New Era Of Comparative
Effectiveness Research
Hyatt Regency Capitol Hill
Washington, D.C.
October 5, 2010
With thanks to:
An Evaluation Of Recent Federal
Spending On Comparative
Effectiveness Research
Priorities, Gaps, & Next Steps
Joshua S. Benner
Marisa R. Morrison
Erin K. Karnes
S. Lawrence Kocot
Mark McClellan
The Recovery Act provided unprecedented
federal spending on CER—in a 2 year period
Actual and Projected CER Funding, FY 2009-2011
1200
Millions of dollars
1000
PCORI
800
600
OS
400
NIH**
200
AHRQ*
0
2009
2010
2011
Fiscal Year
Sources: AHRQ: http://www.hhs.gov/recovery/reports/plans/pdf20100610/AHRQ%20CER%20June%202010.pdf;
http://www.ahrq.gov/about/cj2011/cj2011.pdf; NIH: http://report.nih.gov/rcdc/categories/; Office of Secretary:
http://www.hhs.gov/recovery/reports/plans/pdf20100610/OS%20CER%20June%202010.pdf
*FY09/10: Based upon ARRA
actual/projected obligation
amounts
**FY09/10: Based upon actual
and projected spending
amounts
As of Aug 4, nearly all of the $1.1B was spent, but
tracking it is challenging
Sources:
• Recovery.gov
• Grants.gov
Accounted for
Not
accounted
for
AHRQ:
$299M
OS:
$282M
17%
83%
NIH:
$330M
• FBO.gov
• NIH Research Portfolio
Online Reporting Tools
Expenditures and Results
(RePORTER)
• NIH Research, Condition,
and Disease Categorization
(RCDC) system
• AHRQ Grants On-Line
Database (GOLD)
•AHRQ website
More than half of funds were spent on
improving capacity to do CER
Millions of dollars
500
Recovery Act CER spending, by project type
(As of Aug 4, 2010)
400
OS
300
NIH
AHRQ
200
100
0
Ev idence
dev elopm ent
and sy nthesis
Infrastructure
and m ethods
dev elopm ent
T ranslation and
dissem ination
Priority setting
Stakeholder
engagem ent
Recovery Act CER studies funded through NIH
span the disease areas prioritized by IOM
Implications for future CER spending
• Areas where more funding is needed
–
–
–
–
–
Experimental research
Evaluation of system-level reforms
Identification of effects in subgroups
Inclusion of understudied populations
Dissemination of results
• Need timely, comprehensive, “dashboard”
–
–
–
–
–
Opportunities and ongoing projects
Specific aims, including priorities areas addressed
Methods and data sources
Budget amount, duration, and granting agency
Linked to CER Inventory
Creating A High
Performance System For
Comparative Effectiveness
Research
Lynn Etheredge
Rapid Learning Project
George Washington University
A Rapid Learning Health System
• Designed to learn as rapidly as
possible about the best treatment
for each patient
• Key concept: in silico research (on
large computerized databases)
– Complements in vitro and in vivo methods
– Studies many more patients, richer (EHR) data, more
researchers; more, different, faster studies
– Researchers: multi-year data collection  log-on to
world’s evidence base. Databases designed for CER
A High-Performance CER System
• Learning about new technologies
– Registries, research plans, reporting
– E.g. new cancer therapies, off-label Rx use, surgery
• Comparability standards for studies
– To accelerate comparative research, data networking
• A National Database For
Effectiveness Research Studies
– Clinical studies submit standardized data-sets, a
growing world “evidence base” for CER
– Established by Presidential order
A High-Performance CER System
• A national network of clinical
research registries and databases
– Fill gaps in the evidence base for elderly, children,
pregnant women, persons with multiple chronic
conditions, minorities, rare diseases, surgery
• Getting results
– HHS public work plans, w/ due dates, accountability
– Create a national rapid-learning culture for CER, w/
websites, networks, rapid-learning communities
Comparative Effectiveness
And Personalized Medicine:
Evolving Together or Apart?
Robert S. Epstein
J Russell Teagarden
The Worry
• Comparative Effectiveness can help
determine the average effects
between alternative treatments
• But this may mask results from
important sub-populations (e.g.
