(605) EHR-integration of an evidence-based calculator for cesarean risk during induction: impact on utilization and outcomes

Objective: We developed a validated calculator to predict individualized likelihood of cesarean delivery (CD) for patients undergoing induction (IOL) [Levine 2018]. In a prospective cohort, calculator implementation via an external website was associated with reduced maternal morbidity and CD [Hamm 2020]. Here, we aimed to determine if Electronic Health Record (EHR) integration of the CD risk calculator would further improve utilization and clinical outcomes.

Study Design: This prospective cohort evaluated calculator utilization, maternal morbidity, and CD for 6 months PRE (10/2022-4/2023) vs. 6 months POST EHR-release of the CD risk calculator (4/2023-10/2023) in 2 units. Prior to EHR-integration, Site#1 had a 2018 push for calculator website use, while Site#2 had not. EHR-integration recognized eligibility, allowed clinicians to confirm and utilize EHR variables for score automation, visible on electronic display. Patients were calculator eligible if undergoing term IOL with a singleton, intact membranes, and dilation ≤2cm without prior CD. Calculator utilization was measured as a score recorded in the EHR. Analysis was stratified by site and risk of CD based on calculator results (< 20%, 20-39.9%, 40-59.9%, ≥60%).

Results: 2018 patients met eligibility (PRE:998; POST:1010) with more IOLs for fetal indications, earlier gestational age, and higher rates of hypertension in the POST group. Calculator utilization significantly increased POST-EHR implementation (Table). There were no differences PRE to POST in maternal morbidity overall (aRR 1.18 [0.91-1.53]), by site or CD risk strata. There was no difference in CD overall (aRR 1.19 [0.99-1.43). However, an increase in CD was found at Site#2 POST EHR-integration (aRR 1.30 [1.02-1.66]), specifically in the ≥60% CD risk strata.

Conclusion: EHR-integration of a CD risk calculator was associated with substantial increases in utilization. While CD increased at one site PRE to POST, PRE rates were lower than anticipated, and POST rates consistent with institutional data. Care should be taken this calculator is not used to perform CD outside of clinical indications.