Inter-observer agreement in clinical decision-making for abnormal cardiotocogram during labour: a comparison between CTG and CTG plus STAN
Article first published online: 10 DEC 2009
© 2009 The Authors Journal compilation © RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology
BJOG: An International Journal of Obstetrics & Gynaecology
Volume 117, Issue 1, pages 121–122, January 2010
How to Cite
Vayssière, C., Arnaud, C., Pirrello, O. and Goffinet, F. (2010), Inter-observer agreement in clinical decision-making for abnormal cardiotocogram during labour: a comparison between CTG and CTG plus STAN. BJOG: An International Journal of Obstetrics & Gynaecology, 117: 121–122. doi: 10.1111/j.1471-0528.2009.02392.x
- Issue published online: 10 DEC 2009
- Article first published online: 10 DEC 2009
- Accepted 28 August 2009.
Sir, We thank JM Grant for his interest in our paper.1
His comments raise the issue of the difficulty of choosing judgment criteria for evaluating decisions made during labour. Hypoxic-ischaemic encephalopathy is indisputably a severe complication that we must aim to prevent. Although its prevalence can be used as a judgement criterion in large-scale studies, it cannot be used for evaluating the appropriateness of decisions made during labour. The objective in such cases is not only to prevent the most serious complications, but also to preserve fetal wellbeing, for which the most widely used biological marker is pH. The majority of experts would agree with JM Grant that a pH of 7.05–7.10 is a significant degree of acidosis, usually indicating delivery; however, this does not generally result in encephalopathy.
The objective of our study was to determine if ST waveform analysis (STAN) could improve the inter-observer homogeneity of decisions during labour compared with cardiotocography (CTG) alone. In order to do this, we selected a series of tracings from an existing database. Given that our total sample had to include a sufficient number of cases justifying intervention, we decided that ten of our 30 cases should have a pH < 7.05. In order to analyse the homogeneity of decision-making, we decided to use the method described in a previous article by JM Grant.2 As in this initial article, we calculated confidence intervals by taking into account the total number of trials of agreement, considering that the decisions of the different observers were independent. It is true that the observations were not all independent, as each obstetrician gave several opinions, and each tracing was analysed several times, but calculating the confidence interval from the number of tracings as suggested would only be relevant if there were only two observers.
As a post hoc analysis, we wanted to subdivide the sample into two groups, taking into account the appropriateness of intervention decisions. It is not always possible to judge a posteriori the appropriateness of a decision made during labour. Indeed, when an intervention is carried out and the pH is normal at birth, it is not possible to distinguish between a false alert having brought about an unnecessary intervention, and an effective intervention. We therefore considered that all cases with pH < 7.05 would constitute the ‘justified intervention’ group, but that the group of cases with pH ≥ 7.05 would be subdivided. In fact, we only considered the cases without a low pH where no intervention had been carried out as being justified non-interventions. In agreement with JM Grant’s comment,3 we considered that the limit of 7.05 was too low for this group, which is why we only kept the 12 cases with a pH > 7.10.4
We agree that this work is only a preliminary step that, as we underlined in the discussion section of our article, must be verified by larger studies.