C O M M E N TA R Y
Paying for Outcomes: Innovative Coverage and
Reimbursement Schemes for Pharmaceuticals
Josh J. Carlson, PhD; Louis P. Garrison, Jr., PhD; and Sean D. Sullivan, PhD
‘‘The customer really doesn’t want a drilling machine, he wants a hole-in-the-wall.’’
—Theodore Levitt1
R
ecently, there has been significant interest in novel coverage and reimbursement schemes for medical products that
involve the concept “pay-for-performance,” or “risk sharing.” The emergence of these schemes could have far-reaching
implications for both payers and manufacturers, especially with
the current policy emphasis on evidence-based methods for technology assessment, pricing and reimbursement reform, and cost
containment. A few high-profile cases have brought the concept
further into focus within the health care industry, including the
U.K. National Health Services’ (NHS) scheme to reimburse drugs
to treat multiple sclerosis in the context of evidence development,
Johnson and Johnson’s scheme to refund the NHS for multiple
myeloma patients who fail to respond after 4 cycles of bortezomib
(Velcade), and UnitedHealthcare’s scheme to tie reimbursement
of Genomic Health’s Oncotype DX assay to a program of data
collection examining the link between test results and treatment
choices in women considering adjuvant breast cancer therapy.2-4
These and several other schemes both in the United States and
worldwide have been profiled in the news media and suggest this
may be an emerging trend.3-5 Although the majority have been
implemented outside the United States, a handful have been put
into practice in the United States and may portend additional
arrangements on the horizon. As these models continue to gain
momentum, their potential impact on U.S. health care system
increases. As such, a discussion of the underlying drivers, barriers, and potential implications of these new models for U.S.
health care payers is warranted.
Why the Interest?
Performance-based reimbursement schemes appear to have
arisen in response to a number of external factors such as (a) a
limited evidence base and associated uncertainty for pharmaceuticals and other medical products at the time of market introduction owing, in part, to accelerated drug approval; (b) increasing
cost pressures from those who fund and pay for health care; (c)
the increased use of external reference pricing globally; and (d)
an increased emphasis on policies to control new technologies
such as health technology assessment programs. These schemes
have an intrinsic appeal because they have the potential to alleviate some of the negative consequences that can arise from these
www.amcp.org
health care trends. The relevant uncertainty arises in the transition from efficacy-oriented clinical trial estimates derived during
product development to realized performance (effectiveness) in
a given patient population. These uncertainties include those
related to who will get treated (i.e., patient and subgroup heterogeneity), how much product will be used (both individually and
at the population level), and what the impact of the intervention
will be on long-term individual and population outcomes. These
uncertainties are directly linked to a product’s ultimate budget
impact and realized value. In addition, they are particularly
salient at product launch when coverage and reimbursement
decisions are typically made.
Cost pressures on pharmaceutical manufacturers and payers
continue to grow with the rising cost of drug development; the
emergence of new medical technologies, including diagnostics;
and the general expansion of treatable conditions,6,7 which is
exemplified by the potentially expanded indication for statin therapy to include healthy individuals with high levels of
C-reactive protein.8 In this environment, health care payers are
under great pressure to contain costs in order to remain competitive. As such, performance-based schemes that link payment to
outcomes may be a means to create incentives for manufacturers
to participate in targeting their products toward those most likely
to benefit, since they would expect to receive a lower price per
unit in less targeted patient populations. In addition, utilizationrelated schemes (e.g., price-volume and per patient utilization
caps) limit total expenditures and budget impact.
Another motivation for performance-based schemes outside
the United States can be to provide a means to give a discount to
payers without changing the global price—the lowering of which
can have negative impacts on revenue due to external reference
pricing (i.e., benchmarking drug prices across countries). For
Performance-based reimbursement schemes
Reimbursement schemes between medical product manufacturers
and payers in which the performance of the product is tracked in a
defined patient population over a specified period of time and the
level of reimbursement is tied by formula to the outcome.
