(275) The Effects of Dynamic Patient Trajectories on Prenatal Care Utilization and Clinical Operations

 To assess the effects of a new tailored prenatal care policy on clinical capacity and efficiency, accounting for dynamic patient trajectories in pregnancy.

Study Design: We conducted a discrete event simulation study of traditional versus tailored prenatal care policies. We compared the traditional care model (14 in-person visits for all patients), to a schedule tailored to medical risk (high-risk: 13 visits; low-risk: 9 visits), and preference for care modality (in-person vs. virtual). We also introduced dynamic patient trajectories to account for medically low-risk patients who develop complications (e.g. gestational diabetes, gestational hypertension) in pregnancy, requiring transition to the high-risk schedule. Data inputs were derived from 4,992 real patients who received prenatal care and gave birth at a single academic institution from March 1 2022 to March 1 2023. Simulation parameters included probability of diagnosis of complications, preference for hybrid care (derived from local survey data), patient risk level, and dynamic patient arrival times. The simulation was run in C++ for 100 replications of 92 weeks each. We reported on measures of clinical operations including capacity utilization, patient delay, and overbooking.

Results: In the dynamic patient simulation, 12.18% of low-risk patients became high-risk. Of patients who switched, the average number of appointments added to the pathway was 1. Following a traditional visit schedule showed in-person capacity utilization of 107%, implying overbooking, and patient delay of 0.41 weeks per patient on average. When compared to a traditional visit schedule, implementing a tailored prenatal care policy showed average capacity utilization of 92% (including in-person and virtual appointments), eliminating overbooking. Patient delay was reduced to 0.36 weeks per patient.

Conclusion: The implementation of tailored prenatal visit schedules allowing for a dynamic patient type model eliminates in-person overbooking and reduces patient delays. Further study is needed to determine the optimal proportion of clinic capacity to allocate to virtual care.