Complementing Coverage and Reimbursement Decisions With Multi-Criteria Decision Analysis

This article was co-written by R. Brett McQueen, PhD; Melanie D. Whittington, PhD; Zoltan Kalo, PhD, MD; Roger Longman, MA; and Jonathan D. Campbell, PhD

Introduction to US Pharmaceutical Coverage and Reimbursement Decision Making

Coverage and reimbursement decisions are hardly predictable or consistent within and across US payers. Payers are in the position of determining which drugs to pay for, for whom, and at what level to offer coverage (i.e. formulary coverage decisions), as well as how much to pay for each drug (i.e. reimbursement decisions). Coverage and reimbursement decision making is increasingly complex, in which judgement calls are made regarding numerous treatment options. Generally, this decision-making process involves committees reviewing traditional clinical and economic evidence of a drug and its alternatives. However, these decisions involve considerable subjectivity, in part due to non-traditional evidence criteria, for example, equity. The committee’s subjectivity results in unpredictable and inconsistent coverage and reimbursement decisions within and across payers. While coverage decisions will always involve human deliberation, there are ways to quantify aspects of non-traditional criteria that may be important to a committee’s decision.

Multi-criteria decision analysis (MCDA) is a useful technique to enable more objective decision-making by quantifying and weighting decision-making criteria. It’s an old approach—Ben Franklin described it in a letter to his friend Joseph Priestly, the British polymath and discoverer of oxygen.

He told Priestly he made decisions by dividing a piece of paper into 2 columns: “…writing over the one Pro, and over the other Con. When I have thus got them all together in one view, I endeavor to estimate their respective weights… when each is thus considered, separately and comparatively, and the whole lies before me, I think I can judge better, and am less liable to make a rash step, and in fact I have found great advantage from this kind of equation.”

Franklin’s MCDA approach has been advanced further by economists, mathematicians, software engineers, and operations researchers. It has been used to inform simple to very complex decisions across all sectors of our economy—from forest and environmental planning to investment banking.

The basic requirements of MCDA are fairly simple. First, identify the criteria to be considered in valuing a drug. The criteria should be measurable, identifiable, and distinct from each other. Each criterion should be defined so that a total score can be calculated across all criteria — helping determine the preferred treatment among available alternatives (the highest score wins). And as MCDA has imparted rigor to decision-making in other areas, we hypothesize that MCDA tools can also improve the predictability and consistency of coverage and reimbursement decision-making.

The University of Colorado’s Pharmaceutical Value (pValue) initiative will test this hypothesis and define where and when MCDA can be most useful.

Existing Practices for US Pharmaceutical Coverage and Reimbursement Decision Making

Currently, private payers primarily rely on evidence on the comparative effectiveness and cost-effectiveness of specific drugs to make coverage and reimbursement decisions. Information on cost-effectiveness offers estimates of the cost required to achieve specified improvements in health. For example, CVS Caremark announced in August 2018 that they planned to use cost-effectiveness findings reported by the Institute for Clinical and Economic Review (ICER) to allow clients, such as health plans and employers, to exclude any commonly-used drug launched on the bases of whether the drug has a cost-effectiveness estimate greater than $100,000 per quality-adjusted life year (QALY) gained.

The idea is controversial. The Second Panel on Cost-Effectiveness in Health and Medicine, which provides guidelines and recommendations for the use of cost-effectiveness in health care decision making, suggested that cost-effectiveness is a tool to inform decision makers about the economic value of interventions, not necessarily a strict rule for coverage and reimbursement decision making. Moreover, recent recommendations by leaders in the field of health care decision making suggest that other non-traditional criteria outside of cost-effectiveness could impact decision making, including hope for current and future patients (e.g., opportunity for a cure) and equity or well-being for all health plan members. Notably, ICER panel value votes include deliberations of other potential benefits, disadvantages, and contextual considerations. The explicit impact of such deliberations on value is difficult to tease out. While QALYs and other cost-effectiveness metrics are by definition quantified and remain a starting point for value discussions, the measurement and impact of non-traditional criteria remains understudied.

At present, payers, employers, and other health care decision makers are asking: what decision tools can help us consider a wide range of medical, social, economic, and ethical judgements to help create a sustainable system that provides quality and valuable drugs to everyone in need? Or in other words: how can we make better value-based pharmaceutical coverage and reimbursement decisions?

MCDA as an Emerging Health Care Decision Tool

MCDA is particularly helpful in an area like coverage and reimbursement decision-making, where the available alternatives are characterized by multiple, sometimes conflicting, criteria, some of which are judged objectively, some subjectively, and by multiple decision-makers, each with his or her own views on a particular criterion’s relative importance. MCDA has the opportunity to build in non-traditional elements of value to a decision tool where cost-effectiveness analysis by definition, has fallen short. Criteria such as equity or severity of disease, generally missing from cost-effectiveness analysis, can be included in a MCDA assessment, and thus, be assigned a numeric level of importance. Measuring the value of an intervention in this way can create incentives for pharmaceutical manufacturers to develop new drugs to treat severe diseases with no alternatives or develop interventions that lessen health equity disparities. By creating a system that rewards innovations by including important elements of value that are often overlooked, MCDA can incentivize research into disease areas with unique treatment gaps.

And the impact of MCDA tools improves with repeated use. If MCDA tools are developed for repeated use, they can achieve predictability by communicating and quantifying a criterion’s level of importance and can achieve consistency by applying the same levels of importance across similar drug evaluations.

MCDA has by no means been absent from healthcare with patient-level diagnosis and treatment decisions being the most commonly published cases. In the United States, MCDA tools have been studied in different cancer care settings to help determine appropriate screening guidelines for populations and to help patients and their clinicians weigh options when chemotherapy treatments often include tradeoffs between quantity and quality of life. As patients weigh different criteria to varying degrees, MCDA can be customized to reflect patient preferences and aid in more effective and patient-centered decision making. Outside the United States, MCDA methods are being studied in both patient/clinician shared treatment decision-making level and in coverage and reimbursement decisions. For these and other reasons, MCDA methods are beginning to take hold as an emerging method that can aid decisions related to a treatment’s value.

Leaders in the field suggest that traditional measures of value are a good starting point for discussions and deliberations. Traditional measures of value used in cost-effectiveness analyses are not, by definition, fully comprehensive or sufficiently flexible to allow for the inclusion of all the criteria that patients, payers, clinicians, or other health care stakeholders care about. To get back to our question: how can we make better value-based pharmaceutical coverage and reimbursement decisions? Benjamin Franklin may have had the answer well before we asked the question.