Vol. 15, No. 8
October 2009
JMCP
Journal of Managed Care Pharmacy 683
Paying for Outcomes: Innovative Coverage and Reimbursement Schemes for Pharmaceuticals
example, the list price for bortezomib remains unchanged in the
current performance-based scheme with the NHS, but because
the NHS will be refunded for patients failing to respond, the net
price paid per unit of drug will be lower than the list price—in
effect, a discount.9
Finally, the increased worldwide emphasis on health technology assessment has created an environment in which payers
and decision makers are requiring more and better evidence to
inform coverage and reimbursement decisions. Performancebased schemes can add to the evidence base through the associated programs of data collection, especially if they are made
publicly available.
Performance-Based Schemes in the United States
There are 4 performance-based schemes between payers and
manufacturers in the United States of which we are aware (Table
1). There have been other schemes aimed at patients covering
products for blood pressure, hair loss, smoking cessation, and the
ever popular erectile dysfunction—but these were more akin to
money-back guarantees for other common products (e.g., brake
pads).5,10 The first scheme in the United States to include insurers was an older scheme implemented in 1998 that is no longer
in practice in which Merck offered to refund both patients and
insurers up to 6 months of their prescription drug costs if simvastatin (Zocor) plus diet did not help them lower LDL cholesterol to
target concentrations identified by their doctors. The scheme was
offered as part of Merck’s “get to goal” campaign and appears to
have been more of a patient-focused marketing plan rather than
a major shift in the coverage and reimbursement policy for simvastatin.5 Nonetheless, it predates many of the current examples
and indicates that although interest and uptake of these schemes
may be relatively new, the concept itself has been around for a
while, even in the United States. The other 3 U.S. examples are
more recent and include schemes between UnitedHealthcare
and Genomic Health for the Oncotype DX genomic test used to
predict recurrence of breast cancer;3 a scheme between CIGNA
and Merck regarding 2 related products, sitagliptin (Januvia)
and sitagliptin + metformin (Janumet), in the treatment of type
2 diabetes;11 and a scheme between the manufacturers of risedronate (Actonel, Proctor & Gamble, and sanofi-aventis) and
Health Alliance in the treatment of osteoporosis.12 The details
that underlie these schemes are worth exploring.
UnitedHealthcare agreed to reimburse the Oncotype DX test
at list price ($3,460 per test) for 18 months while it and Genomic
Health track the results to determine if the genetic test is having
the anticipated effect on actual clinical practice. If the number of
women receiving chemotherapy exceeds an agreed-upon threshold, even if the test suggests the patients would not benefit from
therapy, the insurer will receive a pre-negotiated lower price.3
The savings for the payer, and that which underlies Genomic
Health’s value message and pricing, is in women who would
have received expensive (approximately $15,000, including costs
684 Journal of Managed Care Pharmacy
JMCP
October 2009
Vol. 15, No. 8
associated with infusion, patient time, use of colony-stimulating
factors to prevent myelosuppressive complications, and management of chemotherapy-related side effects)13 adjuvant chemotherapy under current treatment guidelines, but for whom this
assay suggests it is not necessary. If, however, women and their
doctors do not follow the test results and proceed with adjuvant
chemotherapy, the payer receives no cost savings. In fact, they are
worse off because they have also paid for a relatively expensive
genomic test. It therefore appears that UnitedHealthcare wanted
to ensure that they were getting sufficient value and cost offsets
to warrant the coverage and reimbursement amount for the assay
by implementing an internal study on the proportion of women
whose treatment choice coincides with their genetic test results.
If this proportion is not in line with the agreed-upon estimates, a
pricing change would then be warranted to align the reimbursement amount with the actual value received.
The details for the scheme between Merck and CIGNA involve
the coverage and reimbursement for two type 2 diabetes drugs—
sitagliptin and sitagliptin + metformin. Under the scheme, CIGNA
will assess the blood sugar levels using hemoglobin A1c lab values for patients on any oral antidiabetic medications. If the A1c
values, in aggregate, have improved by the end of the agreement
period, the discounts offered by Merck on sitagliptin and sitagliptin + metformin will increase by a pre-agreed amount. In addition,
CIGNA will use claims data to determine if patients are taking
sitagliptin and sitagliptin + metformin as prescribed by their physicians. If so, Merck will further increase the discounts it offers
CIGNA on the 2 products. Finally, the agreement also includes
better placement for sitagliptin and sitagliptin + metformin on
CIGNA’s formulary tiers with a lower copayment versus that for
other branded drugs.11,14 This agreement differs from many other
schemes we have encountered in that there is a deeper discount
when patients on any diabetes drug improve their A1c lab values.
