#hasummit14 session #12: sneak peek: improving patient engagement and outcomes with predictive...
TRANSCRIPT
#HASummit14
Session #12:Sneak Peek: Improving Patient Engagement
and Outcomes with Predictive Analytics
Pre-Session Poll QuestionDoes your organization currently share predictive analytic results with patients?
a) Yesb) Noc) Unsure or not applicable
Lou CervoneDirector of Business IntelligenceCrystal Run Healthcare
Gregory Spencer, MDChief Medical OfficerCrystal Run Healthcare
#HASummit14 2
Our Organization
Physician-owned MSG in NY State, founded 1996
350+ providers, 30+ locations 40,000 commercial lives at risk 12,000 attributed beneficiaries Joint Venture ASC, Urgent Care,
Diagnostic Imaging, Sleep Center, High Complexity Lab, Pathology
Early adopter EMR (NextGen®) Accredited by Joint Commission Level 3 NCQA PCMH Recognition
#HASummit14 3
The Pain Point
How can we leverage the capabilities of our analytics platform to better engage and activate patients?
Can existing data be used interactively to help modify patient behavior?
#HASummit14
The Concept of a Flight Path
Example
Diabetes Cohort
Good Profile
Poo
r P
rofil
e
1. A1c < 72. LDL < 1003. BP < 130/80
1. A1c > 7 2. LDL > 100 3. BP > 130/80
$ COST per member per year (charges)
For > 1 year of encounters
(disease specific)(5 years, 26k patients)
4
#HASummit14
Poll Question #2
How are you currently engaging patients using data?
a) We actively use predictive analytics with patients to show them the predicted impact of their lifestyle choices
b) We share lab values and results with patients and verbally counsel them on long-term health implications of lifestyle choices
c) We share results only
d) Unsure or not applicable
5
#HASummit14
Our ApproachAnd
Results
6
#HASummit14 7
Our Goal
Using disease-specific: metrics, costs, analytics, simulation, and predicted outcomes… to engage both the
patient and clinician in more efficient diabetes care.
Our Approach
#HASummit14
What factors can predict “health”?
Lab Values Complications Risk Scores
Family History Substance Use Demographics
10
#HASummit14 9
Flight Path – Risk
Risk Prediction
Given everything we know about the patient, what is his expected “risk score”?
“What-if” analyses
Sliders will show how changing X (e.g., BMI) will affect the overall risk score
#HASummit14 10
Flight Path – Future complications
Predict the likelihood of developing one of 14 diabetes-related
complications and display the “next most likely”
Possible complications Ranked by “next likely”
Cataracts
Coronary Artery Disease
Diabetic Ketoacidosis
Diabetic Retinopathy
End Stage Renal Disease
Glaucoma
Peripheral Neuropathy
#HASummit14 11
Flight Path – Recommendations
• Compile list of recommendations for each complication
• Calculate recommendation score
• Sort recommendations from highest to lowest
• Present in both a patient view and clinician view
Likelihood of
developing
complication
Complication
Severity
Recommend Impact
Recommend Score
#HASummit14 12
Flight Path – Recommendations
Categorize recommendations by type/theme to facilitate patient’s
ability to process and remember
#HASummit14
Leveraging Predictive Layers
11
#HASummit14
Poll Question #3
Based on what you’ve seen, is this something you could envision implementing in your organization?
a) Yes
b) No
c) Unsure or not applicable
13
#HASummit14
Expected Results/Measurable Analytics
14
• Gain patient understanding of the life choices and things within their control that can impact their potential clinical outcomes
• Show measurable improvement in patient engagement and clinical outcomes
• Inform future application development
#HASummit14
Future Plans
• Deploy application into clinical areaso Endocrinology, primary careo Diabetic nurse educator
• Evaluate effectivenesso Follow the cohort that has used the toolo Follow cost, Hemoglobin A1c, quality measure complianceo Patient Activation Measure (PAM)?
• Begin work on heart failure and subsequent additional applications
15
#HASummit14
Lessons Learned
1. Select your clinical conditions carefully
2. What you learn informs future applications and saves time
3. Manage the data
4. Decide on time parameters and how to treat values over time
5. Consider how the data is to be displayed for best effect
16
#HASummit14
Analytic Insights
AQuestions &
Answers
18
#HASummit14
Choose one thing…
Write down one thing will you do differently after hearing this presentation
19
#HASummit14
Thank You
20
#HASummit1421
Session Feedback Survey
1. On a scale of 1-5, how satisfied were you overall with this session?
1) Not at all satisfied
2) Somewhat satisfied
3) Moderately satisfied
4) Very satisfied
5) Extremely satisfied
2. What feedback or suggestions do you have?
#HASummit14 22
Upcoming Speakers
3:45 PM – 4:35 PM
16) Delivering Excellence at Stanford Health Care
Amir Dan Rubin, President and CEO, Stanford Health Care
4:35 PM – 5:00 PM
17) The Future World of Value-Based Healthcare (Documentary featuring Michael Porter)
Caleb Stowell, MD, Vice President, Research and Development, International Consortium for Health Outcomes Measurement (ICHOM, Senior Researcher, Harvard Business School)
Location
Grand Ballroom
Grand Ballroom