Strengths and Limitations
The main strength of this study was the careful selection of medical records. Ten medical records were selected among forty others by four reviewers, including one expert in court. Clear instructions were given beforehand to the reviewers to only retain records whose outcome was unequivocal. In other words, the four reviewers had to agree that the actual mode of delivery (either a caesarean section or a vaginal delivery) was the best possible outcome in the clinical context with knowledge of the entire medical record, including the results of the foetal scalp blood sampling and the neonatal status. This rigorous selection process legitimises the claim that choosing an alternative mode of delivery without using a second-line method would at least result in a nonoptimal outcome for the patient or her newborn. The exhaustive and careful presentation of all the information necessary for the participants to make a decision should also be emphasized. Real CTGs were presented either entirely since the beginning of labour or partially since before the beginning of CTG abnormalities (when the tracings exceeded three hours). Hence, this online questionnaire allowed an original simulation of real clinical situations, and we believe that it is a reliable evaluation of the participants’ professional attitudes.
However, several considerations mitigate the conclusions that can be drawn from the results of this study. Despite the aforementioned precautions, it is impossible to assert with absolute certainty the outcome of an alternative scenario. Although we consider it unlikely, it is not formally impossible that among the 4 records that resulted in a caesarean section, one or more records could have ended differently. Several participants also pointed out that reducing the choice to three options was too restrictive and did not give a fair representation of reality. In particular, it is possible that some participants would have opted to continue labour only for some time before ultimately performing a caesarean section or a second-line method. Similarly, some participants suggested that other options were available, such as positioning the patient differently or stopping oxytocic administration. As stated before, although four reviewers deemed that the actual outcome was the best possible option, alternative scenarios cannot formally be excluded.
In addition, there is no way of verifying that the participants would have actually made the stated decision in a real-life situation. It is possible that the context of the anonymous questionnaire, devoid of any stress related to the consequences and presented as a test, encouraged participants to take more risks in their decision-making. It is also worth mentioning that the majority of the participants of this online survey were residents. Although we only included residents in their fourth year or higher who are ordinarily already self-reliant in obstetrical decision-making, we can hardly extrapolate the results to more experienced physicians.
Furthermore, all modes of training for the CTG physiology-based interpretation were placed on an equal footing in the training group. The number of participants was insufficient to provide subgroup analysis for each type of training. Although the vast majority of the participants from the training group declared that they used the precepts of CTG physiology-based interpretation in their daily practice, it is possible that our results are biased by the insufficient training of some participants.