Big Data: How Can it Guide Practice and Health Care Reform Eric D. Peterson, MD,MPH Executive Director, Duke Clinical Research Institute April 2015
The Best of Times The Worst of Times Science and IT are advancing at incredible speeds, but: Yet Poor transition of bench to bedside Lack of evidence for much of what is done in medicine Slow adoption of useful therapies Medical costs rising exponentially All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 2
Years 3 Pharma s compound translation failures 6 9 12 15 Source: PhRMA, Industry Profile All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 3
Evidence- or Ignorance-Based Medicine CV Guideline recommendations based on RCTs AF Heart failure PAD STEMI Perioperative Secondary prevention Stable angina SV arrhythmias UA/NSTEMI Valvular disease VA/SCD PCI CABG Pacemaker Radionuclide imaging Tricoci P et al JAMA 2009 0.3% 6.4% 6.1% 3.5% 4.8% 11.7% 15.3% 13.5% 12.0% 9.7% 11.0% 19.0% 22.9% 23.6% 26.4% 0% 10% 20% 30% All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 4
Slow and Varied Adoption of Evidence It takes an average of 17 years for 14% of original research findings to lead to changes in care that benefit patients Ballas E & Boren S. Yearbook of Medical Informatics: Patient Centered Systems. 2000:65-70. MI Care at 430 US Hospitals Peterson et al, JAMA 2006;295:1863-1912 All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 5
Rising US Healthcare Costs and Questions about Value? http://ucatlas.ucsc.edu/spend.php All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 6
Harvesting Big Data Turning Big Data into Knowledge Turning Knowledge into Practice/Policy All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 7
How Big is Big Data Just in 2010, 4 exabytes (4 x 10 18 ) of unique data was generated, which was more than in the preceding 5,000 years altogether "Every 2 days we now create as much information as we did from the dawn of civilization up to 2003 (Eric Schmidt, Google CEO) The amount of new data is now doubling every 13 months -and will soon double every 12 hours according to IBM For college students in technical degree, half of what they learn in their first year of study will be outdated by their third year Brett King, Huff Post Tech; June 4, 2014 [http://www.huffingtonpost.com/brett-king/too-much-content-a-world-_b_809677.html]; Ray Kurzweil, The Law of Accelerating Returns; March 7, 2001 [http://www.kurzweilai.net/the-law-of-accelerating-returns]; David Russell Schilling, Knowledge Doubling Every 12 Months, Soon to be Every 12 Hours; industry tap into news, April 19th, 2013 [http://www.industrytap.com/knowledge-doubling-every-12-months-soon-to-be-every-12-hours/3950]; http://www.youtube.com/watch?v=9xxazrhhmxy All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 8
Data Driven Medicine: Not a New Concept Collecting Data, Asking Questions, & Translating it to Practice Chronic diseases can be studied, but not by the methods of the past. If one wishes to create useful data computer technology must be exploited. Eugene Stead, MD 1969 Led to the concept of computerized textbook of medicine Formed foundation of the Duke Databank for CV Diseases Spurred a generation of clinical and quantitative researchers All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 9
Value of Aggregated Medical Data Database Size Examples Clinician s Experience 10-100 s Osler, Harvey Single Center Database 1,000-10,000 s Duke, Emory Epidemiological Cohort 10,000 s Framingham, MESA National Registries 1,000,000 s AHA, ACC, STS EHR / patient-powered Registries 10,000,000 s PCORnet, Health eheart registry All Rights Reserved, Duke Medicine 2007
Recent Transformations in Health IT EHR Adoption mhealth Remote Monitoring All Rights Reserved, Duke Medicine 2007
The New Era of Precision Science January 2015 All Rights Reserved, Duke Medicine 2007
Baseline Study A comprehensive study of human health and the transition to disease A longitudinal cohort study to extensively characterize participants at baseline and serially using a battery of clinical, imaging, psychosocial, behavioral, socioeconomic, geospatial, physiometric, and molecular tools. All Rights Reserved, Duke Medicine 2007
Baseline: Human Health and Transition to Disease All Rights Reserved, Duke Medicine 2007
Too Little or Too Much Data All Rights Reserved, Duke Medicine 2007
IT and Healthcare Promise and Challenges DATA INTEGRATION ANALYTICS ACTION 90% of world s data has been created in the last 2 years. But 80% of that data is unstructured and stored in separate systems. Big data analytics is coming to medicine: Google, IBM Watson Will need to integrate these data into practice. SCALABILITY FRAGMENTATION DECISION SUPPORT LEARNING SYSTEMS All Rights Reserved, Duke Medicine 2007
Evolving Informatics for EHR- based Clinical Research Network Research Site A EHR Internal Data Warehouse Research Datamart Study specific Clinical Research Network Research Site B EHR Internal Data Warehouse Research Datamart Research Site C EHR Data Warehouse Research Datamart Clinical Study Database Centralized disease registry EHRs can contribute some basic data to CT database All Rights Reserved, Duke Medicine 2007
