Paying for value is all the rage in health care, and recently the spotlight has been brightest on prescription drugs (http://bit.ly/2a0Oy5E). It’s hard to argue with the notion that how much we pay for a drug should be related to the value it provides. Hard to argue, that is, until you try to pin down whose value counts, what value means, or how much to pay for it.
Austin B. Frakt, PhD
Expressions of a drug’s value are implicit in current norms and policies. For example, insurers cover some drugs more generously than others. Many patients prefer to switch from a less-covered brand-name drug to a more fully covered generic version. This also communicates the drug’s value to the insurer in terms its manufacturer can well understand: (http://brook.gs/29B12vK): “Produce more and we will pay for it.”
But insurers rarely link the extent to which they cover a drug to the value it provides to patients (http://bit.ly/1HQoMeV), a value-based insurance design. (http://bit.ly/1ohymvi). They more typically use a price-based design in which a cheaper drug is more generously covered than a more expensive one, even if the latter would provide greater health improvements. This privileges the insurer’s value (to spend less) over patients’ (to pay less for more effective drugs).
The US Food and Drug Administration also communicates value when it grants a certain number of years of market exclusivity to a drug. It’s during those years that a manufacturer can charge the highest price, which is clearly of great value to it. Price, in this case, may not be an accurate expression of a drug’s value to patients, insurers, and health systems.
That raises the question, How should a drug’s value be assessed and reflected in its price?
Several US organizations have developed “value frameworks” to answer that question. Late last year, Peter J. Neumann, ScD, and Joshua T. Cohen, PhD, of the Institute for Clinical Research and Health Policy Studies at Tufts Medical Center, in Boston, summarized the methods of 5 of them (http://bit.ly/1QWajQJ). Reflecting the fact that value means different things to different stakeholders, the frameworks’ goals vary, as do their methods. (Similar frameworks exist in other nations (http://bit.ly/29SKfqk), with perhaps the United Kingdom’s National Institute for Health and Care Excellence being the most familiar [http://bit.ly/1EtB2fV].)
The American College of Cardiology and the American Heart Association issued a statement (http://bit.ly/1RlDkHI) in 2014 about including cost-effectiveness, value assessments, and recommendations in practice guidelines and performance measures “to facilitate the achievement of the best possible health within the constraints of available resources.” Its proposed framework follows a textbook definition of cost effectiveness, measuring the cost of therapies relative to the number of quality-adjusted life-years (QALYs) (http://bit.ly/29KCq7G) that they provide. QALYs are based, in part, on differences in people’s preferences across various states of health.
But combining cost and QALYs into cost-effectiveness ratios has a number of limitations (http://bit.ly/29T7bJ1). One of them is that they give rise to the ordering of treatments that is inconsistent with value as people would normally judge it. People do not always perceive as equivalent 2 therapies with identical cost-effectiveness. Relative to therapies for healthier individuals, society strongly prefers those that offer the same degree of health improvement to those who are more seriously ill (http://bit.ly/29Bfune), even at the same or greater cost. The “rule of rescue” (http://bit.ly/2acYEMV)—the overwhelming tendency to expend disproportionately greater resources to assist people close to death—is the most extreme expression of this phenomenon.
In other words, cost per QALY may be how policy makers or managers of a public program would assign value but not how the populations it serves or that pay for it would do so.
Perhaps recognizing the problems with a purely cost effectiveness–based value framework, other organizations take different approaches. The American Society of Clinical Oncology (ASCO) (http://bit.ly/29zA48l) and the National Comprehensive Cancer Network (NCCN) (http://bit.ly/29Rxldq) both have goals to inform physicians and patients about the value of cancer drugs. They assess a drug’s value across a number of domains of quality and cost. Similarly, the Drug Abacus (http://bit.ly/1fk2up0), developed by Memorial Sloan Kettering’s Cancer Center (MSKCC), is an interactive tool that allows a patient to explore and understand the value of a cancer drug to him or her. Users of the tool apply importance weights to various drug characteristics including efficacy (on survival), toxicity, novelty, research and development costs, target disease incidence, and annual life-years lost to the target disease. What the 3 frameworks—MSKCC’s, ASCO’s, and NCCN’s—have in common is that they are not designed to compute a specific measure of cost-effectiveness.
The goal of the value framework developed by the Institute for Clinical and Economic Review (ICER) “is to improve the reliability and transparency of value determinations made by insurers (http://bit.ly/2ad1p0J).” ICER’s approach does consider cost-effectiveness (cost per QALY) but not to the exclusion of other benefits, disadvantages, and ethical and contextual factors associated with therapies it evaluates (http://bit.ly/29M1me7). These include, for example, whether treatment outcomes reduce disparities across patient groups, whether treatments facilitate greater productivity through, for instance, more rapid return to work, or whether there are any treatment alternatives.
The proliferation of value frameworks and intensity of the conversation about the price of prescription drugs suggest that the way we currently price drugs is unacceptable. In a JAMA Viewpoint, Peter Bach, MD, of MSKCC, New York City, and Steven Pearson, MD, MSc, of the Institute for Clinical and Economic Review, Boston, suggested ways that the output of value frameworks might be used by insurers to align drugs’ effectiveness with cost sharing and adjust lengths of market exclusivity periods according to value, among other ideas (http://bit.ly/29FaUE9). Likewise, A. Mark Fendrick, MD, of the University of Michigan School of Public Health, Ann Arbor, has long advocated that the cost sharing for drugs be lower for those that are more effective (http://bit.ly/1qt1p3R).
All of these are good ideas but with many details to be worked out. Which framework is used to assign value and how that assignment translates into policy change are uncertain, in large part because value is not well defined (http://bit.ly/29VFHiP). It varies among stakeholders and even within them. The value of a drug to an insurer or its manufacturer is not the same as its value to patients. Your value of a therapy is likely not the same as mine. Thus, many might agree that something must be done about drug prices, without agreeing on what that something is.
Corresponding Author: Austin B. Frakt, PhD (firstname.lastname@example.org).
About the author: He blogs about health economics and policy at The Incidental Economist (http://theincidentaleconomist.com) and tweets at @afrakt.
Published Online: July 13, 2016, at http://newsatjama.jama.com/category/the-jama-forum/.
Disclaimer: Each entry in The JAMA Forum expresses the opinions of the author but does not necessarily reflect the views or opinions of JAMA, the editorial staff, or the American Medical Association.
Additional Information: Information about The JAMA Forum is available at http://newsatjama.jama.com/about/. Information about disclosures of potential conflicts of interest may be found at http://newsatjama.jama.com/jama-forum-disclosures/.
Frakt AB. Determining Value and Price in Health Care. JAMA. 2016;316(10):1033-1034. doi:10.1001/jama.2016.10922