Kaiser Permanente is recognized as one of America’s leading healthcare providers and not-for-profit health plans. Currently serving more than 10 million members, KP is a trusted partner in total health—collaborating with members to help them thrive and create communities that are among the healthiest in the nation.
Analyzing Risk Among Disparate Data Sources
Kaiser Permanente is focused on delivering the best health care possible at the most affordable cost. A key factor in reducing costs and improving outcomes is determining readmission risk for patient discharges and predictively classifying readmissions as potentially preventable or not preventable. Potentially preventable readmissions can be further sorted into categories and tracked over time and across hospital units to support improvements and develop organization-wide best practices. Classifying readmissions as potentially preventable or not preventable can also be used to establish accountability.
To determine readmission risk and develop ever-better discharge protocols over time, Kaiser Permanente needed to be able to seamlessly access high volumes of both new and existing data from disparate sources and analyze it for patterns and trends.
Easy Access to Insights
Zaloni developed a Hadoop-based data lake for Kaiser Permanente using its Bedrock data management platform. The data lake housed more than 60 million records of historical patient data, enabled real-time data ingestion of new records, as well as updates to existing records. More importantly with Bedrock, Kaiser Permanente had a clear view of what data was in the data lake; could track its source, format and lineage; and enable users to search, browse and find the data they needed for analytics.
Specifically, the solution:
- Ingested all existing data into Hadoop
- Prepared all data, including patient information, encounter data (including ICD-9 codes), ICD-9 to clinical classifications software (CCS) mapping data
- Consolidated data from various sources, including institutional, ambulatory, skilled nursing, home health/hospice, dialysis, and ambulance transport encounters into a single view for processing
- Transformed and mapped the data so that the output datasets could be used for creating predictive models
Reduced Costs and Lower Readmission Rates
Zaloni’s solution enabled Kaiser Permanente to reduce costs and improve outcomes by ultimately lowering readmission rates, due to:
- Significantly improved understanding of potentially preventable readmissions and not preventable readmissions, which supported development of protocols for typically high-risk cases
- Ability to develop better algorithms to more accurately predict readmissions
- New ability to identify patients with the highest risk of readmission early in their initial hospitalization and proactively adjust treatment plans sooner to account for that risk