Data Analytics:Powering Clinical and Operational Excellence at the Cleveland Clinic
January 15, 2019
Craig Henderson
Manager, Business Intelligence
Agenda
• Introduction to Cleveland Clinic
• Analytics Strategy- Pillars of analytics
• Protocol Adherence & Effectiveness- Clinical Care Paths
• Emergency Services Case Study- Uncovering trends
• Patient Access Case Study
• Measuring success
• Founded in 1921
• 7.1 million patient visits
• 165,000 admissions
• 207,000 surgical cases
• 550,000 ED visits
• 4,435 inpatient beds
• 150 outpatient locations
• 51,000 Caregivers
• 3,000+ physicians and scientists
FHC
Elyria Family
Health CenterWestlake Family
Health Center
FHC
Lorain and Avon
Family Health
Centers
FHCLakewood
Family Health
Center
FHC
Strongsville Family Health
and Surgery Center
FHC
Brunswick Family
Health Center
FHC Wooster Family
Health Center
FHC
Independence Family
Health Center
FHC Solon Family
Health Center
FHC
Chagrin Falls Family
Health Center
FHC
Beachwood Family
Health and Surgery Center
FHC
Willoughby Hills
Family Health Center
FHC
HFairview Hospital
Avon Hospital
H
Lutheran Hospital
H
Ashtabula County Medical Center
H
Euclid Hospital
H
Hillcrest Hospital
H
South Pointe
Hospital
H
Marymount Hospital
H
Huron Health Center
FHC
Twinsburg Family Health Center
FHC
Cleveland
Clinic
Akron General Hospital
H Lodi Hospital
H
888 W Bonneville Ave, Las Vegas, NV
The Cleveland Clinic Lou Ruvo Center for
Brain Health, opened on July 13, 2009 in
Las Vegas, Nevada. It is operated by the
Cleveland Clinic and was designed by the
world-renowned architect Frank Gehry.
Cleveland Clinic Mission, Vision, Values
Our Mission:• To provide better care of the sick, investigation into their problems, and further
education of those who serve.
Our Vision:• Striving to be the world’s leader in patient experience, clinical outcomes,
research and education.
Cleveland Clinic – Strategic GoalsPatients
First Caregivers Affordability Growth Impact
Analytics Strategy
Enterprise AnalyticsBusiness Intelligence
• 70 caregivers responsible for clinical data warehousing and reporting
• Report through Medical Operations (not IT)
• Enterprise department; open to all 51k Caregivers
• Clinical reporting, research, web solutions, dashboard analytics
• 100+ data sources, 75TB
Commitment Flow Down
Enterprise Goals:
5 Pillars
Institute/Hospital
Department
Frontline
Performance Management
Monthly Executive Review/Strategic Initiative
Dashboards
Institute/Hospital Analytics
Center/Service Line
Frontline Management
Institutional Alignment
Enhance the value of CC data assets
Develop an analytically
capable organization
Promote standard processes and a culture that value
analytics over instinct
Create a modern analytics platform
DATA PEOPLE
PROCESS Technology
Comprehensive Analytics Strategy
Data Integration &
Quality
• Data Governance &
Stewardship Support
• Data Certification Capability
• Master Data Management
• Data Quality Capability
(Profile, Scorecard)
• Data Integration (ETL)
• Data Lineage and Meta-
data available
Data Aggregation &
Management
• Role-based Flexible
Security
• Efficient Storage of Large
Volumes of Data (structured
and unstructured)
• Quick Response Data
Retrieval
• Patient Entered Data
• Clinical Vocabulary
(Semantic) Support
Data Visualization,
Interrogation & Analytics
• Self-Service Capabilities
• Advanced Analytics &
Visualization Capabilities
• Data Exploration &
Discovery
• Support Structured and
Dashboard-style Reporting
• Geospatial Capability
• Simulation & Optimization
• Natural Language
Processing
How to meet the varying needs of our customers
Architecting the Platform
Analytics Challenges• Multiple audiences
- Executive, manager, analyst, statistician
• Internal and external sources
• 300+ analytics requests per month
• Accessibility to understandable information- Right accessibility for the right
person
- Summary vs detail
Sophis
tication
Ability to
Produce
Ability to
Consume
Maturity
Make analytics easier to consume
Increasing
understanding of
analytics
Analytics Platform
• Analytics quicker to market• Decrease visualization development
• Increase creators
• Executive usability complimented by deep drill
• Shift to self-service
• Empower overall data analytics
Enterprise Data
Sources
People
Any Data
Anywhere
Desktop Tools Analytics
Consumers
Business Objectives
Self-Service Capability
Protocol Adherence & Effectiveness
Creating a Path to Improved Patient Care
Process Management
Improvement for Healthcare• Optimize patient care
• Reduce practice variation
• Follow established guidelines
• Improve outcomes
• Use of the EMR
• Lower overall care costs
Patient Experience
Quality
Safety
High Reliability
Efficiency
Value
Care Path and EPIC
Care Path Dashboards
Summarized data
Detail
Flexibility
Trends
Comparisons
Measure impact• Why are we doing this?
