You are here
Computer Analytics To Reduce Hospital Readmissions
The Baltimore score (B score), a novel machine-learning model developed by researchers at the University of Maryland Medical System (UMMS), may help hospitals better predict which discharged patients are likely to be readmitted.
Using the new predictive score, researchers analyzed the data of more than 14,000 patients from three hospitals to determine their likelihood of being readmitted.
The research, published in JAMA Network Open, may help improve patient care and reduce the number of hospital readmissions.
Almost 20% of U.S. patients are readmitted to hospital, increasing the risk for harm—such as infections, falls, and delirium––as well as increasing expenses. Previous studies indicate that clinicians are not well equipped to identify which patients will be readmitted, and many readmissions are believed to be preventable.
By better targeting time and money toward the goal of appropriate discharge, hospitals could lower the number of patients who are forced to return. Also, rates of unplanned readmission within 30 days after discharge are used to gauge a hospital’s performance and quality of patient care.
Existing tools for assessing readmission risk consider limited variables for each patient: length of hospital stay, type/severity of admission, medication types/amounts, other possible chronic conditions, and previous hospital admissions.
The experimental B-score algorithm was individualized for each hospital in different settings, and the final model drew from 382 variables, including demographics; lab test results; whether the patient required breathing assistance; body mass index; affiliation with a specific church; marital status; employment; medication usage and substance abuse.
Researchers then compared the B-score readmission-risk ranking to actual readmissions at the three hospitals, and to the predictions scored by other programs, including the LACE index, HOSPITAL score, and Maxim/RightCare score.
Despite the different settings, the B score was better able to identify patients at risk of readmission than the other scores. It was particularly accurate among highest-risk patients. Patients in the top 10% of B-score risk at discharge had a 37.5% chance of 30-day unplanned readmission, and patients in the top 5% B score had a 43.1% chance of readmission.
Source: MedicalXpress, June 5, 2019