Advanced Data Analytics for Improved Decision-Making at a Veterans Affairs Medical Center

Area of Science:

• Healthcare Management
• Health Services Research
• Data Analytics in Medicine

Background:

• Veterans Affairs medical centers (VAMCs) utilize Strategic Analytics for Improvement and Learning (SAIL) reports for data visualization and trend analysis.
• Improving patient outcomes, specifically the 30-day standardized mortality ratio (SMR30), is a key objective for VAMCs.

Purpose of the Study:

• To present a data-driven methodology for decision-making to enhance patient outcomes at a VAMC.
• To identify key performance indicators correlated with SMR30 and model the impact of targeted interventions.

Main Methods:

• Utilized over 4 years of VAMC data, including 17 quarters of SAIL report information.
• Employed correlation algorithms to identify factors most closely linked to SMR30.
• Developed a model to predict the necessary reduction in in-hospital complications (IHC) to achieve a 1% decrease in SMR30.

Main Results:

• In-hospital complications (IHC) were found to be the most significant factor correlated with SMR30.
• A 1% reduction in SMR30 necessitates an 18.6% decrease in IHC.
• Potential positive outcomes include improved methicillin-resistant Staphylococcus aureus mitigation, while potential unintended consequences on catheter-associated urinary tract infections and patient safety indicators require monitoring.

Conclusions:

• This methodology enables healthcare leaders to make informed decisions by anticipating both positive and unintended consequences of interventions.
• The study highlights the need for continuous monitoring and discussion among stakeholders regarding resource allocation and progress tracking.
• Implementing a weekly progress dashboard, rather than relying solely on quarterly SAIL reports, is recommended for more agile management.