Bringing Big Data to the Point of Care by Creating the App Store for Health @mandl Kenneth D. Mandl, MD, MPH Professor, Harvard Medical School Director, Boston Children s Hospital Computational Health Informatics Program chip.org
Virginia 2011
r, Mandl, ne JAMA Boston Children s Hospital Computational Health Informatics Program
No Linker
? How do we learn from All the Patients... All the Time? (Instead of Some of the Patients All of the Time... )... or really, just a couple of the patients, not that often.
Brownstein, Mandl PLoS One
Introduction of COX-2 Inhibitors Brownstein, Mandl PLoS One
Autism Phenotype Clusters Seizures Psychiatric Disorders
WE USE LITTLE (AND BIASED) DATA TO MAKE OUR INFERENCES
Pick a number Bourgeois, Mandl PLoS One
Predictive Medicine: Domestic Violence High Impact Paper!! Impact on practice? Nada Reis, Kohane, Mandl BMJ
13 EHRs Data Returns to Health System + 13 i2b2 Sidecars SMART Apps for Patient Review Trial Matching Contact Surveys PRO Collection = SHRINE Query for Network-wide Patient List
Federated EHR Networks Mandl, Nature Biotechnology 2015
??Data IN and Insights BACK to Point of Care??
2000 Conclusion: Computerized physician order entry is a powerful and effective tool for improving physician prescribing practices.
$48 BILLION DOLLAR FEDERAL INVESTMENT IN HIT
But 6 years later, we have an infrastructure that shows the clinician, one patient at a time, what he or she typed in in the past--a very narrow slice of big data
INNOVATORS AND INNOVATIONS CAN T GET TO THE POINT OF CARE
Motivation: Innovation Med-tastic is a well-funded NewCo which has developed an elegant medication list application that has physician and consumer facing functionality. Patients can annotate medication lists for accuracy. To work, Med-tastic needs Prescribing history Dispensed medication history Allergies Problem list diagnoses
Personalization of Therapy
BIG DATA COLLECTED AT POINT OF CARE DISAPPEARS
Turns out it Matters
DATA ARE NOT INTEROPERABLE
BIG DATA DON T GET TO THE POINT OF CARE
Contextualization Annals of Internal Med 2011
Designing the App Store for Health
API Can EMRs behave like iphones or Androids in that innovators readily create and widely distribute SUBSTITUTABLE apps across thousands of installs?
18.12 Issue
State-of-the-Art???
1 Design + 1 Developer + 1 Week
1 SMART App in 3 SMART Systems
SMART BP Centiles App Running on Cerner
41
Decision Support Integration e.g., Immunization Forecasting
Monograph app e.g., Diabetes
Merge App e.g., Microsoft HeatlhVault and Diabetes
Precision Medicine App
Linkage to external decision support services Pharmacogenomic rule sets Decision support for doctors Big Data Mashups: e.g., Personalized Medicine: There s an App for that
Clayton Christensen, HBS Susanna Fox American Medical Association The Advisory Board Company AARP BMJ Canadian Institutes of Health Research CMS HCA Lily MyHealthBook Polyglot Systems Premier Surescripts UK NHS Vecna
i2b2 the side car in action: EMR-like function
Some SMART Hotbeds
Activities now Jason I and II HL7 Argonaut The SMART App Gallery Meaningful Use 3
gallery.smarthealthit.org
gallery.smarthealthit.org @mandl Kenneth D. Mandl, MD, MPH Professor, Harvard Medical School Director, Boston Children s Hospital Informatics Program @bos_chip