Concordance of Self-Reported Obstetric Outcomes With Electronic Health Record Documentation: A Secondary Analysis of the Nulliparous Pregnancy Outcomes Study Monitoring Mothers-To-Be.

backgroundTo compare participants' perceptions of reasons for admission, labor induction, and rationale for cesarean birth to those documented by clinicians in the electronic health record (EHR) and if these comparisons differed by race and ethnicity.

methodsWe conducted a secondary analysis of data from the Nulliparous Pregnancy Outcomes Study: Monitoring mothers-to-be (2010-2013). We calculated kappa to evaluate agreement between participant postpartum interviews and EHR notes (n = 6085).

resultsOverall, agreement was variable. There was substantial agreement (κ = 0.717, 95% CI 0.704, 0.724) on the reason for obstetric admission, but concordance for the admission reason was lower among preterm births. Only fair agreement was observed (κ = 0.290, 95% CI 0.220, 0.360) for whether labor was electively induced or medically indicated. As a whole, moderate agreement was observed between interview and EHR data on the indication for cesarean birth. EHR documentation on elective labor induction was moderately concurrent with interviews from non-Hispanic White participants; however, it was poor to fair among all other subgroups. DISCUSSION: There were varying degrees of concordance between patient perceptions and what is documented within the EHR. Clinicians should ensure effective communication regarding obstetric procedures and the rationale for interventions, particularly those that are elective. Decision making in later pregnancies should consider all sources of data (EHR and self-reported), particularly where patient and EHR data are discordant around labor dysfunction.