Abstract
At 10 Danish obstetrical departments, 116 residents (42 senior and 74 junior) participated in a study to assess inter-observer and regional variability in the description and evaluation of and clinical decision regarding 11 fetal heart rate patterns. The 11 traces included normal as well as pathological patterns, and normal as well as clinically asphyxiated babies. Five antepartum and six intrapartum patterns were included. A total of 1,276 descriptions and evaluations were obtained. The degree of agreement in description of fetal heart rate changes was high regarding the baseline and the presence of silent or sinusoidal pattern (87–94% on an arbitrary 0–100% scale), and low regarding the assessment of variability and type of deceleration (50–72%). The degree of agreement in interpreting heart rate patterns was 59% (on an arbitrary 0–100% scale). Senior residents generally interpreted the changes as indicative of less serious fetal stress than did their junior colleagues, explaining why junior residents 30% more frequently than their older colleagues found an indication for Cesarean section. Relatively low regional inter-observer agreement scores were primarily due to low agreement between departments, especially between departments far apart. It is concluded that we still need a scientific clarification of which specific heart rate changes are the best predictors of fetal stress. Artificial intelligence programs for interpreting fetal cardiotocograms and ECG signals constitute one promising prospect.