A Rapid-Learning Healthcare SystemIn silico research

A national HIT infrastructure of computable dataAdopting best practices

Lynn EtheredgeWolfram Data Summit 2011

September 8, 2011

A Rapid-Learning Health System

• Objective: a health system that learns as rapidly as possible about the best treatment for each patient– Addresses personalized medicine, major health system problems: clinical

practice variations, poor quality, lack of comparative effectiveness research (CER), high & rising costs, ineffective markets, inefficient regulation. Advances translational science & evidence-based medicine.

• Key concept: in silico research -- using large computerized databases and research networks for 21st century science– Complements in vitro (lab) and in vivo (experimental) methods

– Enables study of many more patients, richer & longitudinal data, many more researchers, more, different & faster studies

– Researchers: multi-year data collection log-on to & use the

world’s evidence base of computable data

A Rapid-Learning Health System

• RL = R + C + D + Apps

• We have: R (great researchers), C (very fast computers)

• D (databases) = a national system of high quality, clinical research registries & databases, pre-designed, pre-built, and pre-populated for RL– Most RL questions are applied research – treatment “A” vs treatment “B”

for (sub) population “C”: can specify data that will answer the question

– Electronic health records (EHRs) enable RL databases & registries, a data-rich environment, a computable world clinical evidence base

• Apps: user-friendly software to acquire, share, analyze & use computable data, quickly & efficiently

A Rapid-Learning Health System

Toward A Learning Healthcare System

• Recent initiatives:

– Comparative effectiveness research (CER) ($1.1B). A national system of “gap-filling” registries and networks for sub-populations usually not included in clinical trials: children, seniors, minorities, persons with multiple chronic conditions, rare diseases. Independent CER agency (PCORI);

– A national EHR system for all Americans ($27B+);

– FDA Sentinel Network (100 M patient records);

– A Medicare/Medicaid Innovation Center for national pilots and implementation of best practices ($10 B);

Toward A Learning Healthcare System

• Recent initiatives:

– A national biobank + bio-repository (Kaiser, NIH, RWJF), 500,000 patient EHRs w/ genetic & environmental database, bio-specimens;

– NIH eEMERGE network for genetic & clinical studies, personalized medicine, with shared “open science” data;

– NIH “Collaboratory” with HMO Research Network (19 HMOs, 14 M patients (28M) w/ EHRs, Virtual Data Warehouse w/ standardized core data elements;

– VA “one million veterans” initiative w EHRs, genetic information;

– EHR4CR: European universities & industry (33 partners)

Toward A Learning Healthcare System

• Recent initiatives:– Physician societies & specialists building rapid-learning systems:

• Real-time oncology learning network (ASCO) – Sept 16

• National cardiovascular research infrastructure & registries (ACC)

• Pediatrics studies networks (PEDSNet, PTN, PQMP)

• National quality registry network (NQRN) – Nov. 2011

– Advances in predictive models, e.g. Archimedes-ARCHeS & IndiGo (40% fewer strokes/heart attacks, 70% less cost) (NYT, 5/19/2011)

– New statistical & research methods: VA adaptable trials, accelerated multi-cohort/rapid cycle strategies, Bayesian & decision-theory statistics, mixed-methods, multiple-source research, randomized cluster designs

– National Center for Accelerating Translational Science (NCATS): new NIH institute (Oct. 2011), CTSA Consortium (60 academic centers)

EHRs, Registries & Apps

• National goal: EHRs for all Americans; $27 B in subsidies for federally-qualified EHRs; a learning healthcare system

– EHR Status: rapid-learning leaders w/ EHRs & research programs in organized care systems (Kaiser, VA, Geisinger, HMORN) ~ 35 M patients; larger multi-specialty practices & institutions. Elsewhere fragmentary (25% physician practices, 10% hospitals).

• An evolving HIT strategy for computable data– National EHR standards, with a core dataset & data exchange capabilities

– A national system of research registries for all patients & conditions

– A vibrant Apps marketplace to acquire, share & use data (aka modules, plug-ins, links, downloads, add-ons)

EHRs, Registries & Apps

• Rationale for basic EHRs + Apps– Vast differences between leaders & average healthcare, economy, politics,

more reasons to purchase & use EHRs; vastly expand potential for acquiring & using computable data

• Growing use of private sector EHR standards, registries, modular/specialized HIT & Apps: – HMO Research Network, its Virtual Data Warehouse, EPIC systems

– Specialty & disease-focused learning networks (cancer, cardiology, pediatrics, medical devices, rare diseases)

– Patient-focused Apps: Health 2.0, M-Health (500 m users by 2015? Star-Trek “tricorder” medicine?)

– NIH’s Apps-development challenge (300 databases, 8/31/2011), FDA Sentinel Apps-standard algorithms, CDC flu & reportable diseases Apps

Rapid-Cycle Learning & Adoption of Best Practices

• Health Reform legislation: $10 billion & models to be tested by CMS & rolled-out, if successful (examples): accountable care organizations, pay-for-performance, medical home, bundled payment, community-based care transitions, independence at home, hospital re-admission reductions, global safety net system payment, national quality reporting system, all-payer systems, integrated care for “dual eligibles” payment incentives for evidence-based cancer care, improved post-acute care through continuing care hospitals, comprehensive payments for “innovation zones”, payment reforms to advance care coordination, collaborations of high-quality, low-cost health care institutions, patient decision-support tools, etc.

Rapid-Cycle Learning & Adoption of Best Practices

• HHS new $ 1B Medicare initiative to produce $50B in 10 yr savings from improved safety, fewer re-admissions (April 2011)

– Partnership For Patients uses research and data to identify “best practices” and interventions

– Financial incentives through payment reforms, pilot projects

– Target: 60,000 lives saved over next 3 years

• Recently announced: Accountable care organizations (ACOs), bundled payments

• Many more rapid-cycle learning initiatives, coming soon, e.g.

– Maternal & infant care

– Cardiovascular disease - 1 million heart initiative

Key Messages

• US national policy is adopting a new way of thinking -- a learning healthcare system

• The key idea is to learn, as quickly as possible, about “best practices” – and then to have these adopted throughout the healthcare system. And to create a continuous learning cycle, and a continuously learning system

• All parts of this new system are now open for discussion & creation – a great time for innovators, computable-data people, and do-gooders !!