OUTCOMES: THE INTEROBSERVER VARIABILITY SUNEET CHAUHAN, CHAD KLAUSER, THOMAS WOODRING, MAUREEN SANDERSON, EVERETT MAGANN, JOHN C. MORRISON, Society for Maternal-Fetal Medicine, West Allis, Wisconsin, University of Mississippi Medical Center, Jackson, Mississippi, University of Mississippi Medical Center, Obstetrics & Gynecology, Jackson, Mississippi, University of Texas Health Science Center at Houston, Brownsville, Texas, Naval Medical Center, Obstetrics & Gynecology, Portsmouth, Virginia, University of Mississippi Medical Center, Obstetrics and Gynecology, Jackson, Mississippi OBJECTIVE: The purpose of this study was, among patients with non-reassuring (NR) fetal heart tracings (FHT), determine the interobserver variability (IV) in interpretation of FHT (using the definitions in ACOG practice bulletin on the topic) and the ability to predict 1) patients requiring emergency cesarean delivery (ECD); 2) umbilical arterial [UA] pH 7.00; 3) base excess [BE] -12 mmol/L; and 4) Apgar score [AS] 3 at 5 min. STUDY DESIGN: Five clinicians (1 generalist and 4 MFM), not involved with the care of cohorts, reviewed 100 FHT. Ten tracings with objective evidence of fetal compromise were matched with 9 other FHT from the same time period. The FHT were divided into two phases: an hour prior to periodic decelerations and, if applicable, the hour before delivery. Weighted kappa coefficients (WKC) were calculated to assess inter-observer variability for episodic decelerations, whether the FHT is reassuring or uninterruptible, and the four outcomes. Spearman correlation coefficients (SCC) were calculated to assess the correlation between predicted and actual outcomes overall and by observer. RESULTS: Among 100 cohorts, 46% had CD for NR FHT; 2%, UApH 7.00 (acidosis); BE 12 in 14% and 3% had AS 3 at 5 min (low AS). The WKC for classifying the FHT as being reassuring in early labor was 0.12 and before delivery 0.15. The WKC for the 5 clinicians for predicting ECD was 0.26; for acidosis, 0.21; for abnormal base excess 0.16, and; for low AS at 5 min, 0.17. The SCC for each observer was 0.005 to 0.21 for ECD; for UApH, 0.01 to 0.46; for BE 0.03 to 0.29 and; for low AS 0.0 to 0.31. In the prediction of the UApH 7.00, three observers differed significantly (P 0.05) from one of the clinicians. The SCC range from 0.10 to 0.19 for the 4 outcomes. CONCLUSION: The reproducibility of and utility of FHT as a predictor of neonatal hypoxia is questionable because, there is a poor agreement among clinicians regarding interpretation of FHT and they cannot reliably identify newborn with hypoxia.