Personalized Medicine)
There is some evidence to support
this concern
• A recent random sample of 5 major
journals and 319 trials – revealed:
• Only 29% of the studies tested
comparative effects within subpopulations
Source: Gabler NB et al: Dealing with heterogeneity of treatment effects: is the literature up to the challenge? Trials 2009;10:43.
Comparative trial showed 16% reduction in CV
events if on high dose statin vs. std. dose at 2 years
27
26
25
24
% CV events
23
22
21
20
Std dose
High dose
Source: Cannon CP et al: Intensive versus moderate lipid lowering with statins after acute coronary syndromes. NEJM 2004;350(15):1495-1504.
Personalized medicine finding:
40% CV drop in genetic sub-population!
Source: Iakoubova OA et al: Polymorphism in KIF6 gene
and benefit from statins after acute coronary
syndromes: results from the PROVE-IT TIMI 22 study.
JACC 2008:29;456-8.
Comparative effectiveness can also
provide evidence of clinical utility
8
7
6
5
Control
Tested
4
3
2
1
0
Clin Dx
Confirmed
Source: Mallal S et al: HLA-B*5701 screening for hypersensitivity to abacavir. NEJM 2008;358:568-79.
Conclusions
• Comparative effectiveness can
– Uncover findings not available
elsewhere due to studying diverse
patient populations
– Provide evidence of utility in the ‘real
world’
• Must remember to structure CE
studies with broad entry criteria and
include genetics – to help each
discipline grow together
A Patient Advocate’s Perspective
On Patient-Centered Comparative
Effectiveness Research
Tony Coelho
Former Chair, Board of Directors, Epilepsy
Foundation of America
A Flexible Approach To
Evidentiary Standards For
Comparative Effectiveness Research
Louis P. Garrison, Jr.
Department of Pharmacy, University of Washington, Seattle WA
Peter J. Neumann
Center for the Evaluation of Value and Risk in Health, Tufts Medical
Center, Boston, MA
Paul Radensky
McDermott Will & Emery LLP, Washington, D.C.
Sheila D. Walcoff
McDermott Will & Emery LLP, Washington, D.C.
Research support: National Pharmaceutical Council
Patient-Centered Outcomes Research
Institute (PCORI)--Responsibilities
• Setting priorities
• Developing methodological standards
• Communicating research results to
decision makers
Question: How can the institute support the different
standards for acceptable evidence used by various
government agencies, providers, patients, and other
decision makers?
Evidentiary Standards
Rules that decision makers apply in using different
types of evidence to carry out their mission
Evidentiary standards vary among decision
makers:
•
•
•
•
•
FDA for new drugs: “substantial evidence”
FDA for devices: “reasonable assurance”
FDA for promotion: “substantial clinical experience”
CMS: “reasonable and necessary”
FTC: “Competent and reliable scientific evidence”
Comparative Effectiveness Research in the
Decision Maker-Evidence-Policy Loop
Key Issues And Implications
• The information produced through CER will be a
“public good”, benefiting all decision makers.
• Models are a valuable tool for synthesizing data.
• Value-of-information analysis can help PCORI
with research prioritization.
“Experiment, observation, and mathematics, individually and collectively,
have a crucial role in providing the evidential basis for modern therapeutics.
Arguments about the relative importance of each are an unnecessary
distraction. Hierarchies of evidence should be replaced by accepting—
indeed embracing—a diversity of approaches.”
Sir Michael Rawlins, NICE
Methodological Standards
• Best practice for a given type of methodology
• Calls for a “translation table” to link methods to
research questions
• ACA: “provide specific criteria for internal
validity, generalizability, feasibility, and
timeliness of research”
There is a need for balance—and
tradeoffs—among these objectives.
Conclusion
• PCORI should take a balanced and
flexible approach to the types of research
it sponsors, being careful not to let
rigorous scientific methods become a
rigid evidentiary standard.
Designing Comparative Effectiveness
Research On Prescription Drugs:
Lessons From The Clinical Trial
Literature
Aaron S. Kesselheim, M.D., J.D., M.P.H.
Co-authors: Dave A. Chokshi, M.D., M.Sc., Jerry Avorn, M.D.
