Pharmaceutical manufacturers—arguably health care’s last holdouts for pay-for-volume—are being pulled into a value-based world. Risk-based contracts between pharma and payers, while still relatively few, are becoming more common, and interest among payers is heating up.
According to the 2016 “EMD Serono Specialty Digest,” 14% of payers in 2015 had at least one outcomes-based contract in place with manufacturers, up from 10% the previous year. Last summer, 63% of health plans surveyed by Avalere indicated “high” or “very high” interest in entering into outcomes-based contracts for the expensive hepatitis C drugs like sofosbuvir (Sovaldi), and 53% said the same for oncology medications.
Wanting to leap in is one thing, actually jumping is another. Experience from the field suggests that measuring the outcomes that underpin risk-based contracts is not always as simple as it sounds. “It’s certainly not easy in the sense that certain conditions take a long time to manifest,” says Wayne Dix, New York-based vice president of SSA & Co. The difference a particular drug is making is also difficult to tease out in complex patients, Dix adds, “and you may not be able to measure it on a patient-by-patient basis but can only reasonably measure it, statistically, over a population of qualified patients.”
MACRA, which makes interoperability among the EMR systems a priority, will eliminate the need for chart-based reviews and will accelerate the writing of value-based contracts, says consultant Wayne Dix.
Moreover, he says, there are many practical challenges to value-based contracts: administrative overhead, the intricacies of data collection and validation, and understanding how or why the data are relevant to the agreement, to name a few.
Cigna, for which value-based contracting is a core business strategy, is one of the most active payers in the area of outcomes-based pharmaceutical contracting. Seven years of experience has enabled Cigna to boil criteria for outcomes-based agreements down to a checklist (see the box below). That checklist, says Chris Bradbury, senior vice president for integrated clinical and specialty drug solutions at Cigna Pharmacy Management, lends itself to outcomes-based agreements across a number of conditions, including diabetes, cardiovascular disease, hepatitis C, and, increasingly, oncology.
In its contracts with pharmaceutical manufacturers, the outcomes Cigna tracks tend to be similar to the endpoints that were measured in the important clinical trials of a drug. For instance, in its agreements with Amgen for its PCSK9 inhibitor, evolocumab (Repatha), and Sanofi and Regeneron for theirs, alirocumab (Praluent), Cigna looks for LDL cholesterol reductions equal to or exceeding those seen in the pivotal trials. If reductions meet the LDL target on a population basis, the negotiated price remains in place. Reductions that come up short trigger a rebate.
But Cigna’s data collection often extends beyond the thresholds in question, says Bradbury. “In the case of PCSK9 [inhibitors], you start to look at event rates, where, if you see cholesterol reduction, are you seeing reductions in hospitalization rates?” This sort of offset is not part of a contract’s financial terms—manufacturers’ legal teams hesitate to price for an outcome that’s not on a drug’s label—but the data-collection exercise helps Cigna learn how a drug may influence real-world, downstream costs.
Cigna’s size and structure as an integrated health plan, PBM, and specialty pharmacy give it an advantage many payers don’t have: a vast data repository that makes the operational challenges of outcomes-based contracting relatively easy. “Depending on the drug class, we may have all of the data within our claims information and other patient profiles,” says Bradbury. But in some cases, he concedes, the company still needs to collect additional data from the individual customer or physician to fill in gaps in claims data. A good example, says Bradbury, is with hepatitis C cure rates. For them, Cigna needs additional data to analyze what’s happening in actuality and get the right population and subpopulation analyses, he says.
For some payers, that can be a heavy lift. For starters, you have to know about the integrity of the data you collect from third parties. “Data quality is almost always an issue when we get involved in a data analytics project,” says Dix. “Assessing whether the data are accurate and cleansing the data often means the difference between having a successful understanding and well-founded insights into the data or not.” Assurance that third-party data were captured correctly are “pretty important,” he adds, when millions of dollars are at stake.
Then there are diseases in which outcomes can be downright hard to measure. Take cancer, where many treatments gain FDA approval on surrogate endpoints, such as progression-free survival (PFS). PFS isn’t always tracked in an electronic health record, however, and when it’s not, a manual record search may be required to determine whether the cancer came back or something else led to early discontinuation of a drug.
Any outcome that is not easily traceable is likely to make risk-sharing agreements impractical, because it will be too time-consuming to figure it out. “The reality is that every single one of these experiments has collapsed under its own weight because the administrative overhead ate up the potential savings,” Express Scripts Chief Medical Officer Steve Miller told a National Comprehensive Cancer Network Policy Summit in 2015, speaking of the complexities of gathering third-party data.
Processes that don’t fit into provider workflows may also jeopardize the ability to collect critical third-party data. Consider a physician who must monitor, say, outcomes and side effects in a subset of patients who take a particular drug. It becomes just one more thing in a clinician’s busy day. “Our experience in health care and other sectors suggests that if the data can be gathered passively, or from sources that are already captured actively at the point of patient engagement, you’re going to be more successful in gathering accurate and complete data,” says Dix.
As data collection and analytics capabilities improve, presumably many of these obstacles will be overcome. Dix believes that MACRA, which makes interoperability among the EMR systems a priority, will eliminate the need for chart-based reviews and will accelerate the writing of value-based contracts.
Another influence, he suggests, is public opinion. “Pharmacy companies have taken it on the chin, rightly or wrongly, in terms of pricing,” says Dix. As patients are exposed to greater cost sharing, he says, they will join payers in demanding that manufacturers stand behind the performance of their products.
Other sectors have learned to shift from mere sellers to problem-solvers, and Dix thinks that’s the promise of outcomes-based contracting for medications.
Accountability is a priority at Cigna, which makes outcomes public after it puts a value-based contract in place. Bradbury believes that the public disclosure and dialogue it generates will create a new form of competition for drugs coming to the market: “We’ve seen, pretty much in every instance when there are more forms of competition, the customer benefits through enhanced innovation and products that deliver more value. We believe these types of agreements are a key component of driving pharmacy benefit management forward.”