Development of a Risk Score to Predict Obstetric Anal Sphincter Injuries in Laboring Patients [8OP

Abstract

INTRODUCTION: We piloted the utility of a scoring system to predict patient risk of obstetric anal sphincter injury (OASIs) during labor. METHODS: Among the 9,800 women with singleton term cephalic vaginal deliveries within Kaiser Permanente Northern California between January and December 2013 with complete study data, 8.5% (n=833) of the nullipara and 3.5% (n=343) of the multipara incurred OASIs. Multivariable logistic regression and classification and regression tree analyses were used to identify risk factors. Maternal and gestational age, race/ethnicity, duration of second stage of labor, vacuum delivery, and history of OASIs were associated with OASIs. Stratified by parity, points were assigned to each risk factor as an integer on a 10-point scale based on effect size and impact in the regression analyses, and were externally validated within a similar separate patient cohort using multivariable logistic regression. RESULTS: Different OASIs risk score cut-offs were assessed. Among nulliparous women, a score ≥6 had 40% sensitivity and 91% specificity, whereas a score ≥4 had 51% sensitivity and 84% specificity. Among multiparous women, a score ≥7 had 89% sensitivity and 60% specificity compared to a score ≥10, which had 74% sensitivity and 72% specificity. CONCLUSION: These pilot scores, demonstrating a reasonable predictor of OASIs risk, illustrate the utility of a clinical risk score to inform providers of patients' risk of OASIs prior to delivery and facilitate a shared decision-making approach to labor management.