Inconsistencies in classification by experts of cardiotocograms and subsequent clinical decision

Area of Science:

• Obstetrics and Gynecology
• Fetal Monitoring
• Medical Statistics

Background:

• Cardiotocography (CTG) is crucial for assessing fetal well-being during pregnancy.
• Standardized interpretation guidelines, such as FIGO, aim to improve consistency.
• Inter-observer variability can impact clinical decision-making and patient outcomes.

Purpose of the Study:

• To evaluate inter-observer agreement among experts in classifying cardiotocographic tracings.
• To assess agreement on subsequent clinical decisions based on CTG interpretation.
• To identify areas of significant disagreement in CTG analysis.

Main Methods:

• 33 cardiotocographic tracings were interpreted by multiple experts.
• Agreement was quantified using the kappa statistic (K) and proportion of agreement (Pa).
• Classification categories included normal, suspicious, and pathologic tracings, with clinical decisions analyzed.

Main Results:

• Overall agreement for tracing classification was fair (K = 0.48), with higher agreement for normal tracings (Pa = 0.62) than suspicious (Pa = 0.42) or pathologic (Pa = 0.25).
• Agreement on clinical decisions was slightly higher (K = 0.59), largely driven by consensus on 'no action' (Pa = 0.79).
• Significant disagreement occurred regarding decisions to 'monitor closely' (Pa = 0.14) and 'intervene immediately' (Pa = 0.38).

Conclusions:

• Inter-observer agreement in cardiotocography interpretation and clinical decision-making shows limitations.
• Discrepancies, especially in critical decisions, necessitate careful consideration in clinical audits and legal contexts.
• Further standardization or training may be needed to improve consistency in high-risk CTG interpretations.