Previous schemes have linked discounts to poorer performance
for patients on the specific medical product in question (e.g.,
bortezomib in the United Kingdom).9 The scheme provision that
ties the improved adherence to larger discounts makes intuitive
sense because it can benefit all the key parties—payers, manufacturers, and patients. Diabetes patients who are more adherent
tend to have better outcomes.15 Furthermore, patients with better
adherence and outcomes tend to utilize fewer resources, which
can provide a cost savings for the payer who is already receiving
discounts on the cost of the drug.16 This may in part be attributable to a healthy adherer effect.16 Finally, manufacturers can
improve sales volumes with better patient adherence because
patients will be filling a greater proportion of their prescriptions.
This improved sales volume will likely offset the lost revenues
related to the per unit discount offered by Merck. As a recent New
York Times article stated, “Merck is betting not only that its drugs
prove superior but that CIGNA’s incentives to reap the benefits
of the deeper Januvia and Janumet discounts will prompt the
insurer to try to keep patients on those drugs.”13 This scheme is
www.amcp.org
Paying for Outcomes: Innovative Coverage and Reimbursement Schemes for Pharmaceuticals
TABLE 1
Performance-Based Coverage and Reimbursement Schemes in the United States
Source
Moldrup (1998)5,10
Disease area
High
cholesterol
Product
simvastatin
Pollack (2007)3
Breast cancer
Oncotype Dx
Genomic
Health
UnitedHealthcare
Shearer (2009)11
Type 2
diabetes
sitagliptin;
sitagliptin + metformin
Merck
CIGNA
Merck has agreed to peg what the insurer CIGNA pays for the diabetes drugs sitagliptin and sitagliptin + metformin to how well type 2
diabetes patients are able to control their blood sugar.
Osteoporosis
risedronate
Proctor &
Gamble,
sanofi-aventis
Health
Alliance
Two companies that jointly sell the osteoporosis drug risedronate
agreed to reimburse the insurer Health Alliance for the costs of treating nonspinal fractures suffered by patients taking that medicine.
Anonymous (2009)12
Manufacturer
Payer
Merck
Patients and
insurers
in line with the concept of “value in use,” where value is defined
as “the benefit the customer obtains through use, and compels
the firm to bring in customer usage as part of its responsibility to
deliver the outcome.”17
Finally, the scheme between Proctor & Gamble, SanofiAventis, and Health Alliance for the use of risedronate in osteoporosis has its own unique aspects that differentiate it from other
performance-based schemes.12 In this scheme, the 2 companies
that jointly sell the osteoporosis drug agree to reimburse the
insurer for the costs of treating nonspinal fractures suffered by
patients who consistently take the medication. This appears to
be the first published example of a manufacturer agreeing to
cover the cost of disease-related sequelae as opposed to discounting or refunding the cost of their product. This scheme will
certainly lower the medical costs to Health Alliance, since hip
fractures and wrist fractures cost approximately $30,000 and
$6,000, respectively. The benefit to the manufacturers will be in
keeping patients from switching to cheaper generic versions of
alendronate and through maintaining a lower copayment level
than their competitor, ibandronate. This scheme further illustrates one of the proposed motivations for the implementation of
performance-based schemes—decreasing the risk to the payer
related to product uncertainty—in this case, the uncertainty
about the benefit of risedronate in terms of reducing nonspinal
fractures. This uncertainty was revealed in the clinical trials of
risedronate, which failed to show a statistically significant reduction in nonspinal fractures, whereas some of its competitors have
been able to demonstrate this benefit in their trials.18,19 In essence,
the makers of risedronate are betting that their product will be
able to reduce these nonspinal fractures in actual practice and/
or the cost of treating them will be offset by maintaining or even
expanding their market share in a highly competitive market in
which it may not be the market leader.
www.amcp.org
Description
Merck promised to refund patients and insurers up to 6 months of
their prescription costs if simvastatin plus diet did not help them
lower LDL cholesterol to target concentrations identified by their
doctors.10
UnitedHealthcare agreed to reimburse the Oncotype Dx test for 18
months while it and Genomic Health monitor the results. If the
number of women receiving chemotherapy exceeds an agreed upon
threshold, even if the test suggests they do not need it, the insurer
will negotiate a lower price.
