I spotted a paper published late last year which has helped me to find the language to capture something I was aware of in relation to CTG monitoring but found difficult to put into words. Rogers, Entwistle and Carter have written a thought-provoking paper about one of the ethical issues that arises from the use of screening tests – that of overdiagnosis. What exactly is overdiagnosis?
Overdiagnosis is a counter‐intuitive phenomenon as detecting instances of disease is usually considered to be beneficial (given assumptions about the availability of effective treatments and the value of knowledge about prognoses). Overdiagnosis upends these assumptions as there are more harms than benefits from the diagnosis, often because the detected condition would not have progressed to advanced disease, thus there is no benefit to detection. Overdiagnosis is also counterfactual in that the diagnosis and any associated interventions cause harm on balance because people’s overall health states and experiences are worse than they would otherwise have been without the diagnostic intervention.
Rogers, et al., 2019, p. 237.
Let me give you an example to make this clearer. Joan is a 52-year old woman. On her birthday she decides to stay home and watch reruns of Dr Who. She lives a happy and inspiring life, and then suddenly dies, aged 61, from a stroke.
In an alternate timeline however, 52-year old Joan shuns Dr Who and decides instead to have a mammogram on her birthday. A breast cancer is detected. She has a lumpectomy, chemotherapy, and radiotherapy. The treatment is effective, and Joan never has a recurrence. Her family are so grateful that the mammogram saved their mum’s life, that they donate a large amount of money to set up a mobile mammogram service. Joan lives a full and inspiring life, and then suddenly dies, aged 61, from a stroke.
There was no way for Joan number two or her family to know it, but she has been overdiagnosed. Her cancer was never going to give her a moment’s bother, shorten her life, or cause her death. What could have been a third possible alternate timeline was that Joan died, aged 53 from sepsis as a complication of her chemotherapy. Joan’s family were still grateful that her cancer was detected, because they felt that it would have been awful for her to die from cancer.
See the thing with overdiagnosis is that it is only possible to measure it at a population level in research.
As we can see from Joan’s three alternate timelines, Joan and her family have no way of knowing that Joan number two would have been no worse off if she stayed home to see if the Daleks had learned to climb stairs yet, and Joan number three would actually have been much better off if she had. Unlike Dr Who, we can’t drop into our alternative future timelines to see what the consequences of our choices are. The Cochrane review on screening mammograms highlights that for every individual woman who experiences a better outcome as a consequence of having a mammogram, ten women will experience overdiagnosis. If you receive a breast cancer diagnosis after mammogram screening you have no way of knowing which category applies to you.
So, what does overdiagnosis have to do with CTG monitoring? It is well known that the sensitivity of CTG monitoring is poor. That is, when the CTG is considered as abnormal, the possibility that the fetus is actually at risk of harm and will therefore benefit from being born sooner is really low (astonishingly gobsmackingly low actually). Calculating the sensitivity of CTG monitoring is quite tricky as it depends on what you count as an abnormal CTG, and which outcome you are looking at. Nelson and co-authors (1996) calculated that for every 1000 fetuses exhibiting multiple late decelerations, or decreased beat-to-beat variability of the fetal heart rate, or both, then 1.9 would go on to develop cerebral palsy if corrective action were not taken. In other words, the sensitivity of CTG monitoring is 0.19%, giving a false positive rate of 99.8%. (Intrapartum stillbirth is one fifth as common as cerebral palsy so the sensitivity for detecting this is lower still.)
So for every 1000 women who are told that something must be done because the CTG is very abnormal, there are potentially two whose baby might avoid cerebral palsy. The other 998 were overdiagnosed, and this group will include women and babies who were no worse off, but also many who experienced complications which could have been avoided if the CTG had not been used. But I would bet that almost all of those 998 women, and their care providers, will probably consider themselves lucky that the problem with the baby was picked up early by the CTG so that something could be done about it. Some of them will purchase CTG machines for their local hospital, and one of them might even become the health minister one day.
And that is one of the reasons why the myth that CTG monitoring is effective is so difficult to shift from our collective consciousness. The irony is that if CTGs had better sensitivity, fewer people would believe that their lives were made better because of CTG monitoring.
Gøtzsche, P. C., & Jørgensen, K. J. (2013). Screening for breast cancer with mammography. Cochrane Database of Systematic Reviews (6). doi:10.1002/14651858.CD001877.pub5
Nelson, K., Dambrosia, J., Ting, T., Grether, J. (1996). Uncertain value of electronic fetal monitoring in predicting cerebral palsy. New England Journal of Medicine, 334(10), 613 – 619. https://dx.doi.org/10.1056/nejm199603073341001
Rogers, W. A., Entwistle, V. A., & Carter, S. M. (2019). Risk, overdiagnosis and ethical justifications. Health Care Analysis, 27(4), 231-248. doi:10.1007/s10728-019-00369-7
Categories: CTG, EFM, Reflections
Birth Small Talk 
Many thanks for this review. As a Midwifery educator it is so hard to change the deeply ingrained practice of ‘routine CTGs’
Many thanks
Meredith
LikeLike
Thanks Meredith – education is a major part of the solution.
LikeLike