The Use of Risk-Sharing Agreements to Manage Costs, Mitigate Risk, and Improve Value for Pharmaceutical Products


Alex Bastian: GfK Market Access, San Francisco, CA.

Sahand Mirzahossein: GfK Market Access, San Francisco, CA.

Deeksha Dua: GfK Market Access, Boston, MA.


Alex Bastian, 340 Pine Street, Suite 601, San Francisco, CA 94104, Phone: 415-7621052; Email:



Mr. Bastian is an employee of GfK; has received speaking honoraria from AcademyHealth; owns stock in Genomic Health, Celgene, Pfizer, and Roche, and serves as an advisor to American Society of Clinical Oncology and the Medicaid Evidence Based Decisions Project of Oregon Health and Science University. Ms. Dua is an employee of GfK Market Access. Mr. Mirzahossein is an employee of AbbVie, Inc.

Key Words

Abstract: Recent regulatory changes have made it possible for drugs to be approved by the US Food and Drug Administration (FDA) more quickly and with less stringent data, particularly for treatments that address areas of high unmet need. A result of these changes is that health decision-makers have less data to consider when determining a drug’s value and position with the current treatment setting in order to set pricing for the drug. In such situations, payers have engaged in risk-sharing agreements (RSAs), in which the risk is shared by both the payer, which agrees to pay reimbursement costs for the product despite uncertainties about its clinical or health economic value, and by the drug manufacturer, which typically agrees to make a financial- or outcomes-related concession depending on the future performance of the product. In this article, the authors review the types of RSAs that have been implemented to date, with a specific focus on two European countries (the United Kingdom and Italy) and on the United States. The potential benefits and challenges of implementing RSAs in the current US pharmaceutical landscape are also discussed. 

Citation: Journal of Clinical Pathways. 2015;1(1):43–51. 

Received August 13, 2015; accepted September 16, 2015.

Disclaimer: Mr. Bastian, a member of the journal’s Editorial Advisory Board, was not involved in the editorial review or decision to publish this article. 


The modern age of drug development has been accompanied by significant changes in the regulatory system and the accompanying clinical data that is provided for many drugs at the time of launch. Since 1962, the US Congress has mandated that approval of new medicines in the United States would be based on the requirement that any drug be proven through substantial evidence from controlled clinical trials conducted by qualified experts.1 In 2013, the US Food and Drug Administration (FDA) built upon this mandate, adding to countless regulations, amendments, and guidance documents.2

One such change was the amendment to the Food, Drug, and Cosmetic Act that established review deadlines for “priority” and “fast-track” reviews of drugs that represent significant advances or fill unmet needs for serious illnesses, respectively.3 Further exceptions to the traditional requirement that two clinical trials were necessary to demonstrate benefit were subsequently added. With these accelerated pathways, drug manufacturers could utilize data from a single trial, based on surrogate endpoint measures to demonstrate a sufficient level of benefit to justify authorization. However, with this early approval, additional post-approval studies were often required to confirm the benefits and safety for these products. The FDA set the minimum standard for testing drugs for long-term or open-ended use at a minimum of 300 patients observed for at least 1 year or more without a comparison group.4
Related Content
FDA Hematology-Oncology Reviewers Move on to Consult or Work for Biopharmaceutical Companies
The Role of Patient Navigation in Improving the Value of Oncology Care

By 2014, the Obama administration provided more guidance on two additional designations: “breakthrough” and “accelerated review”, providing drug manufacturers with a variety of options to access the market quicker, with less stringent data.5 Although these changes clearly benefit patients in areas of high unmet need, they also raise a set of questions for health decision-makers as they try to grapple with more limited data sets upon which to make decisions related to a drug’s value and position with the current treatment setting.

Europe has a long history of skepticism towards limited datasets such as those currently accepted in the US. The European Medicines Agency (EMA) often requires additional data to meet its requirements for market approval, as do the national payer organizations that conduct technology assessment to determine acceptable pricing for new health technologies. In the face of increasing scrutiny, payers have engaged in agreements, termed “risk sharing,” in which the risk is shared by both the payer, which agrees to pay reimbursement costs for the product despite uncertainties about its clinical or health economic value, and by the drug manufacturer, which typically agrees to make a financial- or outcomes-related concession depending on the future performance of the product.