Barriers to Implementation
Although these agreements have an intrinsic appeal, tying reimbursement to a product’s actual performance, there are many
barriers to their implementation. Apparent major barriers are:
(a) the associated transaction and administration costs; (b) the
limitations of current information systems in terms of tracking
performance; (c) agreeing on the scheme details (e.g., the appropriate outcome measure or the financial adjudication process); (d)
physician push-back; (e) the “free-rider” problem—other manufacturer or payer competitors may benefit from the information
or schemes developed; and (f) a lack of trust between payers and
developers. The primary transaction costs include the development of processes to track product performance, personnel time
to administer the schemes, and the costs of negotiating these
deals. For a payer to consider implementing a novel reimbursement mechanism, the expected transaction costs must be substantially outweighed by the potential benefits—including sufficient compensation to overcome institutional inertia. As such,
these schemes may be reserved for high-cost drugs with limited
evidence of efficacy. Currently, some health systems give many
such drugs or other medical technologies conditional approval
pending further trials to confirm clinical benefit (e.g., Health
Canada and Centers for Medicare & Medicaid Services).20,21 For
the manufacturers, conditional approval allows early market
access. However, payers may view conditional approval as shifting the costs for some Phase III-IV work onto the payer community. Thus, performance-based schemes may be another mechanism for addressing evidence uncertainty at product launch but
at a reduced level of payer investment.
Many payer information systems remain underdeveloped
in their ability to track clinical outcomes as electronic medical
records still remain a goal rather than a reality for the majority of
U.S. health care payers. Without adequate information systems,
these schemes will be limited as to the types of performance
Vol. 15, No. 8
October 2009
JMCP
Journal of Managed Care Pharmacy 685
Paying for Outcomes: Innovative Coverage and Reimbursement Schemes for Pharmaceuticals
measures that can be tracked and will also add to the administrative cost barriers as they rely more heavily on human effort
and resources. This barrier may be more easily overcome by
large payers with better developed information technology (and
larger resources) and integrated health systems. Agreeing on the
many details of a proposed scheme both large and small can be
a substantial barrier as well. The most important details likely
relate to the appropriate outcome measure(s) and the process for
the financial reconciliation of the scheme. Additional financial
barriers might include Medicaid best-price considerations, especially given the uncertainty in the final price per unit for a new
scheme.
Physician resistance is another possible concern. Physicians,
already pressed for time, are loathe to take on any more responsibilities, paperwork, and effort unless appropriately compensated.
Physicians may also have concerns about being pressured into
prescribing certain products either by their patients, looking for
lower copays, or by the payer, seeking to maximize its discounts
or related incentives. Also, information can be a public good,
and both private payers and manufacturers might be reluctant
to enter into these schemes if competitors will benefit from the
information. First-mover advantage and other network effects
may mitigate this barrier to some extent. The number and nature
of the identified barriers makes it likely that performance-based
contracts will be limited to specific situations in which not only
the incentives are aligned, but also the parties can come to agreement on the key process components under which the contract
operates.
Future Prospects
Performance-based schemes appear to have arisen as a way to
decrease some of the potential negative impacts of uncertainty
in the provision of and payments for medical care by altering the
balance of associated risk between manufacturers and payers.
These potential negative impacts grow as products become more
expensive—a continuing trend, especially in the area of specialty
pharmaceuticals such as those under development in oncology.