Division of Pharmacoepidemiology and Pharmacoeconomics
Brigham and Women’s Hospital, Harvard Medical School,
Boston MA
CER Not A New Phenomenon
• CER existed before recent developments
– Government and commercial funding
• Review of some examples of drug CER
– Goal: Identify methodological issues
• 3 areas of special concern:
– Choice of comparison treatments
– Time frame
– External validity
Sources: Hochman & McCormick, JAMA, 2010; Clement et al., JAMA,
2009
Designing Better CER Trials
• Choice of comparator
– The “straw man” problem
• Ex. #1: Comparator doesn’t represent “standard of
care” in efficacy trial (ASCOT)
• Ex. #2: Comparator with high side effect risk in
safety trial (MEDAL)
• Study time frame
– The “surrogate endpoint” problem
• Ex. #1: Surrogate endpoint not valid (DCCT)
• Ex. #2: Surrogate endpoints may mask safety
problems (RECORD, torcetrapib)
Sources: Dahlof et al., Lancet, 2005; Cannon et al., JAMA, 2006; DCCT Research Group,
NEJM, 1993; Home et al., Lancet, 2009; Psaty & Lumley, JAMA, 2008
Designing Better CER Trials (cont’d)
• External validity
– How are CER findings applied? (If at
all!)
• Ex. #1: Trial design doesn’t anticipate realworld use (RALES)
• Ex. #2: Non-representative enrollment
• Ex. #3: Inadequate effort to change
prescribing practices (ALLHAT)
• Ex. #4: Failure to account for critical
pushback (CATIE)
Sources: Juurlink et al., NEJM, 2004; Jagsi et al., Cancer, 2009;
ALLHAT Research Group, JAMA, 2002; Lieberman et al., NEJM,
2005
Moving Forward
• Options for the new era of CER in US
– Innovative approaches to study
methods
• Ex. #1: Adaptive trial design (STAR-D)
– Alternatives to randomized controlled
trials
• Observational studies
• Meta-analyses and systematic reviews
Sources: Rush et al., NEJM, 2006; Nelson, JNCI, 2010; Lumley, Stat
Med, 2002; Hlatky, Lancet, 2009
Policy Recommendations
• CER must compare the best available
alternatives
• Use validated surrogate endpoints, but follow
with hard endpoints and safety studies
• Emphasize “pragmatic” clinical trials with
ongoing reassessment
• Implement observational studies with
methodological rigor
• Link CER to effective strategies for
dissemination (academic detailing)
• Market findings to health care institutions
Why Observational Studies
Should Be Among The Tools
Used In Comparative
Effectiveness Research
Nancy Dreyer
Chief of Scientific Affairs, Outcome
19
Evidence Is Needed
Goal: Informed decision making about realworld practices and outcomes that result
from that behavior or treatment
Seeking Evidence
Relevant to Patients
and Providers
• Valid
• Feasible
19
The Framework
• Decision-makers need to rely on a full range of
high-quality comparative effectiveness research
– Methods and challenges
– Guidance for determining which type of study to
employ
• Whatever methods are used, research should be
designed and implemented using best practices
– Bodies of evidence
• Systematic reviews, meta-analysis, grading systems
– Individual study types
• RCT, observational studies, other applications
Choosing The Right Fit
• How much certainty is required?
• Time and budget?
• What are the comparators of interest?
– Is there enough variability in treatment
without randomization?
– Is there a reasonable overlap of patient
characteristics between treatments?
– Treatment complexity? Single, multiple, etc.?
• Would it be ethical and/or feasible to
randomize?
Use Observational CE Studies When RCTs
Are Impractical Or Infeasible (Examples)
– Patients and conditions not typically included in
RCT (frail elderly, children, pregnancies)
– Multiple treatment paradigms (hearing loss)
– Treatment adherence differs (asthma)
– Off-label uses result in different pop’ns being
treated (stents)
– Operator training (ICDs)
– Operator or institutional experience (surgical
procedures)
– Large study sizes needed (small differences)
The Most Informed Decisions Will Benefit
From Considering The Full Range Of High
Quality Research
Optimal study design depends on purpose, validity
and feasibility within constraints of time and budget
Adding The Patient
Perspective To Comparative
Effectiveness Research
Albert Wu
Professor, Health Policy and
Management, Bloomberg School of
Public Research, Johns Hopkins
University
Some Questions Cannot Be Answered
Without Asking The Patient
•The main objective of much of health care
is improving how patient feels and functions
• Reduction in pain (hip replacement)
• Improved functioning (cataract
extraction)
•Patient is best judge
•Patient best observer of some events and
health outcomes (complications)
Types of Information Captured
In Data Used For CER
Research
Utilization
Clinical detail
Patient perspective
Data
Clinical Care Data
Administrative
Data
-
+
++
++
+++
+
-
-
-
Linking Data Sources for CER
Administrative
Data
Patient
Reported
Outcome Data
Clinical Research
and EHR Data
Electronic Health Records (EHR)
•How can EHR be structured to include PRO
• As clinical indicator that can be tracked,
profiled
• Useable for CER?