Risk sharing agreements (RSAs) may be referred to by various terms across the world, but general concepts hold that it is a pre-planned contract based on agreed-upon measures—financial, outcomes, or both—that involves the utilization or performance of a health technology which is tracked in a specific population of patients over a specific period of time.6 Some other names for the same concept that are commonly used in various settings include: pay-for-performance; performance-based risk-sharing arrangements (PBRSAs); outcomes-based schemes; coverage with evidence development (CED); access with evidence development; patient access schemes; conditional licensing; and managed entry schemes/agreements.6


Risk-sharing models can be categorized according to the type of risk the payer or health care decision-maker faces (Figure 1).7 The first type of risk is financial, relating to the perceived value of the product, the utilization of the drug, or resulting budget impact. The second type of risk is performance-related, where the uncertainty of clinical outcomes or benefits of a new technology is in focus. Despite drug manufacturers’ reluctance to solely focus on the financial-based schemes, these are important to consider as they represent novel models that are currently being implemented in the United States and global markets and may help to reduce uncertainty facing other health care stakeholders.

RSA Taxonomy

Financial-Based RSAs

Financial-based RSAs are structured in such a way as to mitigate the risks associated with assumptions of real-world utilization or volume for a given product. These structures typically are either “top-down” (when the cost on the population level is unknown) or “bottom-up” (when the cost on the patient level is unknown). Hepatitis C medications are an example of products for which population-level financial-based RSAs in the United States might be appropriate. Payers were concerned about the budget impact of these medications, because of the sheer number of overall patients and uncertainty with regards to the drugs’ effectiveness. In these cases, a budget cap8 or a price-volume agreement9 could be prudent solutions. The alternative approach, a patient-level financial-based RSA, might be best demonstrated for a medication with highly variable dosing. Recently launched immunotherapies in cancer care are intended for use in the “long-tail” portion of patients that have the best response to such therapies– and thus may require treatment over several years. Another example outside of oncology is Lucentis (ranibizumab) for macular degeneration. In this case, the number of injections required in each eye for eligible patients may vary significantly. Patients requiring extremely high injection rates would therefore be much more expensive to treat than others. The risks associated with such circumstances cause uncertainty for payers with regard to the cost per patient. In such instances, patient-level utilization caps or partial manufacturer funding of initial or excess dosing could accommodate mutual risk sharing.

Performance-Based RSAs

Performance-based RSAs correspond to a different set of objectives, focusing on uncertainty or unclear value related to the benefits provided by a medical technology. These schemes have four general features.6 First, there is a program of data collection. Second, this data collection is typically initiated during the time period following the regulatory approval. Third, the price, reimbursement, and/or revenue for the product are linked to the outcome of this program of data collection (explicitly or implicitly). Fourth, the data collection is intended to address uncertainty about one or more of the following: efficacy or effectiveness in the tested population as compared with the current standard of care; the efficacy or effectiveness in a broader, more heterogeneous population than used in registration trials or in pre-license testing; the effects on long-term or more clinically significant endpoints than those included in registration trials (which—in the case of a drug—may have used surrogate markers) or in pre-licensing studies (eg, for procedures or devices); any adverse effects and adherence issues; whether health care providers’ management of the patient will change the relative benefits and harms under conditions of usual care; the size and value of cost offsets, such as those due to fewer hospital visits; the proportion of patients who will respond (ie, achieve a preset, minimum outcome that may be an intermediate or surrogate endpoint); the numbers and types of patients likely to be treated with the new therapy in real-world practice; and whether the patients treated are the “correct” ones (ie, have attributes matching those patients whom, on the basis of current evidence, the payer is willing to fund, including or not including off-label use).

The two primary types of performance-based structures are conditional coverage and performance-linked reimbursement. With conditional coverage arrangements, coverage is granted conditionally at the initiation of a program of data collection.10 Thereafter, the data will be reviewed to support the continued, expanded, or withdrawn coverage for a medical technology. This can be limited to use only within a clinical trial or may be done alongside a trial to help inform future use.

