Hospitals are finding that predictive modeling programs are identifying at-risk patients successfully and keeping them from returning to their facilities unnecessarily, HealthLeaders Media reported.
For instance, Dallas' Parkland Health and Hospital System uses an algorithm based on physiologic, laboratory, demographic and utilization data extracted from a patient's electronic medical record within 24 hours of hospital admission to predict which heart failure patients have high risk for readmission or death. So far, the predictive model algorithm has led to 33 percent drop in readmission of Medicare heart failure patients and 20 percent decline in readmissions for all heart failure patients.
Indianapolis-based Community Physician Network is deploying similar technology to improve care and lower costs by identifying high-risk congestive heart failure patients before they are admitted.
Meanwhile, Mount Sinai Medical in New York City is using admission history data to identify at-risk patients. "The higher the score, the higher the risk of readmission," Jill Kalman, director of cardiomyopathy, told HealthLeaders Media. Thanks to the predictive modeling program, the health system's 30-day readmission rate dropped from 30 percent to 12 percent, while emergency department visits fell by 63 percent (over three-plus months), according to the article.
"We wondered if modeling readmissions was going to require us to use more data and create a complex score, but we're validating that a simple [admission history] approach works, and we believe it can be set up easily, regardless of where [an organization] is located, its size or the level of IT support," Kalman added.
In the Chicago metropolitan area, NorthShore University HealthSystem, Northwestern Memorial Hospital and Northwestern Memorial Faculty Foundation also are embracing predictive modeling technology, reported Crain's Chicago Business.
NorthShore has invested $7 million to establish its Center for Clinical and Research Informatics and hired 20 physicians, computer scientists and statisticians with predictive-modeling proficiency. With the multi-million investment, the hospital system expects to save at least $500,000 a year by only targeting patients identified as most likely to suffer complications from pancreatic surgery or be readmitted, according to the article.