Johns Hopkins Medical Institutions EPR
Clinician Website
Clinician can schedule a patient a survey
Patient Website
Patient is presented with survey(s) to complete
Patient Website
Patient survey – sample question
Clinician Website
Clinician can view patients results
“Listen to the patient: He is telling
you the diagnosis”
- William Osler
Patient-Centered Outcomes Research
Institute (PCORI)
• Capturing patient perspective vital to complete
picture of treatment impact
• Strategies to accelerate development of useful
evidence
• Apply research-grade standardized
questionnaires
• Include more uniformly in clinical trials,
registries
• Inject into EHRs
• Incentivize addition to administrative data,
e.g., pay for collection; require for
reimbursement
Thanks To Co-Authors
Claire Snyder
Carolyn Clancy
Don Steinwachs
Contact
Albert W. Wu, MD, MPH
Health Services Research Center
Johns Hopkins Bloomberg School of Public
Health
624 N. Broadway, Room 653
Baltimore, MD, 21230
awu@jhsph.edu
Four key elements of CER
• Compares all relevant alternatives
for real-world decision making.
• Patients are representative of typical
clinical settings.
• Accounts for differences among
individual patients in responding to
interventions.
• Measures all outcomes that are
important to patients.
Cost-effectiveness analysis
• Inherently comparative: the
incremental C-E ratio (ΔC/ΔE).
• ΔE measured in QALYs (qualityadjusted life years)
• Inputs to CEA:
– Costs
– Outcome measure: typically life
expectancy (LE)
– Preference measure for each outcome
ACA limitations on CEA
“The PCORI shall not develop or
employ a dollars-per-quality
adjusted life year …. as a threshold
to establish what type of health care
is cost effective or recommended.”
Should costs be a part of CER?
• People will need data on utilization
and costs
– It will be important to many patients
– Future changes in organization and
financing of care may make cost
information more relevant
– Private insurers will pay a tax to
support PCORI; CER should benefit
them.
Should cost-effectiveness
analysis be a part of CER?
• No… and Yes
• No
– PCORI should not routinely provide CE
analyses or rankings based on them.
– Rationale:
• ACA says PCORI should not draw a bright
line defining good value.
• The public is concerned about government
standardization of care
Should CEA be a part of CER?
• Yes
– CER should measure comprehensive
health effects, including utilization.
– Others can do the C-E analyses.
– Rationale:
• The law doesn’t proscribe CEA by private
parties (insurers, health plans)
• The public is paying for CER, and these
measurements are a public good.
• We cannot afford to ration information.
How Medicare Could Use
Comparative Effectiveness
Research To Set Payment
Levels At The Time of Coverage
Steven D. Pearson, M.D., M.Sc.
Peter B. Bach, M.D.
Current Coverage And Payment
• Historically separate silos
– Payment is generally set to reimburse
providers or manufacturers for costs of care
plus some profit
• Higher pricing w/o comparative evidence
– Proton beam therapy vs. IMRT for prostate
cancer
– Drug-eluting stents vs. bare metal stents
– ESA’s in cancer
– Lucentis vs. Avastin
Paying Well For Meaningful Innovation
Demonstrated superior
comparative clinical effectiveness
Medicare
Positive
Coverage
Decision +
CER
Determination
Insufficient evidence to judge
comparative clinical effectiveness
Demonstrated comparable
comparative clinical effectiveness
Usual Pricing:
Payment based on
existing formulae
Dynamic Pricing:
Payment based on
existing formulae for 3
years.