Performance-linked reimbursement ties the actual reimbursement to a formula relating to pre-specified measures of clinical outcomes in the real-world setting. This might include guarantees of outcomes such as safety or efficacy, but could also relate to response rates and tolerability in newly initiated patients. For example, Velcade (bortezomib) entered into an agreement in the United Kingdom to retrospectively review data in order to identify non-responders, refunding the payer the costs associated with treating these patients. All patients that responded, however, continued on therapy at the normal price of the therapy.11 

Another type of performance-linked reimbursement is for those relating to the pattern or the process of care. In these situations, the payment is linked to the impact of a technology on clinical decision-making or practice patterns. The best example in the United States is an agreement between United Healthcare, a large insurance company, and Genomic Health, the manufacturer of a lab-developed test called OncotypeDx. The promise of this diagnostic test was to predict chemotherapy response, potentially providing a benefit to patients with a positive test result, who receive more personalized treatment, and cost savings to members and the insurance company by foregoing treatment for patients with a negative test result. In this example, UHC agreed to pay for the diagnostic test while data was collected over a period of 18 months. If the test did not result in reduced utilization of chemotherapy, the insurer would negotiate a lower price for OncotypeDx.12


RSAs are a rarity in the United States as compared to certain markets in Europe. US payers may even view these contracts with skepticism, fearing that they may give rise to perverse incentives. Health authorities in the United Kingdom and Italy, for example, have actively engaged in such agreements for years, with mixed results. Further experiments witnessed amongst European HTA bodies, such as those of various regions in Spain, have also begun to open up additional avenues for exploration of RSAs with drug manufacturers. The concept of risk sharing has been brought back into focus in the United States as well, with the recent launch of the heart failure combination drug Entresto (also known as LCZ696) and the public confirmation by its manufacturer, Novartis, that it is actively exploring performance-based RSAs with payers.13

RSAs might help to mitigate some of the challenges faced by health decision-makers in the United States. Clinical pathways have emerged as a popular tool for decision-makers to encourage and reinforce quality, value and efficiency in health care today. These are particularly prominent in areas of medicine in which risk sharing has also been common, and especially in oncology care. Clinical pathways are meant to define appropriate utilization in various treatment settings, and their development relies heavily on the depth of medical literature that is available. In contrast, risk sharing thrives in data-poor environments, in which little published literature or guidelines may exist and the medical technology’s value in that setting could be called into question. This is one of the main reasons for the emergence of RSAs; to ensure that risk is shared early in the entry of a new product when its value is questionable or uncertain.

Payers and health decision-makers tasked with designing clinical pathways should consider risk sharing when they are outlining clinical guidelines for a health plan. For instance, in a clearly defined pathway, would a novel technology with limited data be placed as front-line therapy if accompanied by an RSA? Alternatively, how would an RSA for a particular product not be adversely impacted if the product was niched to a later-line setting where sicker or more comorbid patients may receive the treatment? For this reason, it is worth reviewing some of the RSAs that have been used in two prototypical European markets (the United Kingdom and Italy) as well as here in the US, in order to draw lessons that may be applied to the future use of RSAs in the United States.

RSAs in the United Kingdom

RSAs are common in the United Kingdom, where they are typically referred to as “patient access schemes.” The presence of so many patient access schemes is primarily enabled by the single payer system that forces manufacturers to seek alternative means of justifying value to payers (e.g. the National Institute for Health and Care Excellence (NICE)) at the time of authorization. Since 2010, the United Kingdom has witnessed a consistent surge in patient access schemes, with financial-based schemes leading over performance-based schemes (Figure 2). Financial-based schemes have been introduced for many drugs, with varying models (Table 1).14 More than 80% of the RSAs implemented required a financial discount to the National Health Service (NHS), rather than a focus on tracking of performance outcomes (Figure 3).