PCORnet- Clinical Research & Patient Engagement on Steroids Clinical Data Research Networks $56 million 8 networks 1 million pts/network Patient Powered Research Networks $12 million 18 networks Selby JV et al. Sci Transl Med 2013;5:182fs13 All Rights Reserved, Duke Medicine 2007
The Missing Link: Those Who Can Use Big Data Harvard Business Review Oct 2012 All Rights Reserved, Duke Medicine 2007
Entering a Age of Big Data Wonder! All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 20
All Rights Reserved, Duke Medicine 2007 - omics will help redefine human disease and treatment targets
22 Post Market Safety Surveillance: Mini-Sentinel 150,000,000 people s claims and pharmacy data 27 Institutions, 200 experts, 1 CC > 60 million people FDA, The Sentinel Initiative July 2010 All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 22
Prediction Tools for Risk of Death after Stroke (circa 2012) Based on 900 US hospitals and 1 Million patients Smith EE. Circulation. 2010;122:1496-1504 All Rights Reserved, Duke Medicine 2007
The Future of Clinical Trials Trial-Registry Hybrids EMR-Trials Cluster RCTs Patient-centric Trials All Rights Reserved, Duke Medicine 2007
Using EHR/Registries to Support RCTs Study planning: #Events/ Inclusion Criteria EHR/Registry Clinical Data Backbone Site/Investigator Identification Patient Identification + Evaluate Generalizablity Concurrent Data Collection Patient Follow-up All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 25
Swedish Registry-Trial Hybrids TASTE Trial: Thrombus-Aspiration in MI IOM, November 27, 2012 All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 26
Taking Research to the Patient Apple Research Kit All Rights Reserved, Duke Medicine 2007
Implementation Innovation Using Data + Behavioral Strategy + Policy to Transform Care Practice Models Learning to promote the rapid and complete uptake of clinical research findings into routine practice, leading to improved quality of health care and outcomes. Bench Patients Populations 1 2 First Block: Translation from concept into first human studies Second Block: Translation from clinical trials into practice 28
The Power of Feedback and Quality Improvement Provider-led feedback and QI can improve CV care! GAP, NRMI, CRUSADE AHA GWTG ACC-NCDR STS Means to achieve better care Motivated advocates Timely, valued feedback Simple tools Collaborative Teams Concept Outcomes Safe, Effective, Long-term Use Clinical Trials Provider Led Quality Improvement Measurement Guidelines Performance Indicators All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 29
Implementation Research Site Screening: Use National Clinical Registries Rapid screening of 100 s of centers to find outlier performance Qualitative/Quantitative Research: Identify care processes linked to better outcomes Empirical Evaluation and QI Formally test using cluster RCTs Disseminate what works through system! All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 30
Impact of Target Stroke: Care & Outcomes All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 31
Mapping Disease in Durham County Top layer shows concentrations of diabetes patients. Next layer down percentage single female head of household. Below that in purple, another indicator of economic status. The bottom layer maps the county boundary and streets. Vertical green spines longitude ad lattitude coordinates of where diabetes patients live and locations of key social or commercial institutions, that can be used to link all of these disparate data sets together based on shared geography. Miranda, Ferranti, Strauss, Neelon, Califf. Health Affairs 2013;32:608-1615 All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 32
Check-It Change-It: AHA-DCRI community implementation treat hypertension Patient OnBoarding Patient Engagement Provider Engagement Clinic Provider Community Clinic Provider and/or Kiosks Pharmacy Kiosks and/or Pharmacist Community Center Employer Intervention Patient Coach Website Kiosks Health Assessment Kiosks Campaign Website Channel Patient empowered to begin disease management behaviorstracking health factors, medications, lifestyle changes; receiving patient education; and partnering with their provider. Provider Connection enabled regardless of recruitment channel. Motivational/Engagement Effort Improved Standard of Care Confidential American Heart Association 33 All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 33
Conclusion: The Future of Big Data, Practice and Policy In the past, clinicians were limited by existing technologies and therapeutic options. Today, our clinical, research & IT capacities can drive amazing progress. However, progress will not occur without the development of novel IT and informatics, more efficient research, and more rapid and effective learning health systems We need to harness the emerging data deluge to create new knowledge and translate this into better care All Rights Reserved, Duke Medicine 2007 sb/strategy & Innovation Group 34