RCOG guidelines: How evidence based are they?

By Gemma McKenzie

AIMS Journal, 2015, Vol 27 No 1

Prusova K, Tyler L, Churcher A and Lokugamage AU (2014) Royal College of Obstetricians and Gynaecologists guidelines: How evidence-based are they? J Obstet Gynaecol. 2014 Nov;34(8):706-11. doi: 10.3109/01443615.2014.920794

The Royal College of Obstetricians and Gynaecologists (RCOG) publishes guidelines in order to assist clinicians with decision making. The present study aimed to assess how many of those guidelines were backed up by best quality medical evidence.

Summary of the research

The researchers analysed the RCOG's 'Green-top Guidelines.' Each of these guidelines deals with a specific topic, and contains recommendations, which have been graded according to the quality of medical evidence backing them up. Prior to December 2007, the grades ranged from A to D, and after this date from A to E. In both cases the best kind of evidence was generally considered to be a randomised controlled trial (A), although post December 2007, this could also include either a metaanalysis or a systematic review. The lowest level type of evidence was considered to be based on the clinical experience of the guideline development group.

The researchers found that prior to December 2007, only 8% of the Green-top obstetric guidelines were based on the highest level of medical evidence, whereas 41% were based on the lowest type. Post December 2007, 8% were based on the highest level of medical evidence and 40% on the lowest. Of the gynaecology guidelines, before December 2007, 18% were graded A and 40% as D. After this date, the figures were 13% and 42% respectively.

The researchers concluded that the evidence backing up the majority of RCOG guidelines is based on clinical experience, expert opinion or low quality studies.

AIMS Comments

Although the researchers' conclusion is sobering, there are some general points worth noting. Firstly, attempting to categorise the evidence with a simple A to D/E grading may be too crude to truly assess the quality of the evidence. Secondly, conducting randomised controlled trials within some areas of obstetrics and gynaecology may cause practical and ethical dilemmas. For example, given that women have been led to believe that longer pregnancies are dangerous, how many would feel confident to be part of a trial that attempted to see what happens when a pregnancy goes beyond 42 weeks? Thirdly, this study does not consider the midwifery guidelines produced by the Royal College of Midwives (RCM).

The researchers also pointed out that the existence of research in a particular area does not necessarily guide best practice. For example, following a review, the Cochrane Collaboration¹ concluded that it could not recommend the use of partograms as part of standard labour care. However, partograms are still considered fundamental to clinical practice and are regularly used.

Given that the majority of the RCOG guidelines are based on expert opinion and not the highest quality evidence, this begs the question of how much weight practitioners are putting on those guidelines. Do they understand and take into account the grading system? Although the RCOG suggests that clinicians should do otherwise, are medical professionals simply following the guidelines blindly, seeing them as rules that must be rigidly upheld? Are the guidelines being followed as a form of defensive medicine, even if a practitioner feels an alternative course of action would be preferable? Are they discussing the quality of the underpinning evidence with pregnant women so that they can make the best decision for themselves and their baby? The researchers' conclusion is also relevant to a pregnant woman who is told that, for example, the RCOG guideline recommends a caesarean section in her situation. She may presume that this recommendation is based on hard evidence.

This has implications for informed consent.

A further issue with the heavy reliance on small numbers of experts' opinion to form guidelines is the risk of bias and consequently of error.

Interestingly, a similar study was carried out in the US.² It found that only 25.5% of the American College of Obstetricians and Gynecologists' guidelines were based on the highest level of medical evidence and that 34.8% were based on the lowest. The study also noted that when the RCOG recommendations were compared to the American College's, only 28% were the same, 56% were not comparable and 16% were opposite. This raises the question of accuracy, and also the issue of whether there is such a thing as 'best practice'. Might it therefore be preferable to focus on woman-centred rather than guideline-centred care? Health practitioners would then be able to draw on the best upto- date research and also use their professional judgement to help women make their own decisions.

Although initially shocking, the results of this study only paint part of the picture. To get a more accurate perspective, we need to consider whether there is any feasible way of creating more robust research, how exactly clinicians are using the guidelines, and whether this is ultimately having a negative effect on women, their pregnancies, babies and births.

Gemma McKenzie is a mother of three and about to embark on a PhD researching informed consent within the maternity services

References

1. Lavender T et al (2008) Effect of partogram use on outcomes for women in spontaneous labour at term. Cochrane Database of Systematic Reviews (4): CD005461
2. Chauhan SP, Hendrix NW, Berghella V, Siddiqui D (2010) Comparison of two national guidelines in obstetrics: American versus Royal College of Obstetricians and Gynaecologists, AM J Perinatal 2010;27:763-70