Reference Pricing:
Payment equal to
relevant comparator
Demonstrated
superior clinical
effectiveness
Continued insufficient
evidence or
demonstrated
comparable clinical
effectiveness
Challenges
• Methods
– A single time limit for all
interventions?
– Reference pricing drugs vs. surgery?
– Grandfather in older interventions?
• Politics
– Implementation of new pricing policies
at Medicare in the setting of PPACA
– Medicare payment changes require
new legislation
Conclusion
• The current Medicare cost-plus pricing system
is obsolete, even harmful.
• Managing the disconnect between price and
results downstream is messy at best.
• Using evidence to avoid perverse incentives
– Not “no” but “yes, and….”
– Use market forces to incentivize better evidence
development
– Reward innovators who demonstrate that their
interventions improve patient outcomes
compared to other alternatives
– Help Medicare shift to a more sustainable path
Applying Results Of Comparative
Effectiveness To Patients:
Obstacles and Solutions to Implementing
CER in a Large Health System – the VA
Experience
Joel Kupersmith
Chief Research and Development Officer
Director, Quality Enhancement Research Initiative
U.S. Department of Veterans Affairs
VA Healthcare System
• 8.1 M enrollees, 5.7 Million patients/yr
• >1200 Sites of Care
– 153 Medical Centers
– 773 Community-based Outpatient Clinics
– 260 Readjustment Counseling Centers
• >90,000 Providers
• Electronic Health Record
• Data driven practice network
• Complex environment
– Older, sicker Pts with substantial co-morbidity
– Dual use of VA and other healthcare
• Intramural research program embedded in health care
system with 35 years of CER accomplishment
VA Approaches to Implementation
• “New Medical Media”
– EHR tools
• Reminders
• Computerized decision support
– Tools for shared decision-making with patients
• “Old Medical Media”
– Guidelines
• Used by individual providers, also create consensus
– CME
• Systems
– Pharmacy Benefits Management
– Performance Measures
– Clinical leadership and Program Office policies
• Evidence Synthesis Program
• Quality Enhancement Research Initiative
– Brings approaches together
Example - VA Diabetes Care
•
•
Translation of individualized diabetes care
– E.g. Tighter control for younger patients, influence of comorbidity
Formulary policy emphasizes those drugs shown to improve
long-term clinical outcomes
– Others available under specific conditions
•
Guidelines emphasize risk-based strategies that vary and
individualize treatment based on age, co-morbidity,
underlying disease
•
Performance measures focus on those where benefits of
tighter control greatest (HbA1c <7-9)
– Research testing performance measures linked to clinical
actions and patient conditions
•
Diabetes database allows facility to look at care for different
strata of patients
– E.g. Elderly vs. younger, newly diagnosed
Challenges/Lessons Learned
•
Evidence base is modest overall
–
Must keep up with fast-changing evidence
–
•
Individualization
–
–
How do we implement results in different settings, different patients etc.?
How much variation is allowable while improving care for populations?
•
Patient centered approaches
•
Need to align approaches to implementation and diminish complexity
–
“Make the right thing the easy thing” for busy practices
•
Need to reward performance that reflects CER data and patient
circumstances
–
Performance measures are constantly evolving
•
Organizational and local characteristics are important
–
As is leadership involvement
•
Implementation should be an element of early research and not an
afterthought
‘Bench To Behavior’:
Translating Comparative Effectiveness
Research Into Improved Clinical Practice
Jerry Avorn, M.D.
Professor of Medicine, Harvard Medical School;
Chief, Division of Pharmacoepidemiology and
Pharmacoeconomics, Brigham and Women's
Hospital
The final translational hurdle
• We are not effectively using most of the
treatments that already exist.
• We already have some excellent drugs to manage
many chronic conditions, but often use them
poorly or not at all.
• Inadequate control of these conditions is
widespread:
–
–
–
–
–
high blood pressure
cholesterol
diabetes
osteoporosis
etc.
• The result: massive amounts of preventable
illness, cost, and death
• Great CER findings are likely to suffer the same
fate.
Why is this so?
• It’s nobody’s job to disseminate purely
evidence-based CE information in an
effective way to practitioners.
• The delivery system deals poorly (if at all)
with QA for clinical decisionmaking.
