Predictive Modeling of Hypertension-Related Postpartum Readmission: Retrospective cohort analysis (Preprint)

Abstract

Hypertension is among the most common reasons for postpartum hospital readmission. Better prediction of postpartum readmission will improve healthcare of patients, allow better utilization of resources, and decrease healthcare costs.

Evaluate clinical predictors of hypertension related postpartum readmission using a novel statistical model. We hypothesized blood pressure during labor, not just postpartum, would be an important predictor.

We conducted a retrospective cohort study from a single Midwestern academic center of all women who delivered from 2009-2018. The primary outcome was hypertension-related postpartum readmission within 42 days postpartum. We used a cost-sensitive random forest method to determine predictors of postpartum readmission.

One hundred seventy were readmitted due to a hypertension-related diagnosis. Random forest method achieved balanced accuracy 76.61% for predicting readmission. The most important variables for predicting readmission were blood pressures in labor and 24-48 hours postpartum.

Timing of blood pressure measurements during labor through 48 hours postpartum can be combined with other variables to predict women at risk for postpartum readmission.

N/A