Medical Technology and
Coverage Policy
Sean Tunis MD, MSc
Chief Medical Officer, CMS
June 6, 2004
Mortality in the 20th Century
3000
2500
Better treatment of cardiovascular
deaths per 100,000
disease, low birth weight infants
2000
1500
1000
500
0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990
Reduced infectious disease mortality (clean
water, sewers, antibiotics, better nutrition)
HEALTH SPENDING AND AGING:
HEALTH SPENDING AND AGING:
SELECTED OECD COUNTRIES 2000
14%
Now over 15%
U.S.
PERCENT OF GDP SPENT ON
13%
12%
HEALTH CARE
11%
Switzerland
10%
Germany
Canada France
9%
Netherlands
8% Australi Sweden
a Japan
7% U.K.
Iceland
6%
11% 12% 13% 14% 15% 16% 17% 18% 19%
PERCENT OF POPULATION OVER AGE 65
SOURCE: OECD Data, 2002
Technology and Spending
David Cutler (1995) estimated 50%
81% of economists identify technology
as primary cost driver (Fuchs 1996)
Project Hope (March 2001) estimates
25-33% of growth is technology
BCBSA report (Oct 2002) estimates
18% of growth is technology
Percent of Medicare Decedents Admitted
to ICU During their Final Hospitalization
(1995-96)
30.0
% Admitted to ICU (1995-96)
25.0
20.0
15.0
10.0
5.0
Desirable new/improved
Medicare benefits
Expanded Rx drug benefit
screening / prevention / wellness
quality improvement, safety programs
coordinated care for chronic illness,
disease management
EHR, e-visits, telemedicine
expanded clinical trials coverage
increased payments to physicians,
pharmacists, nurses
Reasonable and Necessary
Safe and effective (per FDA, if applicable)
Adequate evidence to conclude that the
item or service improves net health
outcomes
– emphasis of outcomes experienced by patients
function, QoL, morbidity, mortality
– generalizable to the Medicare population
– as good or better than current covered alternatives
High cost and/or small benefit generally
looked at more carefully
Evidence assessed using EBM framework
Panelist Topics
Susan Foote: Local CMS Coverage
Peter Neumann: National Coverage
Steven Pearson: Private Health Plans
Tanisha Carino: Whacky Europeans
LVRS and LVAD
CMS subsidized both NETT and REMATCH
Both expensive procedures with modest
clinical benefits
– Almost certainly over $200k/qaly
Evidence that outcome depends on peri-
operative and surgical skill
First non-transplant procedures limited to
subset of potential institutions
– LVRS limited to lung-transplant centers
– LVAD limited to heart-transplant centers with
15+ procedures
Verteporfin for AMD
Covered by CMS in 2002 for “classic”
One completed trial in “occult”
– Several flaws; second study underway
K-M showed 60 day delay to 20/200
– 5% sudden visual loss with tx
Course of tx around $10k
Reconsidered and covered for subset of
patients with occult and min. classic
Implications
CMS will continue working on predictability,
transparency and focus on evidence
Improving quality of evidence for decisions is
important
Increasing engagement of medical
professionals and patients is important
Medicare and other payers and purchasers will
also increase focus on value
– Quality of evidence will remain primary
– Balance of costs vs. benefits unavoidable
– Currently implicit, which has drawback
– Unclear exactly how it can be done explicitly
CASE STUDIES
LVRS and LVAD
Verteporfin for AMD
ICD for primary prevention of SCD
Percutaneous Coronary Interventions
CP1027346-1
Kaplan-Meier Estimates of the Survival
Probability in MADIT II for Patients
with QRS > 120 ms
p-value=0.001
Patients with pacemakers were excluded.
CMS analysis of the MADIT II dataset supplied by Guidant.
Kaplan-Meier Estimates of the Survival
Probability in MADIT II for Patients
with QRS 120 ms
p-value=0.25
Patients with pacemakers were excluded.