• The drug industry (and other vendors)
are very effective in achieving their goals.
– Marketing, continuing medical
education are sometimes counterproductive, CE-wise.
But much can be done
• ‘Social marketing’ of evidence-based
CE recommendations
– e.g., academic detailing
• Better separation of commercial
agendas and content from CME
courses
• Use of prompts in electronic health
records
– Who will control this real estate?
• Incentivize evidence-based practice
For more information
The BWH Division of Pharmaco-epidemiology and
Pharmaco-economics (“DoPE”):
www. DrugEpi.org
Academic detailing programs:
www. RxFacts.org
“Powerful Medicines: the Benefits, Risks, and
Costs of Prescription Drugs”
(Knopf, 2005):
www.PowerfulMedicines.org
Moving Comparative Effectiveness
Research Into Practice:
Implementation Science And The
Role Of Academic Medicine
Ann Bonham
Chief Scientific Officer, Association of American
Medical Colleges
The Measures Of Success
• Real gains in patient and population
health outcomes through adoption
of timely and relevant evidence.
• Gains that reach diverse groups.
• Metrics that matter to patients,
providers, and policymakers in
2020, as well as 2012.
Implementation Science:
Helping Insure Success
• Using the power of science to
promote the uptake, dissemination
and endurance of discoveries from
comparative effectiveness research.
• Real changes in routine practice and
behaviors of health care systems,
care providers, and patients.
Role Of Academic Medicine
• Tripartite mission.
• Learning laboratory for testing and
implementing change.
• Access to diverse and underserved
populations.
• Collaborative models across broad
disciplines.
The Way Forward
• From silos to solutions; integrating
missions around patient outcomes.
• Integrated leadership.
• Strategic allocation of resources and
human potential.
• New transdisciplinary partnerships.
The Case For A National
Patient Library
Authors: Jeffrey C. Lerner, Daniel M. Fox, Sheryl B. Ruzek,
Gail E. Shearer
Jeffrey C. Lerner, Ph.D.
President and Chief Executive Officer
ECRI Institute
Perspective
• Health services researchers put relatively little time,
intellectual effort, or money into sophisticated
communication of evidence-based information to patients
and their families, and towards influencing the
clinician/patient encounter.
• Manufacturers of medical products spend almost as much
on advertising as on R & D, including billions on direct-toconsumer promotions.
• Hospitals and clinicians advertise to patients.
• Advocacy groups focus on communicating with patients.
• A National Patient Library of evidence-based information
can help create a better balance in the information
“available” to the public.
Background
• The Patient Protection and Affordable Care Act calls for
“shared decision making” by patients and clinicians. See
sections 931 and 936.
• The Act establishes the Patient Centered Outcomes
Research Institute (PCORI). The National Patient Library
could promote public awareness and the use of findings of
PCORI’s independent research, and underscore the
central role of public finance in conducting that research.
What A Library Would Deliver And To Whom
• Primary purpose—a repository and clearinghouse of
evidence-based information designed for patients and
their lay and professional caregivers.
• Other central purposes:
– assess patients information needs and the degree to
which they are met effectively.
– serve as a center of innovation for health
communication across the spectrum of media.
– assess and disseminate evidence about the effects of
patients’ use of information on the use and cost of
healthcare.
Challenges And Opportunities
• Meet patients’ needs for trustworthy, accessible,
and timely information.
• Achieving sufficient initial and sustainable
funding.
• Ensuring that the public believes that the
purpose of the National Patient Library is to
serve them.
Politics of Comparative
Effectiveness Research
Gail Wilensky
Senior Fellow, Project HOPE
Politics of Comparative
Effectiveness Research
Gail R. Wilensky
Project HOPE
October 5, 2010
87
Comparative Effectiveness
Research
♦ Convergence of Interests: -Major focus on ing uninsured
Need to value, slow spending
♦ To change from where we are, need
Better Information; Better Incentives
(These used to be good Republican concepts)
Explosive Politics Around
CER
CER has become a “lightening rod” for criticism by
some conservatives; some in industry
“gov’t run health care”; “socialized
medicine”; “gov’t monitoring your doctor”;
“code words for denying care by birth date”
More to come? Pre-election? Post-election?
