Interobserver agreement of intrapartum cardiotocography interpretation by midwives using current FIGO and physiology-based guidelines

Objectives: To investigate the interobserver agreement among midwives for the interpretation of intrapartum CTG traces, using both the 2015 FIGO guidelines and the physiology-based guidelines.

Material and methods: Retrospective, single-center, observational study conducted at the Maternity Hospital of the University of Parma, Italy, between November 2022 and July 2023. We selected 100 nonconsecutive intrapartum CTG traces from women undergoing operative delivery (cesarean or instrumental vaginal delivery). Participating midwives received the last 60 min of CTG recording prior to the decision for an operative delivery was made. Five midwives with different professional experience classified the CTG tracing according to the 2015 FIGO guidelines and the physiology-based guidelines at the time of the decision for the operative delivery. Each midwife was categorized as senior or junior if her clinical experience was equal or more or less than 5 years. All participating midwives were blinded to the demographic and clinical variables of each case, as well as to the labor and neonatal outcome. The main outcome of the study was the interobserver agreement among the 5 midwives with either guideline. The secondary outcome was the interobserver agreement based on clinical experience. The agreement between the observers was estimated through the Krippendorff's alpha statistics.

Results: The interobserver agreement for the interpretation of intrapartum CTG traces according to the 2015 FIGO guideline was in the upper limit of the "unacceptable" classification, with a Krippendorff's alpha of 0.632 (95% CI, 0.545-0.723). The interobserver agreement of the physiology-based CTG guideline was classified as "unacceptable" with a Krippendorff's alpha of 0.359 (95% CI, 0.125-0.582). The interobserver agreement among senior midwives according to the 2015 FIGO guidelines was considered as "tentative acceptable agreement" with a Krippendorff's alpha of 0.724 (95% CI, 0.636-0.805). Among junior midwives, the interobserver agreement according to the latter guideline was classified as "unacceptable" with a Krippendorff's alpha of 0.569 (95% CI, 0.359-0.751). Using the physiology-based guideline, the interobserver agreement of the senior midwives - Krippendorff's alpha 0.493 (95% CI, 0.320-0.623) - and "junior" midwives - Krippendorff's alpha 0.359 (95% CI, 0.100-0.581) - were considered both "unacceptable".

Conclusions: Our study showed that intrapartum CTG interpretation is characterized by an overall unacceptable level of interobserver agreement among midwives, with the 2015 FIGO guidelines showing better agreement compared to physiology-based guidelines. The interpretation of intrapartum CTG by "senior" midwives using the 2015 FIGO guidelines showed the highest, interobserver agreement, indicating a "tentative acceptable agreement".