Though it is too early to say whether performance-based
schemes will become a lasting trend, they attempt to address
key health policy issues related to increasing cost pressures,
uncertain effectiveness, and value in health care spending. In
fact, the emergence of performance-based schemes may be seen
as another step in the evolution of health care toward one based
on value—in this case, realized value—and the desire to improve
the level of patient benefit per dollar spent. Recent health care
financing reform initiatives, including value-based insurance
design and value-based pricing, are reflective of this movement
toward value-based health care.22,23 In addition, major health care
payers, including UnitedHealthcare, CIGNA, and the NHS in the
United Kingdom as well as large pharmaceutical manufacturers
such as Merck, Novartis, and sanofi-aventis, have demonstrated
a willingness to engage in performance-based schemes. The
686 Journal of Managed Care Pharmacy
JMCP
October 2009
Vol. 15, No. 8
ultimate impact of these schemes in the United States will likely
be limited by the identified barriers, especially those related
to administrative costs. But given the interest shown by the
involved parties to date and the potential to meet the goals of
individual stakeholders—patients, manufacturers, and payers—
performance-based coverage and reimbursement schemes have
the potential to become a more widespread practice in health
care.
Authors
JOSH J. CARLSON, PhD, is a Senior Postdoctoral Fellow; LOUIS
P. GARRISON JR., PhD, is Professor of Pharmacy and Associate
Director; and SEAN D. SULLIVAN, PhD, is Professor of Pharmacy
and Health Services and Director of the Pharmaceutical Outcomes
Research & Policy Program, University of Washington, Seattle,
Washington.
AUTHOR CORRESPONDENCE: Josh J. Carlson, PhD, Senior
Postdoctoral Fellow, Pharmaceutical Outcomes Research &
Policy Program, University of Washington, Box 357630, Seattle,
WA 98195-7630. Tel.: 206.685.6065; Fax: 206.543.3835;
E-mail: [email protected].
DISCLOSURES
This manuscript was sponsored in part by unrestricted funding to the
University of Washington Pharmaceutical Outcomes Research & Policy
Program through the School of Pharmacy Corporate Advisory Board. Sponsors
include Amgen, General Electric Company, Eli Lilly & Company, Genentech,
Novartis AG, F. Hoffmann-La Roche Ltd., Johnson and Johnson Services,
Inc., GlaxoSmithKline PLC, sanofi-aventis, and Pfizer Inc. Publication was
not contingent upon the sponsor’s approval. The authors report a consulting
relationsthip with a pharmaceutical manufacturer in the subject area of this
manuscript.
Carlson performed the majority of work in the literature search, writing,
and revision of the manuscript.
ACKNOWLEDGEMENTS
The authors acknowledge input from colleagues at the University of Washington
(David L. Veenstra, PharmD, PhD; Brian Bresnahan, PhD; Rick Carlson, JD),
the United Kingdom Office of Health Economics (Professor Adrian Towse),
and Tufts Medical Center (Peter J. Neumann, ScD) who provided critical input
on a previous version of the paper. The authors thank the JMCP editor-in-chief
and 2 anonymous reviewers for their helpful comments.
REFERENCES
1. Levitt T. Production-line approach to service. Harvard Business Review.
1972;50(5):41-52.
2. Pickin M, Cooper CL, Chater T, et al. The multiple sclerosis risk sharing
scheme monitoring study—early results and lessons for the future. BMC
Neurol. 2009;9:1. Available at: http://www.biomedcentral.com/content/
pdf/1471-2377-9-1.pdf. Accessed September 24, 2009.
www.amcp.org
Paying for Outcomes: Innovative Coverage and Reimbursement Schemes for Pharmaceuticals
3. Pollack A. Pricing pills by the results. New York Times. July 14, 2007.
Available at: http://www.nytimes.com/2007/07/14/business/14drugprice.
html. Accessed September 24, 2009.
4. Whalen J. Europe’s drug insurers try pay-for-performance. Wall Street
J. October 12, 2007. Available at: http://online.wsj.com/article_email/
SB119214458748556634-lMyQjAxMDE3OTEyMjExNDI0Wj.html. Accessed
September 24, 2009.
5. Moldrup C. No cure, no pay. BMJ. 2005;330(7502):126264. Available at: http://www.pubmedcentral.nih.gov/picrender.
fcgi?artid=558101&blobtype=pdf. Accessed September 24, 2009.
6. DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new
estimates of drug development costs. J Health Econ. 2003;22(2):151-85.
Abstract available at: http://www.ncbi.nlm.nih.gov/pubmed/12606142.
Accessed September 25, 2009.
7. Baker LC, Atlas SW, Afendulis CC. Expanded use of imaging technology and the challenge of measuring value. Health Aff. (Millwood).
