatment is better than treatment, that it is best to avoid induction, ARM, active management of all stages of labour and so on and so forth. Such RCTs do not constitute midwifery evidence per se, but can be used by midwives in their role as women's advocates and to enable them to decide whether or when to perform an obstetric intervention.
Maggie Banks is similarly uneasy about the place of obstetric evidence being used as a basis for midwifery practice: "The underlying theme seems to have been to pick apart obstetric practice in the hope of exposing what midwifery care is about". Perhaps this sounds rather harsh and idealistic but as the ARM Normal Birth group point out (p 15-16) there is a real lack of midwifery research as opposed to obstetric research. Maggie Banks is an independent midwife practising in a country with a maternity service that has come full circle. Midwifery in New Zealand was all but moribund when midwives and women managed to break the obstetric stanglehold over childbirth. Also no fault compensation may have made it easier for service providers to give more power to midwives.
The situation is very different in the UK; midwives have very little power to affect obstetric and management decision making and the midwifery guideline approach is an attempt to change the system from within, this means using research known to be acceptable by the powers that be. Unfortunately the power structure in the UK is such that midwives are forced to use obstetric research in order to validate midwifery care.
Another related issue concerns getting research through the hoops of ethics committees which are dominated by medical men. For example, a question came up on the ukmidwifery mailing list concerning intermittent ausculation. Currently NICE recommends quarter hourly ausculation; this timing was establised in a RCT comparing intermittent ausculation with EFM (electronic foetal monitoring). It is very doubtful that a protocol which deviated from current standard practice, for example, comparing hourly or half hourly ausculation with the standard quarter hourly ausculation would be approved. In the current litigious climate, ethical committees would tend to err on the side of medical caution whereas the midwifery bias is towards less intervention.
Third, and perhaps most important, by their very nature RCTs should be ethically unthinkable in the context of woman-centred childbirth. Where is maternal choice and control? Relinquishing control to a brown envelope must be one of the worst forms of loss of control and I believe that perception of control on the part of the mother is the most important aspect of birth.
Moreover, childbirth is such an intimate, individual and personal event that the effects of being part of a research trial will distort the mother/midwife relationship to such an extent that the results are likely to be seriously flawed. Which arm of the trial would the mother or her midwife prefer to be in? How would being randomised to the non-preferred group affect the labour? Disappointment may well have an effect on the labour. And there are other serious concerns; it is very important that women should be able to trust their caregivers and merely being asked to participate in a RCT could jeopardise that trust. All too often midwives are involved in recruiting women for obstetric trials - I would be happier with a midwife who advised me not to take part in a trial but to keep all my options open.
I asked myself if there was any aspect of care where I would have been happy to participate in a trial and came up with two: comparing different suture materials and type of prostaglandin agent for induction of labour. I think it is significant that both these trials compare different materials for doing the same job, i.e. they are looking at one tiny aspect of care that is unlikely to be affected by the personality of either the woman or her caregiver. Research on complementary therapies in labour such as aromatherapy might also fall into that category.
Of course there may be women more altruistic than me who would be quite happy to take part in RCTs, women who could embrace uncertainty quite happily; it may be that my knowledge and beliefs about the psychodynamics of labour have prejudiced me forever against planned uncertainty. I suppose that this must be the case since obstetricians seem to be able to get maternal consent for RCTs.
I have deep misgivings about putting too much trust in RCTs which, after all, despite the admirable qualities of prospectivity and randomisation are already history on publication - they are still rooted in particular times, places and cultures and as such are as likely as any other research to become outdated. The respect for RCTs is dangerous in itself; if EFM has proved impossible to dislodge from obstetrics, despite the lack of RCT support for it, how much more difficult it will be to dispel the notion that caesarean section is the only safe option for breech delivery after the latest trial was stopped earlier than planned?
On the other hand, for any RCT that hasn't gone the way we hoped (e.g. the Hitchingbrooke third stage study) we tend to find reasons for disbelieving the outcome. This goes for all of us, midwives, obstetricians and women alike. It is human nature to cling on to one's own beliefs in the face of evidence to the contrary.
There was another question on the ukmidwifery mailing list about best management of premature rupture of membranes at term and one of the list members gathered together the RCTs on the subject, although most came down on the side of watchful expectancy there were two that found earlier induction of labour beneficial. There is still scope for incorporating one's own opinions by choosing which RCTs on which to base one's clinical practice.
While it is tempting to believe that research has all the answers, women are not robots who conform to the laws of statistics. RCTs, however rigorously conducted, cannot tell midwives and obstetricians the ideal course of management for all women at all times. While midwifery should be based scientific evidence it should not be bound by that evidence. The evidence of the woman herself, her clinical history, her hopes, her wishes, her personality, her family circumstances, are equally important. To ignore all these in favour of treatment prescribed by an RCT is to ignore her individuality and deny her humanity.
Midwives and obstetricians are also human, with individual strengths and weaknesses; there is an art to giving maternity care which RCTs will never be able to tap. This art lies within individual midwives themselves. Just how do you measure the art of enabling, empowering, 'being with', and so on? And if you could measure it, how on earth could you test it in an RCT? Midwifery must remain an art informed by but not controlled by science.
Allison J (1996). Delivered at Home, Chapman and Hall, London.
Banks M (2001). 'But whose art frames the questions? Practising Midwife, 4, 9, 34-35.
Jung C (1958). The Undiscovered Self, Routledge and Kegan Paul, London.
Tew M (1998). Safer Childbirth? (3rd edn) Free Association Books, London.
This article was originally published in Midwifery Matters ISSUE 91 Winter 2001, p9-10
LW updated February 4, 2005