patient access schemes

financial based patient access schemes

patient access schemes

Financial-based schemes are sometimes combined with performance-based schemes, as was the case with Votrient (pazopanib), an advanced renal cell carcinoma drug, which was initially contracted at a disclosed discount (a discount on a pharmaceutical’s sales price, negotiated between a manufacturer and a health plan, that is made publically available). Additionally, it was decided that price modification would follow once comparative effectiveness data was available from an ongoing trial.15 Thus, RSAs may evolve over time to account for new evidence or to incorporate newly approved indications for a drug. For example, in 2008, Lucentis (ranibizumab) was initially under a price cap arrangement, but when additional indications (macular degeneration and macular edema) were added in 2012, this was changed to a confidential discount agreement,16 in which the manufacturer sold the drug to the payer at a discounted price that is not disclosed to outside stakeholders. Similarly, Roche–Genentech’s Tarceva (erlotinib) was provided at a disclosed discount to the NHS in 2008,17 but this evolved into a confidential discount scheme when additional indications were added in 2012.18

In the United Kingdom, 34 of the 35 most recent schemes (since October 2011) have been confidential discounts.19 This shift from performance-based to financial-based agreements can be attributed in large part to the administrative burden and high financial costs associated with data collection to demonstrate patient outcomes. The return on investment (ROI) of such RSAs has been frequently called into question.

An example of a performance-based access scheme is for the multiple myeloma drug Velcade (bortezomib), which requires a rebate of the full cost of the drug for patients not responding to treatment. This is measured by a common lab value to define treatment success.20

Another, more prominent example of a performance-based scheme in the United Kingdom is with multiple sclerosis (MS), in which coverage with evidence development was pursued. Initially, NICE rejected disease-modifying MS beta interferon therapies, as they were not considered to be cost-effective over the short-term. However, NICE recognized that the ability of these drugs to reduce disability over long-term may improve their cost-effectiveness in the longer term. A risk-sharing scheme was set up to assess their cost-effectiveness over a 10-year period by prospectively collecting data from United Kingdom patients treated with MS therapies. This scheme was ultimately considered a failure by most observers when the first outcomes report in 2009 showed that disease progression with the new drugs was worse than predicted by the NICE model, and even worse than in the untreated control group.21 The evidence-based scheme was further criticized for not including a price reduction based on these patient outcomes. The total annual cost associated with building and running the scheme was calculated to be close to £50 million, making it one of the most expensive publicly funded health studies in the United Kingdom.21 Additionally, despite the patient access schemes, the United Kingdom has the one of the lowest uptakes of newer MS therapies in Europe, with less than 40% access in England, 36% in Scotland, and 30% in Wales.22

RSAs in Italy

The United Kingdom is not the only major European market employing RSAs. Italy is known for the frequent use of such schemes, referred to as managed entry agreements (MEAs), which are similar to the United Kingdom’s patient access schemes. AIFA, or the Italian Medicines Agency, collects data for monitoring patient outcomes through online registries that are set up on a national level; these maintain records for each prescription given to patients on innovative and specialist drugs, particularly in oncology. Since 2006, these online registries have recorded 78 indications associated with 66 drugs (Figure 4).6

number of risk sharing agreements

Of the 78 indications with monitoring programs, 28 of them involve conditional reimbursement agreements (Figure 5).6 Unlike in the United Kingdom, where straightforward discounts are most common, schemes in Italy are more nuanced. AIFA categorizes schemes in which a price reduction is required until patients show signs of responding to treatment as “cost sharing,” whereas schemes in which the manufacturer reimburses payers for the cost of a drug in patients not responding to treatment are designated “payment-by-results.” “Risk sharing” specifically refers to a scheme that requires manufacturers to reimburse half the costs of treating patients that are not responding to treatment. Also distinct from the United Kingdom, the latter two schemes involve a payback mechanism to the Italian government to compensate for treatment costs, whereas cost-sharing schemes apply general discounts, similar to the schemes frequently seen in the United Kingdom. Of the 28 schemes involving conditional reimbursement, all involve evaluation of treatment efficacy as part of the overall data tracking effort.6

conditional reimbursement

Similar to the United Kingdom, there have been concerns in Italy regarding the return on investment of evidence-based schemes, which are associated with higher costs than financial-based schemes. Cost-sharing schemes are considered to generate higher revenue than risk-sharing and performance-based schemes.23 There are also concerns around dual entry of patient data by physicians, given the sheer number of MEAs in Italy. This reflects the burden of collecting information with the currently inadequate data systems and potentially poor designs of many RSAs in today’s health marketplace.

RSAs in the United States

Experiments with risk sharing in the United States date back to 1998, when Merck agreed to compensate patients and payers for their prescription costs for simvastatin if it failed to lower their LDL cholesterol to target concentration.24 Other RSAs in the Unites States include an agreement between Merck (MSD) and insurer Cigna, in which signs of improved patient adherence to treatment would lead to preferred placement on the latter’s formulary and lower copayment for anti-diabetic drugs Januvia (sitagliptin) and Janumet (sitagliptin and metformin).25

Another example was in the osteoporosis setting. The manufacturer of Aclasta (risendronate), a long-acting agent, failed to show statistically significant reduction in non-spinal fractures in clinical trials. Hip and wrist fractures cost ~$30,000 and ~$6,000, respectively, in the Unites States. As a result of this, the manufacturer crafted a scheme to reimburse the insurer for costs of treating non-spinal fractures experienced by patients who consistently take medication. This reduced pressure on the insurance company to move patients to less-expensive generics and allowed Actonel to keep its preferred tier status (and resulting smaller patient co-payment) compared with competing brand-name drug Boniva (ibandronic acid).26,27 The payer received a performance guarantee and addressed the total cost of care, creating a win-win situation.

Entresto, the aforementioned Novartis drug used for the treatment of heart failure, showed statistically significant mortality-benefit in a clinical trial over ACE inhibitors, the current standard of care.28 Recognizing the low cost of generic treatments available for heart failure, Novartis is publically advocating for RSAs with some health plans to provide Entresto at a discounted rate initially, followed by an additional charge in cases where the drug shows improved patient outcomes, such as reduced hospitalizations.13 Although Novartis’ proposal for Entresto appears to be performance- or outcomes-based, alternative models such as financial-based agreements may address US decision-maker concerns related to the cost impact of this novel therapy.

Financial arrangements have also been implemented related to patient-level price caps based on a variety of features. Avastin (bevacizumab), a Roche–Genentech drug used for the treatment of colon, lung, kidney, and brain cancers, received FDA approval for breast cancer in the United States in 2008 after facing resistance due to a disproportionately low perceived benefit (improved median progression-free survival of 0.9 months) for its estimated annual cost of more than $100,000.29 The company reduced the cost of the drug to $55,000 annually for advanced breast cancer patients who met certain criteria, such as annual income below $100,000.

To date, there have been at least 20 documented RSAs in the United States that are publically acknowledged;30 however, it is likely that many more were executed but in a confidential manner. Most contracts in the United States are between two commercial entities, such as an insurance company and a pharmaceutical company. As discussed elsewhere,31 this is most common for medical devices and surgical procedures. Due to the data rights and protections that either party has within the RSA contract, we are unlikely to ever know the full landscape of RSAs. It also means that the RSAs we are aware of were successful, whereas the failures likely were not publicized.


The rising focus on pay-for-performance in the US health care system may create the impetus for RSAs for pharmaceuticals; yet they are more likely to be unique exceptions rather than an emerging trend. Despite the recent rise in RSAs, particularly in certain European markets, there are many hurdles that have tempered the enthusiasm for this novel pricing model. While certain RSAs, such as Merck’s agreement with Cigna for its diabetes drugs Januvia and Janumet, have been successful in aligning the interests of all stakeholders through a simple scheme, most RSAs are much more complex in execution. The implementation and logistics associated with RSAs can be cumbersome, and the burden of generating the necessary evidence base may not be equally shared between different stakeholders. Collaboration between key stakeholders and sharing of data is essential to the successful implementation of RSAs. The fragmentation of stakeholders in the United States complicates these two requirements.

Collaboration Between Stakeholders

Revolutionary medical technologies such as gene therapies are transforming the health care space, bringing about curative treatments for diseases such as progressive vision loss, hemophilia B, and fat metabolism disorders with a single administration of therapy. Manufacturers such as UniQure, which developed Glybera (alipogene tiparvovec) for a fat metabolism disorder and is approved at a list price of €830,000 in Europe32 and Spark, which is conducting Phase 3 trials for SPK-RPE65, a gene therapy for inherited retinal dystrophies, are exploring alternative pricing models that could help US payers to fully reimburse these expensive yet innovative technologies when they reach the market.32

Yet, serious doubts remain as to whether a risk-sharing model offers the potential solution for such technologies in a system that is more inclined to reward repeated treatments over one-time-cure drugs. In the US, health plan beneficiaries typically switch insurance every few years. Thus, insurers are more focused on short-term health costs because they are unlikely to benefit from long-term health savings produced by curative treatments. Spark’s upcoming single-dose hemophilia drug, SPK-FIX, is projected to be priced at $4–6 million,32 but would a US employer be willing to bear the burden of lifetime costs without reaping equivalent rewards? In such scenarios, a performance-based model of payment might be helpful in not only monitoring the treatment’s efficacy, but also sharing the burden of expenses across subsequent payers who will benefit from the previously treated patient.

In addition to the lack of consensus between payers and manufacturers, prescribing physicians may pose an additional challenge to the execution of RSAs, as they may not be interested typically in investing their time and energy into data collection and reporting efforts. Moreover, obtaining consensus from all stakeholders on scheme details (eg, appropriate patient outcome to be measured, financial adjudication process) can be challenging, because payers and drug developers do not always have the same end goals.24 For instance, the implementation of risk-sharing schemes may face resistance from manufacturers, given the Medicaid “best price” provisions in the United States. Pay-for-performance schemes could reduce the effective price to zero, and that price would then be the “best price,” hence upsetting the entire Medicaid rebate arrangement. Even after overcoming the initial inertia by developing a risk-sharing strategy and building the requisite infrastructure, industry and payers fear that later competitors may “free-ride” and unduly benefit from this information in the future.10

Thus, designing and executing risk-sharing contracts requires concerted efforts from all stakeholders, including payers, manufacturers, patients, and physicians, with substantial investment of personnel, time, and money. The idea behind this approach is to promote innovation in drug development by rewarding new, high-performing therapies, while checking undue spending on drugs that do not prove to be beneficial to patients. These agreements are not only beneficial for patients and payers, who can forego paying for ineffective treatments and receive better therapies as they are developed, but also for manufacturers, who can gain faster market access while the drug is still in trials. Despite these benefits, a key question to consider is whether the payback from these agreements outweighs the costs that go into implementing them. 

Data-Sharing Limitations

In the United States, the lack of data sharing among commercial and public payers, and the limited ability of electronic medical records to track patient outcomes, makes the administration of a large-scale RSA even more challenging.24 During a drug’s developmental stages, the manufacturer is largely responsible for generating evidence through clinical trials, whereas in the post-trial stage, payers and manufacturers both may need to generate this evidence via monitoring outcomes to assess a treatment’s performance in the real world setting. For payers, this could require extensive IT infrastructure and investment. The US health care system does not and cannot currently accommodate complex or sophisticated tracking efforts required for many performance-based agreements, for example, due to lack of connectivity in the distribution system.33 Although data and IT capabilities have improved dramatically over the past decade, fragmentation and poor integration of data sources persists.

The type of outcome measures (i.e. direct versus surrogate outcomes) used to determine adequate “performance” and routine measurement (i.e. standard measurement that is visible with current tracking versus not measured or uncommon) practice must also be considered. In Europe, closed health systems and single-payer markets allow for long-term and complete views of costs and outcomes of care. In contrast, US payers often cycle through patients every few years, and budgets may or may not include direct and indirect cost visibility. This presents unique challenges in the alignment of goals and appropriate measures to which RSAs must be anchored. 

Yet extensive data collection and analysis has been identified as the key to successful disease management.34 As the use of genomics and personalized medicine increases, RSAs could be used in the United States to address the associated challenges, including the rapid pace of innovation, the lack of regulation, the limited evidence for clinical utility, and the uncertain beliefs that genomic tests could have personal utility without having clinical utility. Utilizing this documented experience over the continuum of cancer care, for example, could provide a stronger foundation for the selection of standard of care and critical pathway development as well as enhancing risk-sharing capabilities in providing care for a complex set of patients. In oncology care, electronic medical records and cloud-based companies (eg, CancerLinQ (, OncologyCloud ( offer potential transformations in more linked and holistic data. As big data and more powerful analytics become more seamlessly interwoven into care delivery and process development, the opportunities to utilize this real-world data may enable more complex and sophisticated RSAs by a variety of health care decision-makers.