• What is the impact?
Measureable Impact
60%35%
• Referral to physical therapy
• Imaging studies
• Use of opioids
Cost Savings Spine
• Less images ordered
Patient Care
• Less orders for opioids or tramadol
• Spine Care Path
• Cost variables in patient management
• Improved outcomes
• Total Hip/Total Knee Replacement• Elevated hemoglobin
• Diabetes, heart disease, anemia
• Appropriate antibiotics
46%
• 30 day readmissions
• 90 day reoperations
• Discharge to home
Patient Care
• More patients discharged to home
Cost Savings
• Less nursing home days
Measureable Impact
COPD Care PathAdherence & Effectiveness
Measure (Aug 2017) Care Path Non Care Path
Rate of LABD 86% 58%
Rate of follow up appointment 99% 78%
Rate of completed appointment 40% 32%
Cost per case reduction $1,285
“Clinical Care Pathways are one of the most powerful tools ensuring high quality, lower cost care.
Standardization and powerful visualizations provide insight into practice patterns and meaningful outcomes allowing us to make real change.”
• Dr. Timothy CroneMedical Director, Enterprise Business Intelligence & Enterprise Analytics
Break for Questions
Emergency ServicesCase Study
Driving Institute PerformanceProblem Statement:• 15 emergency departments using EPIC ASAP
• Accessibility to encounter information supplemented with ancillary services
Current Conditions:• Wealth of data
• Frequent flyer of analytic requests
Solution:• Partnered Governance
• Technology enablement
• Self service analytics
Emergency Services Data Model
ESI AnalyticsExternal
files
Decentralized data
Staging
Oracle
SQL Server
EPIC
ASAP Inpatient
ResultsDashboards that provide value:• Data replaces anecdotal evidence
• Visualizations highlight opportunities and drive performance improvement
• Transparency and comparative analysis – promotes friendly competition
Use Cases:• Monitoring of Call Back program
• Imaging and Medication Turnaround Time - Reducing ED LOS
Use Case: Call Back Program
Use Case:• Call patients post discharge to assess
wellbeing
• Address questions / follow-up
Current Conditions:• Captured in EPIC flow sheet
• Leadership can track call back performance
Measurable Impact:• Increased patient satisfaction
• 25% increase in call back rate
• 33% of ED’s surpass 75% target
Use Case: Imaging Turnaround TimeUse Case:• Turnaround time increases ED
LOS
Implications:• Lower patient satisfaction
• Higher rate of Left Before Treatment
• Increased “boarding”
Measureable Impact:• Dashboard used in Culture of
Improvement model areas
• ED’s reducing TAT in focused cohorts
• Identification of outliers
“Tableau dashboards have produced downstream benefits; we are able to more quickly and efficiently meet the demands of our customers.”
• McKinsey MuirSenior Operations Analyst, Emergency Services Institute
Patient Access
No Shows/Same Day CancelsOverall Impact
• Negatively Affected Patient Care
• Artificial Access Problems
• Office Productivity
• Lost Revenue
1%
DECREASE
Summary Dashboard
No Shows & Same Day Cancellations
Use Case: Lorain & TransportationUse Case:
•High percentage of chronic no shows/same day cancels
•Patients cancel due to transportation
38% in Lorain vs
19% overall
Use Case: Lorain & TransportationUse Case:
•74% live within 10 miles of the Lorain FHC
Implications:•Higher health costs
•Higher acuity health issues
•Low patient/provider satisfaction
•Lower productivity
Measurable Impact:•Pilot program initiation
Creative SolutionSame Day Cancel• Patient calls to cancel appointment
• Patient Service Representative asks for reason
• If reason is transportation offer ride program
Forecasting Future Access
• Forecast - Available
- Appointments scheduled
• Variables- Medical specialty
- Seasonality
- Day of week fluctuation
- Historical trends
Measuring Success
Implementation is a journey…
Divert ad hoc and project requests:• Analytics delivery up 146%, Distinct users up 54%
• 86% increase in projects of “High” complexity by enterprise resources
Continue self-service direction:• Increase data accessibility through Enterprise Data Vault
• Certify and publish data sources to Tableau Server
• Build Power User community for best practice sharing
• Implementation of Analyst University – increase Analyst acumen
Conclusion
Questions
https://www.linkedin.com/in/craig-henderson-0430907/