2008;27(6):1467-478.
8. Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl
J Med. 2008;359(21):2195-207. Available at: http://content.nejm.org/cgi/
reprint/359/21/2195.pdf. Accessed September 25, 2009.
13. Hornberger J, Cosler LE, Lyman GH. Economic analysis of targeting chemotherapy using a 21-gene RT-PCR assay in lymph-node-negative, estrogen-receptor-positive, early-stage breast cancer. Am J Manag
Care. 2005;11(5):313-24. Available at: http://www.ajmc.com/media/pdf/
AJMC05MayHornbergr313to.pdf. Accessed September 25, 2009.
14. Pollack A. Drug deals tie prices to how well patients do. NY Times.
April 22, 2009. Available at: http://www.nytimes.com/2009/04/23/
business/23cigna.html. Accessed September 25, 2009.
15. Cobden DS, Niessen LW, Barr CE, Rutten FF, Redekop WK.
Relationships among self-management, patient perceptions of care, and
health economic outcomes for decision-making and clinical practice in type
2 diabetes. Value Health. 2009 Aug 20 (epub). Abstract available at: http://
www.ncbi.nlm.nih.gov/pubmed/19695005. Accessed September 25, 2009.
16. Simpson SH, Eurich DT, Majumdar SR, et al. A meta-analysis
of the association between adherence to drug therapy and mortality. BMJ. 2006;333(7557):15. Available at: http://www.bmj.com/cgi/
reprint/333/7557/15. Accessed September 25, 2009.
17. Gordon S. Performance-based contracting. IQ Report 2001:33(6).
Available at http://bookstore.icma.org/product1.cfm?DID=7&Product_
ID=682. Accessed September 25, 2009.
18. Cohen S, Levy RM, Keller M, et al. Risedronate therapy prevents corticosteroid-induced bone loss: a twelve-month, multicenter, randomized,
double-blind, placebo-controlled, parallel-group study. Arthritis Rheum.
1999;42(11):2309-18.
9. National Institute for Health and Clinical Excellence. NICE technology
appraisal guidance 129: Bortezomib monotherapy for relapsed multiple
myeloma. 2007. Available at http://guidance.nice.org.uk/TA129. Accessed
September 25, 2009
19. Felsenberg D, Miller P, Armbrecht G, Wilson K, Schimmer RC,
Papapoulos SE. Oral ibandronate significantly reduces the risk of vertebral fractures of greater severity after 1, 2, and 3 years in postmenopausal
women with osteoporosis. Bone. 2005;37(5):651-54.
10. Anonymous. Merck announces get-to-goal guarantee for patients taking
zocor. November 5, 1998. Available at: http://www.pslgroup.com/dg/bed72.
htm. Accessed September 25, 2009.
20. Anonymous. Notice of compliance with conditions -NOC/c (therapeutic
products). Health Canada. Available at: http://www.hc-sc.gc.ca/dhp-mps/
alt_formats/hpfb-dgpsa/pdf/prodpharma/noccfs_accfd-eng.pdf. Accessed
September 25, 2009.
11. Shearer L. CIGNA and Merck sign performance-based contract. Cigna
news release. April 23, 2009. Available at http://newsroom.cigna.com/article_display.cfm?article_id=1043. Accessed September 25, 2009.
21. Centers for Medicare & Medicaid Services. National coverage determinations with data collection as a condition of coverage: coverage with evidence
development. July 12, 2006. Available at: http://www.cms.hhs.gov/mcd/
ncpc_view_document.asp?id=8. Accessed September 25, 2009.
12. Anonymous. Health Alliance news release. Health Alliance announces
first fracture protection program for Actonel (risedronate sodium) tablets.
Available at: https://healthalliance.org/news_20090414.asp. Accessed
September 24, 2009.
www.amcp.org
22. Chernew ME, Rosen AB, Fendrick AM. Value-based insurance design.
Health Aff. (Millwood). 2007;26(2):w195-203.
23. Claxton K. OFT, VBD: QED? Health Econ. 2007;16(6):545-58.
Vol. 15, No. 8
October 2009
JMCP
Journal of Managed Care Pharmacy 687