CMS analysis of the MADIT II dataset supplied by Guidant.
Sudden Cardiac Death
SCD-HeFT
Trial
Heart Failure
Mortality by Intention-to-
treat
0.4
HR 97.5% CI P-Value
Amiodarone vs. Placebo 1.06 0.86, 1.30 0.529
ICD Therapy vs. Placebo 0.77 0.62, 0.96 0.007
0.3
Mortality
0.2
0.1 Amiodarone
ICD Therapy
Placebo
0
0 6 12 18 24 30 36 42 48 54 60
Months of follow-up
Kaplan-Meier Survival by Treatment Group
Total Mortality
CONV: 19.8% Hazard Ratio = 0.69
ICD: 14.2% Adjusted P=0.016
31% reduction in risk of all-cause mortality
ICD:CONV
Variable # Pts.
Hazard Ratio
Age
65 yr 659
Gender ANALYSES II:
Male 1040
Female 192 Mortality
LVEF
0.25 401
NYHA Class
II 350
QRS There is no significant
0.15s 264
Beta-blockers
hazard ratios between
Yes 769 subgroups
No 463
All patients 1232
ICD Better 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Conventional Better
CMS policy on ICD prophylaxis
MADIT-II published 2002; prevalence pool 600k
(about half >65)
FDA approved: reasonable assurance of S&E
ACC/AHA/NASPE gave this IIa recommendation
– single trial
– possible selection bias
– need for risk-stratification
MCAC concluded that evidence was adequate
– Unaware of QRS data
CMS decided to cover wide-QRS subgroup, revaluate
after SCD-HeFT results
SCD-HeFT confirmed large ICD benefit in CHF pts
– Narrow QRS group much smaller benefit than overall
New Cancer Drugs
Increasing pipeline of anti-cancer
biologics with price $20k+
Rituxan, Zevalin, Bexxar for NHL
Irinotecan, Oxaloplatin for CRC
Velcaide for multiple myeloma
For advanced, relapsed, refractory dz
MMA will replace AWP with ASP
CMS looking at policy regarding off-
label use
AGE, SEX, ILLNESS, PRICE-ADJUSTED TOTAL MEDICARE
PAYMENT PER NON-CAPITATED MEDICARE ENROLLEE, 1996
Appleton, WI $3,404
Salem, OR $3,647
Minneapolis, MN $3,700
Rochester, MN $4,148
UNITED STATES $4,993
Providence, RI $5,056
San Diego, CA $5,678
Boston, MA $5,949
Baton Rouge, LA $7,700
Miami, FL $7,783
McAllen, TX $9,033
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000
SOURCE: Dartmouth Atlas of Health Care, 1999
Improvements to NCD
process
MMA timeframes
Posting of draft decisions
Guidance documents
BIPA appeals
Proactive engagement with range of
key stakeholders
Expedite high value and
true breakthroughs
Horizon scanning and expedited policy
development for high value services
(Steve Berkowitz)
FDA-CMS coordination
– Crawford and McClellan supportive
– staff level discussions underway
MTC (CTI) revised charter under
development
Improve quality of clinical
research
Medical devices evaluation Think Tank
last fall (Califf, Temple, Zarin, Tilson,
Kessler)
Annotated bibliography underway
White papers on key issues will be
developed
Exploring more NIH-CMS-Industry
collaboration for studies of promising
technology
Considering costs in
technology policy
Awareness of costs and CEA inform
contour coverage and set prices
– LVAD (coverage and price)
– iFOBT (price linked to CEA)
– Stents (payment rate and speed)
Other possible approaches
– CEA as criterion for coverage
– Evidence-based cost sharing
Medicare coverage:
Improvements since 1999
Coverage process described
– With specified timeframes for review
Explicit adoption of rules of evidence
– Increased technical sophistication
Increased transparency
– Public advisory committee (MCAC)
– Decision memos
– Highly interactive with stakeholders