Healthcare Analytics: Creating a Prioritized Improvement System with Performance Benchmarking.

Area of Science:

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

Background:

• Healthcare improvement is critical for patient outcomes and cost reduction.
• Existing performance benchmarking methods require enhanced analytical approaches for prioritization.

Purpose of the Study:

• To develop and implement a prioritized improvement system using performance benchmarking.
• To leverage analytics and modeling for assessing departmental performance metrics.
• To translate data-driven insights into actionable, enterprise-wide improvements.

Main Methods:

• Collaborative development of a prioritized improvement system with Seattle Children's experts.
• Application of analytics and modeling to U.S. News and World Report benchmarking data.
• Comparison of diverse performance metrics: patient outcomes, structural/process metrics, survival rates, clinical practices, and subspecialist quality.
• Development of a predictive model using simulated transformations and imputation methods.

Main Results:

• A predictive model achieved high accuracy with average rank correlation of 0.98 and average score correlation of 0.99.
• The system effectively translates data into prioritized departmental and enterprise-wide improvements.
• Demonstrated a data-to-knowledge-to-outcomes paradigm for sustainable healthcare enhancement.

Conclusions:

• The developed prioritized improvement system, utilizing performance benchmarking and predictive modeling, is effective.
• This data-driven approach facilitates significant improvements in patient care and cost reduction.
• The methodology offers a scalable solution for addressing diverse healthcare challenges.