CER: Now Part of Reform
2010 Legislation
But -♦ Concept remains controversial
♦ And with a variety of limitations
♦ And an uncertain role in decision-making
CER – Still Very Fragile
Early ARRA Results
Will Be Important
♦ Interesting mix of studies
♦ Important mix of methodologies
-- IOM top 100: ½ RCT; ½ not
♦ Some results could be controversial
-- Need input from affected parties
♦ Dissemination/explanations will be important
Ultimate Challenge:
Making Use of CER
♦ Primary stated purpose: improved health outcomes
♦ “Secondary” purpose: building block to
“spending smarter”
-- CER more suited to reimbursement decisions
-- Value-based reimbursement/value-based
insurance principles
Using CER to Moderate
Spending
A major challenge!
♦ PCORI: can’t mandate coverage or
reimbursement
♦ Other limitations on PCORI
♦ CMS can’t use cost in coverage decision; no
authority to use CER in reimbursement
First Steps to
Moderating Spending
♦ Encourage private payers to use CER in making
reimbursement decisions
♦ Encourage use of value-based insurance by
private payers
♦ Promote pilots/demo’s of VBI for Medicare
reimbursement
Need to focus on the long
run
Will CER Be Politically
Sustainable?
Insights From National Public Opinion
Surveys
Alan S. Gerber, Eric M. Patashnik,
David Doherty and Conor Dowling
Support Rating for Funding CER
75
Overall, the Public Supports Government
Funding of CER
50
Overall
Age
Survey conducted 5/21-24/2010 (N=2017)
Education
Race
Female
Male
Hispanic
Black
White
High
Low
65+
18-37
All
25
Gender
A Sharp Partisan Divide Over CER Among Voters
Support Rating for Funding CER
75
50
25
Did Not Vote 2008
Survey conducted 5/21-24/2010 (N=2017)
Democrats
Voted 2008
Independents
Republicans
The Public is Enthusiastic About Some Uses of CER,
Opposes Others
To what extent would you support or oppose using comparative effectiveness
research to:
0%
25%
50%
provide information about whether a given treatment
works better than alternative ways of treating patients
with the same condition.
create warning labels for treatments that are not
supported by strong scientific evidence.
determine whether Medicare and private insurance
companies will cover new treatments that have just
become available.
66.4%
8.9%
63.9%
11.2%
51.4%
21.0%
determine what groups of patients should be protected
from budget cuts in Medicare and other government
health programs.
charge a patient more to get a treatment that research
has not shown to be effective, even if the patient's own
doctor recommends it.
Survey conducted 5/21-24/2010 (N=2017)
75%
34.1%
34.1%
15.4%
61.0%
Support
Oppose
Americans Fear Guidelines will Lead to Rationing
and "One-Size-Fits-All" Medicine
Mean support
for Board
issuing
mandatory
guidelines: 34
on 0-100 scale
(May 21-24,
2010 survey)
Treatment Guidelines (For)
...doctors are affected by economic incentives...and this
can lead doctors to give patients unnecessary care…
...doctors in a given area may be unaware that better
treatment approaches are being used elsewhere…
Treatment Guidelines (Against)
No outside group should come between doctors and
patients in making treatment decisions.
...doctors will be unable to tailor care to the needs of
individual patients.
The government and insurance companies will use
treatment guidelines as a way to control costs and ration
care.
