Unified Health System
Identifying Risk Factors for Re-admissions with More Complete Data
Our client is a unified health system with more than 100 locations and 11,000 employees.
80% Of Data Was Unavailable to Existing Systems
Per the Center for Healthcare Quality & Payment Reform, “one of the best ways for communities to reduce healthcare costs quickly and improve patient care in the process is to implement initiatives to reduce hospital readmissions.”
The U.S Department of Health & Human Services estimates that avoidable hospital readmissions make up more than $17 billion in Medicare expenses. To help address these rising expenses, the Hospital Readmissions Reduction Program adjusts Medicare payments for hospitals with higher-than-expected 30-day readmission rates for targeted clinical conditions such as heart attacks, heart failure, and pneumonia. By reducing payments, hospitals and healthcare providers are financially incented to improve the quality of care and identify patients or cases with a high likelihood of readmission.
To meet the goals set by the Hospital Readmissions Reduction Program, this unified health system needed a better way to predict the propensity of a patient to readmit. Unfortunately, 80% of data was invisible to legacy systems because it was unstructured. Being able to uncover insights in that data would prove invaluable when predicting triggers for readmissions.
After identifying readmissions indicators, the health system would be in a better position to develop action plans to improve clinician and patient engagements.
Avoidable Hospital Readmissions Make Up More Than $17 Billion In Medicare Expenses Nationwide
Unlocking data helps predict readmissions
We developed a predictive modeling solution that predicts readmissions across all diseases and conditions. It also looks at the psycho-social needs of the patient and the family rather than the traditional clinical approach, which is the primary industry focus. By incorporating this data with a wealth of unstructured and structured healthcare data, healthcare providers at our client can significantly increase the accuracy of their readmissions modeling and predictions.
The solution leverages a cognitive platform to uncover new evidence in predicting readmission propensity, ushering in a new era of evidence.
Unlocking data helps predict readmissions based analytics. Combining insights from our cognitive exploration and content analysis platform and from healthcare annotators, the data is used to analyze physician notes from the EMR to extract relevant, contextual data and transform that information into structured data points. The healthcare annotators are “reading” this unstructured information, tuned appropriately to best interpret and transform the organization’s data. The analysis result is exported as structured data into a data warehouse, which is then consumed by a predictive modeling tool along with other existing structured data to develop the predictive readmissions model. The resulting readmission risk indicator is then incorporated into the EMR system, indicating a patient’s likelihood of readmission at the point of care.
Results
Providing better care – that’s what it’s all about
The healthcare system is experiencing benefits on two primary fronts: improved patient care and enhanced predictive modeling.
Improved Patient Care
The health system is better able to identify patients who are most in need of interventions, identify which interventions are most appropriate, and begin the intervention process as quickly as possible.
Early intervention gives clinicians more time to provide instruction, increase patient and family member understanding, and allow time for community services to be arranged so they are ready when the patient is discharged.
Ultimately, the solution is designed to reduce hospital readmissions across the entire patient population by educating patients and their families to better manage their conditions while coordinating various services within the community.
Enhanced Predictive Modeling
Tuned cognitive healthcare annotators help extract key information related to patients’ psychosocial factors, such as compliance and demographics.
The psychosocial factors extracted were identified within the top 10 predictors for readmission.
Before the solution, the health provider’s readmissions prediction accuracy was 46%, which was significantly below its goal of 70%. After implementing a cognitive predictive model, that rate jumped to 71%. With cognitive analysis and accounting for unstructured data, the rate surpassed the goal and reached 93%.
