Research in an IT Connected World: Building Better Partnerships – NIH and Health Care Systems
August 21, 2013
Building Partnerships
1. Why? The urgent need
2. Concrete examples of partnerships – the Health Care System Collaboratory Demonstration Projects
3. Preliminary thoughts about IT capacities, methods, rules
IOM: The Imperative for ActionDrivers of the Problem
Scientific uncertainty
Perverse economic and practice incentives
System fragmentation
Opacity as to cost, quality and outcomes
Lack of patient involvement
Underinvestment in population health
Clinical evidence development is not keeping pace with the emergence of
new diagnostics, treatments and insights into individual variation.
IOM- Learning Health Care System
Reversals
Examples:
Mortality was higher with recommended glycemic targets as opposed to more permissive standards (ACCORD, NEJM 2008)
Routine use of pulmonary artery catheters worsened ICU outcomes (NEJM, 2009)
Breast cancer survival was not improved by autologous stem cell transplant and intensified chemo compared with standard chemo (Tallman et al NEJM 2005)
Impermeable bedcovers have no benefit for adults with asthma, in spite reduced dust mite exposure (Woodcock, 2003 NEJM
Take home lesson:
We don’t always know what we think we know.
(hence the need to continue to examine standards of care)
What is a Practical Trial?
Defined Practical (pragmatic) Trials as those in which “the hypothesis and study design are developed specifically to answer the questions faced by decision makers”
Decision makers include patients, clinicians, payers,
& health care system policy makers
Pragmatic vs. Explanatory
Broad eligibility
Flexible interventions
Typical practitioners
No follow-up visits
Objective clinical outcome
Usual compliance
Intent-to-treat
Narrow eligibility
Strict instructions
Expert practitioners
Frequent follow-up visits
Surrogate outcomes
Close monitoring
ITT plus per protocol
Thorpe KE et al. J Clin Epidemiol 2009;62:464-75
Pragmatic Study Elements
Broad eligibility criteriaFlexible interventionsTypical practitionersNo follow-up visitsClinical outcomesUsual complianceIntent-to-treat
Thorpe KE et al. CMAJ 2009;180:E47
The Urgent Need
We need continual improvement in standards of care
To achieve that we need pragmatic trials
We cannot maintain an artificial distinction between research and quality improvement
Building a Partnership
1. Why? The urgent need
2. Concrete examples of partnerships – the Health Care System Collaboratory Demonstration Projects
3. Preliminary thoughts about IT capacities, methods, rules
The overall goal of the Commons Fund’s Health Care System (HCS) Collaboratory is to strengthen the national capacity to implement cost-effective large-scale research studies that engage health care delivery organizations as research partners.
Pragmatic Trial Demonstration Projects
1. Preventing hospital acquired infections: (PI Huang) Do intensified antibacterial bathing measures reduce hospital-acquired infections? Cluster randomized trial - HCA randomizing 50 hospitals – 375,000 patients
“The implications of this study are highly important. The lack of effectiveness of
active detection and isolation should prompt hospitals to discontinue the
practice for control of endemic MRSA in ICUs….”
Pragmatic Trial Demonstration Projects
1. Preventing hospital acquired infections: (PI Huang) Do intensified antibacterial bathing measures reduce hospital-acquired infections? Cluster randomized trial - HCA randomizing 50 hospitals – 375,000 patients.
2. Colorectal cancer screening: (PI Coronado) Does a simple intervention implemented through Federally Qualified Health Centers improve colorectal cancer screening? Cluster randomized trial 18 OCHIN Clinics – 10,000 patients.
3. Suicide prevention: (PI Simon) A pragmatic trial comparing suicide prevention programs. Can patients who admit to suicidal thoughts in a health care encounter be successfully randomized to one of two management strategies and do interventions reduce subsequent suicide attempts? Group Health, Kaiser Colorado – 16,000 patients.
Pragmatic Trial Demonstration Projects
4. Lumbar Spine imaging: (PI Jarvik) Does insertion of epidemiological information into imaging reports reduce subsequent diagnostic and therapeutic interventions? Kaiser, N Ca; Group Health, Mayo, Henry Ford. Cluster randomized 128 clinics,135,000 imaging reports.