0.0%
Survey conducted 11/5-12/31/2009 (N=1026)
25.0%
50.0%
Very/Somewhat Convincing
75.0%
Very Convincing
100.0%
Experiment 1: Effects of Exposure to
Arguments For/Against CER
• Baseline support of CER assessed (100-point
scale)
• Respondents randomly assigned to receive one of
two pro arguments
• The information will improve health care outcomes
• Will help reduce the budget deficit by cutting wasteful spending on
ineffective treatments without lowering quality of care
• Respondents randomly assigned to receive one of
several con arguments
• CER will lead to “one-size-fits-all” medicine
• CER will lead to rationing
Survey conducted 7/30-31/2010 (N=2010)
Experiment 1: Results
Pro-CER
Argument
Anti-CER Argument
One Size Fits All
Ration Care
The information from CER
will improve health care
outcomes
-10
-6.9
CER can help reduce the
budget deficit by cutting
wasteful medical spending
without lowering the
quality of care
-9.9
-8.0
Experiment 2: Best Responses To Anti-CER
Arguments
• Respondents randomly assigned to receive one of two antiCER arguments: “one-size-fits-all” and rationing
• Respondents randomly assigned to receive one pro-CER
rebuttal argument
• For “one-size-fits-all” argument, the four rebuttals were
–
–
–
–
CER is supported by doctors
“One-size-fits-all” is a scare tactic
It is important to learn what works best for most patients
Research is not limited to studying average treatment effects and can
incorporate group differences
• Respondents rated persuasiveness of pro and con
arguments (0=con more persuasive; 100=pro more
persuasive)
Survey conducted 7/30-31/2010 (N=2000)
Experiment 2: Results
“One size fits all”
Doctors Support
65.8
“Scare tactic”
51.3
Knowledge of
typical patient
is valuable
52.4
Can design
research to
study subgroups
59.9
Conclusions
•
•
While many Americans favor CER, public support is conditional
on how research findings are translated into practice
– Uses of CER that are perceived to interfere with the doctorpatient relationship or that smack of rationing will spark
opposition among politically engaged citizens
– It will be critical for the Institute to generate information that
consumers find personally useful and to demonstrate that
CER can produce tailored, nuanced recommendations that
won’t lead to “cookie-cutter” medicine
The public trusts doctors and looks to them for guidance
– The political sustainability of CER will depend in part on
whether individual doctors and medical societies actively
defend CER both in general and with respect to the usefulness
of studies conducted within their practice areas
100%
Many Americans Oppose A CER Board
Establishing Mandatory Guidelines
75%
50%
25%
Voted in 2008
Survey conducted 5/21-24/2010 (N=2017)
Party ID
Strongly Oppose
Age
Oppose
Race
Hispanic
Black
White
65+
18-37
I
R
D
Yes
No
Overall
0%
To Tame Costs, Link CER To Consumer Rewards
Rather Than Penalties
Strongly
Support
100
75
58.9
48.7
50
25
Strongly
Oppose
0
Cash for choosing less expensive
treatment
Survey conducted 5/21-24/2010 (N=2017)
Higher Co-pay for more expensive
treatment
Health Reform’s Tortuous
Route To The PatientCentered Outcomes Research
Institute
Kavita Patel, M.D., M.S.
Director, Health Policy Program
patel@newamerica.net
Process and Obstacles
• Evolution from previous proposals, similar ideas
and language
– Starting in 2007
– Dem control of Congress
• 2009 Health reform debate
• Senate side
– HELP and Finance Committees
• House side
– Leadership Tri-Committee bill
• USPSTF recommendations, Nov. 2009
Creating PCORI
• Structural concerns
– Inside government (AHRQ) vs.
independent outside entity
• Scope of power
– Cost and coverage provisions
in bill language
– Fears of rationing
• Must fill need for unbiased
scientific evidence
PCORI Membership
• Chairman: Eugene Washington MD, MSc
• Vice-Chair: Steve Lipstein MHA
• Clinicians, Researchers and Delivery System Experts
– Sharon Levine MD; Robert Zwolak MD, PhD; Arnold Epstein MD;
Harlan Krumholz MD; Debra Barksdale PhD, RN; Christine Goertz
DC, PhD; Lawrence Becker; Grayson Norquist MD, MSPH; Kerry
Barnett JD
• Consumer groups:
– Ellen Sigal PhD; Allen Douma MD; Andrew Imparato JD
• Industry:
– Freda Lewis-Hall MD; Harlan Weisman MD ; Richard E. Kuntz MD
• State or Federal Government:
– Robert Jesse MD, PhD; Leah Hole-Curry JD; AHRQ & NIH Directors
Source: GAO Announces Appointments to New Patient-Centered Outcomes Research Institute (PCORI) Board of
Governors. Sept 23, 2010. http://www.gao.gov/press/pcori2010sep23.html
Outlook
• What if Scott Brown had not won?
• Potential loss of momentum if Democrats lose
majority in upcoming elections?
• High value of CER to clinicians and patients?
• What will happen to the NIH, AHRQ, FDA
operations around CER and their methods?
– Could there be an independent review process
established by PCORI for NIH?AHRQ CER ?
With thanks to: