Value-Based Contracting for Pharmaceuticals: Getting Ready for Prime Time?

Published on: 
Evidence-Based Oncology, March/April , Volume 19, Issue SP3

The full-court press toward comparative-effectiveness research and results reporting has compelled health plans and insurers to sharpen their focus on incorporating this information into coverage decision making. Managed care executives have long complained that too few evidencebased data of this type are available to improve coverage decisions. If the evidence were available and payers were ready to incorporate it, widespread development of value-based formularies might result.

Any pharmaceutical manufacturer whose small-molecule drug is the third or fourth product of a category to reach the market faces its own value-based quandary: How can it prove the value proposition of its agent in the face of heavy competition? Drug rebates and discounts can provide access to a plan’s membership, but this does not truly address the economic value of a drug. The question is even more difficult (and scrutinized) in the case of expensive biologic medications. A survey published in 2012 demonstrated that managed care executives are not satisfied with the current approach of contracting and discounts to obtaining formulary access.1 Only 4% indicated that they were “very satisfied” with this form of contracting.

One innovation that has yet to become mainstream involves manufacturers offering to back their product’s performance by taking some of the risk. This does not necessarily mean a “money-back guarantee,” but it certainly could. This is in stark contrast to the present situation, where the manufacturer’s “risk” has been defined by the difficulty in successfully navigating the drug development process from preclinical studies through US Food and Drug Administration (FDA) approval, and subsequently, the risk in marketing the drug successfully to payers. This risk must not be minimized—according to past reports from the Tufts Center for the Study of Drug Development (www., the cost to develop biotechnology drugs was approximately $1.2 billion in 2006. In addition, the pharmaceutical industry spends hundreds of millions of dollars on marketing efforts to increase awareness and focus on their products’ benefits.

Today’s price-volume agreements or patient-access initiatives (where patients would be given access to a drug at reduced cost to try to prove value to payers, in the hope of landing preferred formulary positioning) are a form of “value- based contracting,”2 but if outcomes are what the payers are after, these contracts are low on the evolutionary scale. “Outcomesbased contracting” represents the next evolution, encompassing risk-based agreements with performance guarantees or payment scales for attaining not utilization or market share benchmarks, but clinical status improvement or prevention of adverse events. This is truly value-based contracting. It boils down to the following: if the drug doesn’t provide the outcome that is expected, the manufacturer would be at risk in some way—reimbursement would be discounted or possibly even free, or the manufacturer would bear the cost of complications suffered by the patient. It may make it more palatable for payers to cover an expensive agent, and gain credibility for the maker of a specialty drug.

If the pharmaceutical companies will need to eventually move away from volume-based sales, it would be reasonable to expect “the industry to shift into a world in which outcomes-based contracting will be the standard reimbursement methodology,” according to one pharmaceutical executive.1 Yet, this concept has not spread to date. Pay for performance has been utilized for provider reimbursement for a decade.3 Why not apply this concept to pharmaceuticals? Assuming one does not run afoul of federal anti-kickback statutes, and the remuneration, discounts, or incremental reimbursement resulting from an outcomes-based contract that can be coordinated under an accepted safe harbor,4 then one can explore some of the outcomes-based pharmaceutical contracting options being discussed today.

Specialty Pharmaceuticals Shine a Light on the Issue

The costs associated with specialty pharmaceuticals continue to concern payers and represent a difficult challenge for the years ahead. Specialty pharmaceuticals today account for 80% of the total drug spend in the medical benefit, and 8 of the top 10 drug categories in 2014 are forecast to be specialty drugs.5 In the future, specialty utilization is expected to grow rapidly, owing to new drug introductions and rising numbers of patients with chronic diseases for which specialty drugs may be indicated. Robert Kritzler, MD, deputy chief medical officer, Johns Hopkins HealthCare, Glen Burnie, Maryland, believes that “with the rising cost of specialty pharmaceuticals, especially those with variable outcomes based on patient selection, there will need to be a different way of contracting.”

Payers are not yet convinced of the value proposition behind these products. They are actively seeking reassurances that an agent that may cost $10,000 for each month of therapy is money well spent. This is partly behind the push for companion diagnostics, to help determine those patients who may optimally benefit from the use of specific specialty drugs. It is also responsible for the interest in performance guarantees and other ways to link a drug’s costs to its actual outcome in the patient.

“The pharmaceutical and biotech industries have done a bad job of clearly stating the value equation and lifesaving value of their products,” asserted Gerald Clor, MBA, of Clor Training & Consulting, LLC, Bethlehem Township, Pennsylvania. “Health plans do not want to be seen as ‘being too close’ to such a negative industry—especially when it is a business proposition.”

Clor continued, “This potential to help patients, improve care, and lower costs is partially lost owing to the ‘politics’ of the current state of affairs within the pharmaceutical industry. Too many big pharmaceutical companies have been caught behaving badly and brought into the public eye. How then can a health plan expect to partner with them? So politics and public sentiment trump good medicine in this case.”


Myeloma Drug Pioneers the Concept

In 2007, the prototype value-based or performance-based contract for a drug was introduced in the United Kingdom by Johnson & Johnson.6 The National Institute for Clinical Excellence (NICE) had deemed its product Velcade provided too little value for the cost to recommend coverage by the National Health Service. The company, worried about being locked out of the British market, offered a radical contract: The use of the drug must achieve a 50% reduction in serum paraprotein (M-protein) levels by the fourth cycle of therapy or the manufacturer would reimburse the National Health Service the full cost of the drug. Technically, this compensation was most commonly made in the form of additional free Velcade.2 The decision makers at NICE took them up on their offer.

The first value-based pharmaceutical contracts in the United States were seen in 2009. One well-publicized contract was between Cigna and Merck for the diabetes products sitagliptin and sitagliptin/ metformin. Merck provided discounts if Cigna members with type 2 diabetes mellitus lowered their blood sugar levels, and also provided for additional discounts if people who were prescribed Merck’s drugs took their medications according to their physicians’ instructions. In return, Cigna placed the agents on a low copayment tier, in the hopes of optimizing adherence to the medication.6 According to a Cigna press release, “The results demonstrated improved blood sugar levels of more than 5% for those continuously enrolled in the program, regardless of which diabetes drug they were taking. There was also a 4.5% increase in blood sugar lab testing during the period.”7

Also in 2009, Health Alliance Plan contracted with Procter & Gamble and sanofi-aventis on the osteoporosis drug Actonel (risedronate). In this deal, the manufacturers agreed to reimburse Health Alliance Plan for the costs of treating nonspinal, osteoporosis-related fractures in eligible postmenopausal women who were using the product, up to a predefined amount, which greatly limited their financial risk. This gamble worked well, as the effectiveness of the osteoporosis drugs in general were heavily questioned at this time, and the arrangement ensured health plan members had access to risedronate. Nine months after starting the program, the manufacturers found that they had paid far less (79% less) than the predefined limit established in the contract. In addition, the data gained corroborated Procter & Gamble’s original clinical trial data for risedronate’s efficacy in preventing nonspinal fractures.8 Furthermore, it may have delayed the plan’s generic substitution for risedronate, as well as kept it in preferred position relative to the newer branded product Boniva (ibandronate).

In general, non-US payers are interested in seeking alternative methods of payment for pharmaceuticals. Although NICE is currently utilizing cost-effectiveness thresholds in making coverage recommendations, it is seeking to revamp its decision making as it applies to any new drug by 2014, and include other factors as well, such as unmet need, the burden of illness, degree of innovation associated with a new medicine, and additional societal benefits gained with the new medicine. This value-based pricing system will supplant the existing system used by the National Health Service when evaluating all new medications.9

Australia, Germany, and Italy have all undertaken several initiatives aimed at increasing the number of outcomesbased pharmaceutical contracts.1 The Australian drug pricing authority has at least 90 contracts of this type on the books. There is no lacking for motivation among payers worldwide. It is only a matter of time before US payers are greater players in this area.

Hurdles to Overcome in Outcomes-Based Pharmaceutical Contracting

If 1 word characterizes the nature of outcomes-based contracting for pharmaceutical drugs, it is “complexity.” Gerald Clor believes that multidrug therapy, promoted by many clinical guidelines, muddies the waters quite a bit. “It becomes very difficult to measure the impact of clinical improvement when there is more than 1 single aspect to that improvement,” he said. “Few companies will want to enter into a riskshared contract if the measure of success goes past the drug monotherapy.”

Just coming to agreement on which outcomes measures are actually of importance is a notable obstacle. For example, according to one survey, payers focus on measures of longevity and quality of life, but pharmaceutical manufacturers are focused on specific units of clinical efficacy or the “cost-benefit implications of a new drug for overall treatment.” Furthermore, healthcare providers and patients may point to other preferred outcomes measures.9

Adding to this measure of complexity is the desire to move toward new payment methods. “If a ‘bundled’ or ‘global’ payment is used, and the goal is to lower the total cost of care,” Clor wondered, “how do I measure my (pharmaceutical) impact on that cost parameter? This complication adds too many variables and components that the contract (or the drug company) cannot control.”

These challenges are difficult to overcome from a pharmaceutical manufacturer’s perspective. From the health plan perspective, the complexity of these arrangements is no less an issue. Dr Kritzler of Johns Hopkins Health Care explained, “Many plans, including our own, have considered some sort ofvalue-based contracting based on final outcome. Areas that many of us have considered are in the oncology, neurology, and the autoimmune space,” he commented, “but plans are having a rough time implementing this, as the details are extraordinarily complex.”

Not only are the details to implementation complex, but the administration of these types of contracts can be barriers to payers. Assume that they would need a tracking mechanism for particular patients’ clinical status over time. In some cases, they would save less money from the outcomes-based contract than it would cost to launch and manage the patient tracking system needed to analyze the intervention’s effects.

However, Dr Kritzler believes that once the ball begins rolling, it will be hard to stop. “As soon as a few of these arrangements gain traction, it is likely that this type of contracting will take off,” he said.

Oncology-Specific Questions

The idea of outcomes-based contracting for oncology products adds a new level of complexity to an already challenging issue. How to measure outcomes becomes not only a question about the value of overall survival versus progression-free survival or complete response versus clinical response as a clinical or patient-related outcome, but consider the next stage of discussion: Would the terms of the outcomes- based oncology contract change for patients diagnosed at a different stage of cancer? For example, should the successful outcomes of a drug given in patients with stage III tumors be judged differently from the outcomes of a drug given in patients with stage I or II disease? These permutations can boggle the mind. In addition, how might a contract define objective result reporting?

With these questions in mind, it should not be surprising to see initial outcomes- or value-based contracting efforts focused on other diseases. Although oncology is an area that may optimally benefit from this approach— many primary drug treatments qualify as specialty pharmaceuticals—it may not be practical for many cancers.

Still, Johnson & Johnson successfully implemented and administered its Velcade contract with NICE, so there is some experience to build upon. In this example, M-protein may not be a good outcomes indicator of survival.2 Furthermore, up to 15% of all patients do not have measureable M-protein levels.2

A value-based approach might help advance an approach that emphasizes biomarker testing before treatment, to help identify who would be most likely to benefit. Additionally, such contracting could be an incentive for manufacturers to engage in greater efforts to educate providers as to who might be the most appropriate subpopulations of patients for these expensive agents.

The Pace of Change Is Slow

The industry has just begun to consider what payers mean when they say they want more value for the money they spend. A survey by a unit of The Economist found that the biopharmaceutical industry is a harsh critic of the progress made to date (Figure).10 The survey report specifies that “More than one-half of respondents overall (55%) say that the sector is adjusting well to increasing demands for proof of value, but only 36% from traditional biopharmaceutical companies agree.”10 Furthermore, just over half (56%) believe that “The industry will introduce products with demonstrably more value than existing offerings in the next 3 years. Worse still, only 39% believe that the industry is more than just somewhat effective at creating such products.” Not insignificantly, the survey respondents didn’t believe that the industry does a good job of proving value, even when their products may be of real value. Only 25% of payers and government policy makers interviewed expressed a level of confidence about claims of value made by the industry.10

Based on PriceWaterhouseCoopers’ analysis, it is imperative that payers and pharmaceutical manufacturers hold numerous discussions and hold them early in the contracting process “to determine what is of value to each stakeholder.”1 This may be augmented by developing formal, well-defined, consensus-driven value metrics, and by obtaining a consensus on a “common set of principles, policies, and technical methods for the data collection program.” Without this extremely high level of agreement on what constitutes an outcome of value to both payer and manufacturer, the vast effort to develop outcomes-based pharmaceutical contracts (for any disease category) will be destined to fail.

Payer PerspectiveOutcomes-Based Contracting for Pharmaceuticals: A Health Plan PerspectiveJeffery D. Dunn, PharmD, MBA

Outcomes-based contracting for pharmaceuticals has been discussed for years as an incentive to offer preferred formulary positioning for certain drugs and to lower costs. Drug categories and their associated outcomes have included diabetes agents and reductions in hemoglobin A1C; bisphosphonates and decreased fracture rates; statins and cholesterol lowering; and even guarantees against dose escalation in specialty categories like rheumatoid arthritis and psoriasis.

However, true outcomes-based contracting is elusive. Many healthcare plans are incapable of capturing, measuring, and reporting the data elements necessary to comply with the contract terms. Even integrated plans lack the information technology to adequately marry medical claims with pharmacy data. Furthermore, most contracts would require plans to manage down to the individual member level rather than use high-level aggregate data that are typically required for rebates.

It is also often difficult to identify and agree on what outcomes will be measured. It can be done with objective measures such as low-density lipoprotein and hemoglobin A1C levels, but when exploring more intriguing disease states, such as multiple sclerosis, rheumatoid arthritis, and oncology, the outcomes are often subjective. For example, a plan cannot be expected to define what is progression-free survival in a patient with metastatic prostate cancer or what is a clinically significant relapse in multiple sclerosis.

Plans also often find the investment in maintaining and complying with the contracts not worth the return warranted by the effort expended, partly because the pharmaceutical manufacturers have never gone truly at risk. Using a previously mentioned example, a bisphosphonate contract was touted as paying for the cost of a fracture if patients were compliant with the drug therapy. However, there was a cap on the exposure to the manufacturer—a maximum payment was set at a percentage of plan spend on the drug. This was substantially less than the total cost of any 1 fracture. For the diabetes and statin examples, the manufacturers offered a few extra percentage rebate points if a plan’s population met certain A1C or lipid goals. Often this results in a few thousand dollars in incremental rebates, which needs to be weighed against the effort required to code and measure the input data (not even considering the requests to stratify risk, which can’t be accomplished through claims databases). There is also the issue of how a plan addresses, and invests in, the education or incentives necessary to change both prescriber and member behavior to improve the respective lab values or outcome measures. These measures and activities would often have to be above and beyond what is occurring for quality measurements such as HEDIS or Medicare Star ratings.

Despite the poor history of these types of contracts, they will continue to be explored, owing to the shift toward specialty drug management. It will be more feasible to implement outcome-based contracts in these categories, because the sample size is much smaller and can be more easily measured at a member level, and the cost of the drugs is significantly higher, making the return to the managed care plan potentially much higher.

In conclusion, most plans cannot comply with outcomes-based contracts. This is not for lack of desire but rather operational or philosophical shortcomings. However,

we need to figure out a way for all stakeholders to share risk as we move toward more niche specialty drugs, new models of healthcare delivery, and payment reform. Pharmaceutical companies should have an active role in backing up the efficacy, or lack thereof, of expensive medications that currently are paid for by plans and members. The government, FDA, or other body may play a role in requiring manufacturers to provide a value proposition for all medications similar to that played by the National Institute for Clinical Effectiveness in the United Kingdom. All stakeholders (plan, provider, member, manufacturer, and government) must have a role in delivering value in medications and delivering cost-effective healthcare.

Dr Dunn is formulary and contract manager for SelectHealth in Salt Lake City, UT.Funding Source: None.

Author Disclosure: Mr Mehr reports receiving payment for involvement in the preparation of this article.

1. Unleashing value: the changing payment landscape for US pharmaceutical manufacturers. PriceWaterhouseCoopers Health Research Institute. Published May 2012. Accessed February 1, 2013.

2. Adamski J, Godman B, Ofierska-Sujkowska G, et al. Risk sharing arrangements for pharmaceuticals: potential considerations and recommendations for European payers. BMC Health Serv Res. 2010;10:153-169.

3. 2011 annual report. Integrated Healthcare Association, Oakland, California. pdfs_documents/resource_library/IHA_AnnualReport_ 2011_Final.pdf. Accessed January 25, 2013.

4. Lutes ME, Gottlieb DG. Managing regulatory risk in value-based contracting. Presented at Value-Based Contracting and Risk-Sharing Agreements, Sponsored by CBI. Philadelphia, PA: January 30-31, 2013.

5. 2011 care insights: changing rules changing roles. Scottsdale, AZ: CVS Caremark; 2012.

6. Pollack A. Drug deals tie prices to how well patients do. New York Times, April 23, 2009. business/23cigna.html?_r=0. Accessed February 2, 2013.

7. Cigna and Merck help customers better manage diabetes (press release). Cigna, October 28, 2010. cfm?article_id=126. Accessed February 14, 2012.

8. Issue brief: value-based pricing for pharmaceuticals: implications of the shift from volume to value. Deloitte Center for Health Solutions; 2012. Accessed February 1, 2013.

9. Faden RR, Chalkidou K. Determining the value of medications—the evolving British experience. N Engl J Med. 2011;364:1289-1291.

10. Kielstra P. Reinventing biopharma: strategies for an evolving marketplace—the value challenge. The Economist Intelligence Unit. April 25, 2012. Accessed February 19, 2013.

Authorship Information: Concept and design; drafting of the manuscript; critical revision of the manuscript for important intellectual content.