5. Nocturnal blood pressure control: (PI Rosenthal) Does taking anti-hypertensive medications at night reduce CV events? U of Iowa and Duke primary care clinics, 6,000 patients.
6. Collaborative care pain management model: (PI Debar) Study impact of integration of psycho-social supports for patients with chronic pain on pain measures and opioid use. 3 Kaiser regions, Georgia, Hawaii, Northwest. Cluster randomized by practice. Several hundred practices, 6,000 patients.
7. Longer dialysis duration: (PI Dember) Does increasing dialysis duration reduce mortality? Partners Fresenius, da Vita, Cluster randomized trial of 402 dialysis units. 7000 patients.
Test and Strengthen Models for Clinical
Research in Partnership with Health Care
Delivery Systems
Develop capacity to leverage resources of major integrated health care systems for large scale clinical research studies
Test and improve methods to extract research quality data from electronic health information systems
Strengthen relevance and translatability of research results to ‘real world’ health practice
Develop and test more cost-effective models for large scale randomized clinical trials
Building a Partnership
1. Why? The urgent need
2. Concrete examples of partnerships – the Health Care System Collaboratory Demonstration Projects
3. Preliminary thoughts about Health IT capacities, methods, rules to facilitate a learning health care system
Data Steward #1
Authorized user
NIH DRN Secure Network Portal
1
52Data tables
Review & Run Query
3
Review & Return Results
4
6
Datatables
Review & Run Query
3
Review & Return Results
4
1- User creates and submits query (a computer program)
2- Data partners retrieve query
3- Data partners review and run query against their local data
4- Data partners review results
5- Data partners return results via secure network
6 Results are aggregated
Data Steward #2
ONC Query Health
Compliant
The Goal of the IT Approach
Creating secure networking capabilities and analysis tools
Build networking capacity, not a network
Insure that access to data occurs only with data stewards’ permission and active engagement
Leverages existing networks’ data and analysis tools
Can use many data types, e.g., EHR, claims, registries
Can use many data models, e.g., Mini-Sentinel, i2b2, OMOP
Can use existing querying tools, e.g., Mini-Sentinel modular programs
Every use requires the agreement of the data steward
Health Plan 1
Health Plan 2 CTSA 1 Research
Dataset 1
NIH Distributed Research Network Secure Portal
Research Dataset 2CTSA 2 Registry
NIH DRN Secure Portal
Mini-Sentinel Site A
Registry 1Mini-Sentinel
Site B
Medical Practice 1PBRN 1
CTSA 1Hospital
1Research dataset 1
NIH Distributed Research Network Coordinating Center
CTSA 2Health Plan 2
Health Plan 1
Network Managemen
tResearch Support
Query Support
Query Tool Developmen
t
Knowledge DatabaseSoftware
Development
Project Managemen
tData Models & Standards
Consultation
Health System Expertise
Knowledge Management System Cross project lessons learned, query tracking, meta-data capture, search functions, etc
AdministrationQuery ToolsSAS, SQL,
menu-drivenModular Programs
Summary Tables
Project 1PROJECTS
LIRE studyAnalytic ToolsFeasibility study
Query Interface Reporting Tools
Security \ Access ControlFile \ Query Repository
User Administration Workflow Management
Thorny issues
Defining the right questions
Ethical issues surrounding research on standards of care: eg. when is waiver of consent appropriate
Optimum trial design for group or cluster randomized trials
Preserving effective public-private partnerships
https://www.nihcollaboratory.org/
National Center for
Complementary and
Alternative Medicine
Clearinghouse: 1-888-644-6226
Web site: nccam.nih.gov
Twitter: @NCCAM
31
Pragmatic Trials
Pragmatic trials should test an intervention, or compare several interventions.
The question should be important – to patients, payers and with health care delivery systems.
The intervention should be reasonable simple and not require a complex structure for implementation. System level interventions may be particularly suitable.
The trial design should incorporate rigorous controls, prospectively identified, preferable by randomization.
The monitoring and outcome assessment should be as simple as possible – and potentially utilize electronic health records.
Top Related