Finally, for RSAs to be successful, they must be incorporated into clinical pathways to allow prescribing physicians to utilize newer technologies when preferred over treatments that are currently standard of care. But decision makers often find it problematic to incorporate new treatments into clinical pathways because determining exactly when a new treatment becomes necessary and appropriate is a moving target not considered when parties agreed initially on the pathways. In addition to the challenges associated with including RSAs into clinical pathways, payers fear that RSAs might interfere with adherence to clinical pathways by providing incentives, such as lower or no co-pays, that could encourage physicians to prescribe alternative therapies. This is particularly important in oncology care, where pathways and RSAs are likely to cross paths.

Conversely, pathways are important benchmarks for establishing budget-based payment programs. Global payment and episode-based payment models should accommodate various pathways within any given disease or treatment setting. Understanding the various treatment pathways and their corresponding costs, outcomes and complications can inform better, more holistic payment criteria for such arrangements. Initiatives to make data around health outcomes more accessible and to increase transparency in evidence-based decision-making could transform the enthusiasm for RSAs into reality.


As many of the examples from the United Kingdom and Italy have shown, the implementation of RSAs presents several challenges. These would not only be present in the United States as well but may actually be more difficult to manage in our fragmented health system. Despite the complexities of the US health care system being an impediment to RSAs, novel medical technologies will continue to need innovative pricing models such as RSAs to overcome the same shortcomings faced by today’s pharmaceutical products. Based on trends related to the rapid entry of novel technologies in areas of high unmet need, the unlocking of the genome and the explosion of untested or immature genetic markers, and the rapid improvement of data management capabilities, the potential benefits of RSAs will likely outweigh the negative experiences of decades past.

The consequences of not considering RSAs—such as limited access to novel technologies or disproportionate risk placed on payers to fund low-value medications—are too great to forego further experimentation with their use. RSAs could be the cart that leads the horse in the effort to help reframe complex shifts in our system of rewards and definitions of value. Further analysis is justified to examine the role these arrangements play in the current US context. Manufacturers and payers have a role to play in publically disseminating the learnings of such experiments. By learning from the successes and failures, we can begin to address the structural and system-related issues that might prevent RSAs from becoming more common and productive. Such consideration would help all health care decision makers to assess the impact of these novel agreements on the development of pathways and novel payment models in the future. 


1.    21 USC §355. New drugs. Accessed, Aug 10, 2015.

2.    21 CFR §314.3. Applications for FDA approval to market a new drug: definitions. Accessed, Aug 10, 2015.

3.    US Food and Drug Administration. Speeding access to important new therapies—fast track, accelerated approval and priority review. Accessed, Aug 10, 2015.

4.    Expert Working Group (Efficacy) of the International Conference on Harmonisation of the Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH). Guidance for industry: the extent of population exposure to assess clinical safety: for drugs intended for long-term treatment of non-life-threatening conditions. Published March 1995. Accessed, Aug 10, 2015.

5.    US Food and Drug Administration. Guidance for industry: expedited programs for serious conditions—drugs and biologics. Published May 2014. Accessed, Aug 10, 2015.

6.    Garrison LP et al. Performance-Based Risk-Sharing Arrangements—Good Practices for Design, Implementation, and Evaluation: Report of the ISPOR Good Practices for Performance-Based Risk-Sharing Arrangements Task Force. Value Health 2013;16:703–719.

7.    Garrison L, Carlson J. Performance-based risk-sharing arrangements for drugs and other medical products. Accessed, Aug 10, 2015.

8.    Altier J, Hummelt J, Banerjee R, Meiselbach M. “Big Data”: the rising hero of HEOR? Healthcare Insights 2014;7(3):8-11. Accessed, Aug 10, 2015.

9.    Sauvage P. Pay-back and price volume agreements – the French experience. Presented at: European Health Forum Gastein; Gastein, Austria; October 5, 2007. Accessed, Aug 10, 2015.

10.    Remuzat C, Toumi M, Falissard B. New drug regulations in France: what are the impacts on market access? Part 2 – impacts on market access and impacts for the pharmaceutical industry. J Market Access Health Policy 2013;1:20892.

