PeraHealth Insights

Our Collaborative Effort to Predict and Reduce Pediatric Readmissions

As part of PeraHealth’s Strategic Initiatives team, it was my pleasure to host the webinar, “Predicting Unplanned Readmissions with the Pediatric Rothman Index,” on July 19… especially since one of our top research priorities is predicting and reducing unplanned readmissions.

In anticipation of their upcoming publication in Hospital Pediatrics, I interviewed nationally-recognized clinical informaticist and research partner Dr. William Feaster, Chief Medical Informatics Officer at Children’s Hospital of Orange County (CHOC), and Louis Ehwerhemuepha, PhD, a lead data scientist who helped develop CHOC’s readmissions model. Their article, “A Novel Model for Enhanced Prediction and Understanding of Unplanned 30-day Pediatric Readmission” will detail breakthrough findings to help save children’s lives.

4 Key Takeaways

1. Reducing 7-Day Readmission Rates

Dr. Feaster explained how CHOC’s persistent challenge has been reducing its 7-day readmission rate, as it hovers around 3.5 percent, the national average. Implementing typical best practices are not working to lower the rate. Therefore, the CHOC team decided to work with PeraHealth and build the “most predictive model for readmission under the most stringent set of inclusion/exclusion criteria.” With an active inpatient case mgmt. program in place and population health staff on board, CHOC was “primed to have an effective team.”

2. Validity of the Pediatric Rothman Index (RI)

Dr. Ehwerhemuepha shared how the team began with the assumption that PeraHealth’s pediatric Rothman Index (pRI) is a valid measure of patient condition, because “it has been proven.” In fact, in my introductory webinar comments, I explained how, at PeraHealth, we even opened our model to the FDA in order to obtain FDA 510(k) Class II clearance for PeraTrend, one of our clinical surveillance solutions based on the Rothman Index (RI). (If you’re interested in the “making of the pRI” story, Michael Rothman, Ph.D., PeraHealth’s Chief Science Officer, details how the PeraHealth team developed and validated the pRI in his recent blog, (More than body mass – The clinical surveillance difference in children vs. adults.)

3. CHOC’s Model

Starting with the pRI variable, the CHOC team added demographics and socioeconomic data, resource utilization, disease severity, and disease diagnosis to its model. In the analysis, study results showed the higher the pRI, the less likely patients will be readmitted. Dr. Ehwerhemuepha shared, “We got an area under the curve (AUC) of 0.79, which is, to the best of our knowledge, the highest achieved for pediatrics with the exclusion of chemotherapy patients.” Other variables shown to impact readmissions were “length of stay” and the “number of ED visits in the last six months.”

“We got an area under the curve (AUC) of 0.79, which is, to the best of our knowledge, the highest achieved for pediatrics with the exclusion of chemotherapy patients.”

4. Improving Clinical Workflow

CHOC has put these findings in practice within its clinical workflow. On a daily basis, the research team sends real-time predictions of readmission risk, providing a “risk of readmission” score to the care management team for every patient in the electronic medical record (EMR). The team then implements agreed-upon interventions.

It’s exciting for me to see results from CHOC’s innovative approach. In my role with PeraHealth and the Rothman Society for Innovation and Research (RSIR), I’m helping expand our pediatric research.

Please let our team know if you are interested in exploring participation by submitting a contact request. Be sure to indicate interest in the pediatric readmissions research initiative.

CHOC has set a great example of addressing a common challenge, creating a unique model, conducting a peer-reviewed study, implementing meaningful change, and helping spread lessons to other hospitals. If you missed our webinar with CHOC, watch the full on-demand version here.

We look forward to sharing the Hospital Pediatrics publication as soon as it is released, and we welcome your interest in the next phase of pediatric readmission research.

We invite you to join our next webinar, “Proactive Rounding Based on Clinical Surveillance: A Team Approach for Rapid Response and Better Patient Outcomes,” when Carilion Roanoke Memorial Hospital and Penn State Health Milton S. Hershey Medical Center will share innovative approaches for using the Rothman Index to integrate proactive rounding with clinical workflows.

September 6, 2018 | 2 p.m. ET. | REGISTER