11.    McKee S. NICE endorses shared-risk scheme for Velcade. Pharmatimes. Published June 04, 2007. Accessed, August 10, 2015. 

12.    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 Managed Care 2005;11:313–324.

13.    Pollack A. FDA approves heart drug Entresto said to cut death risk by 20%. New York Times. Published July 7, 2015. Accessed, Aug 10, 2015.

14.    NHS Clinical Commissioning Group. Risk Sharing Schemes (RSS) and Patient Access Schemes (PAS). Published May 2013. Accessed, Aug 10, 2015.

15.    National Institue for Health and Care Excellence. Pazopanib for the first-line treatment of advanced renal cell carcinoma. NICE technology appraisal guidance 215. Published February 2011. Updated August 2013. Accessed August 12, 2015. 

16.    INSEAD. Taxonomy of Risk-Sharing Agreements. Published April 2014. Accessed August 12, 2015.

17.    Daniel Beaulieu. Roche Lowers price of Tarceva in UK. First Word Pharma. Published, Ocotber 3, 2007. Accessed, August 12, 2015.

18.    PMLive. Roche’s Tarceva receives NICE backing for lung cancer. Published May 10,2012. Accessed August 12, 2015.

19.    National Institute for Health and Care Excellence. List of technologies with approved Patient Access Schemes. Accessed, August 12, 2015.

20.    Coulton L, Annemans L, Javier J, Brown R, Keskinaslan A. United BioSource Corporation. Risk-sharing Schemes Worldwide: A Landscape Analysis of Health Outcomes-based Reimbursement Agreements. Presented at: ISPOR 4th Asia-Pacific Conference. September 5-7, 2010; Phukel, Thailand. 

21.    Raftery J. Multiple sclerosis risk sharing scheme: a costly failure. BMJ 2010;340:c1672.

22.    NHS News. Most MS patients in UK do not receive drugs. Pharmaceutical Field. Published April 2013. Accessed, Aug 10, 2015.

23.    Garattini L, Curto A, van de Vooren K. Italian risk-sharing agreements on drugs: are they worthwhile? Eur J Health Econ. 2015;16(1):1-3.

24.    Carlson JJ, Garrison LP, Sullivan SD. Paying for Outcomes: Innovative Coverage and Reimbursement Schemes for Pharmaceuticals. J Manag Care Pharm. 2009;15(8):683-687.

25.    Chu R, Pugatch M, Healy P. Sharing the Burden: Could Risk-Sharing Change the Way We Pay for Healthcare? Amazon Digital Services, Inc.; 2010:1–48.

26.    Carlson, J. The next frontier for risk-sharing agreements in the U.S.: does the changing health care landscape bring new promise for partnerships between manufacturers and payers? Presented at: ISPOR Issues PanelJune 3, 2014; Montreal, Quebec, Canada.

27.    Pollack A. Drug deals tie prices to how well patients do. The New York Times. Published April 22, 2009. Accessed, Aug 10, 2015.

28.    Novartis. Novartis’ new heart failure medicine LCZ696, now called Entresto(TM), approved by FDA to reduce risk of cardiovascular death and heart failure hospitalization. Published July 8, 2015. Accessed, Aug 10, 2015.

29.    Nature Medicine. Avastin gets a breather. Spoonful of Medicine. Published September 21, 2010. Accessed August 10, 2015.

30.    Mohr PE, Tunis SR. Access with evidence development: the US experience. Pharmacoeconomics 2010;28:153–62.

31.    Carlson JJ, Sullivan SD, Garrison LP, et al. Linking payment to health outcomes: a taxonomy and examination of performance-based reimbursement schemes between healthcare payers and manufacturers. Health Policy 2010;96:179–90.

32.    Chen C, Bloomfield D, Kitamura M. Drug Projected at $1 Million Has Spark Mulling Installment Plan. Bloomberg Business. Published August 6, 2015. Accessed, Aug 10, 2015.

33.    Bach P. Indication-specific pricing for cancer drugs. JAMA 2014;312(16):1629–1630.

34.    Stanfill PH, Weber DG, Wayne C. Clinical patient information capture: the Columbia/HCA cancer care database experience. J Oncol Manag. 1998;7(2):17–21.