Reconciling MCDA and Augmented Cost-Effectiveness Analysis: Do We Get the Same Decisions?

A new paper explores the equivalence of Multi-Criteria Decision Analysis and Augmented Cost-Effectiveness Analysis in adding elements of value to QALY-based cost-effectiveness decision making. It finds that they are equivalent methods under reasonable assumptions when elements of value can be aggregated into two separate top-level groups: health and financial.

A new paper explores the equivalence of Multi-Criteria Decision Analysis and Augmented Cost-Effectiveness Analysis in adding elements of value to QALY-based cost-effectiveness decision making with an opportunity cost threshold. It finds that they are equivalent methods under reasonable assumptions when elements of value can be aggregated into two separate top-level groups: health and financial.

The potential role for Multi-Criteria Decision Analysis (MCDA) in HTA decision making has been much debated. A 2019 review by Baltussen et al., which included OHE’s Martina Garau as an author, noted that “its implementation to date has been criticised for being “entirely mechanistic”, ignoring opportunity costs, and not following best practice guidelines.” A strong defence of MCDA was made by Phelps and Madhavan, who, in an earlier paper, had set out an approach to adapting a cost-per-QALY threshold to use as part of an MCDA framework.

The ISPOR Special Task Force (STF) on U.S. Value Assessment Frameworks recommended using the cost-per-QALY gained as a starting point in deliberations about including new technology in a health plan benefit, offering two major alternative approaches to including additional elements of value deemed relevant by decision makers — augmented cost-effectiveness analysis (ACEA), and MCDA — while emphasising the need to apply the relevant willingness-to-pay (WTP) or opportunity cost threshold rule to operationalise the inclusion decision.

In this paper, Bernarda Zamora, Lou Garrison, Aig Unuigbe and Adrian Towse look at the theoretical equivalence of these two approaches. The MCDA model uses a multi-attribute utility function. The ACEA model is based on expected utility theory. In both the ACEA and MCDA models, value trade-offs for individual attributes are derived in a hierarchical model within two high-level objectives which measure attributes relating to overall health gain separately from financial attributes affecting consumption.

Even though value trade-offs can be elicited or revealed without considering budget constraints, Zamora et al. demonstrate that, on the basis of the preferences of a health insurance enrollee, value trade-offs reflecting indifference as between receiving and not receiving a medical treatment can be used similarly to WTP-based cost-effectiveness thresholds for resource allocation decisions. WTP thresholds are most immediately relevant in health care systems such as the U.S. where the impact of an increase in expenditure from a new technology is on health insurance premiums, and thus on the level of coverage purchased by enrollees. Indeed, some may drop out of insurance cover altogether. The authors find that value trade-offs derived either from ACEA or MCDA move similarly with changes in likely main factors considered by decision makers — the costs of the medical technology, income, and the severity of the disease being treated. Risk attitudes also impact value trade-offs, with risk aversion reinforcing the effects of these factors on value trade-offs. Reconciling decisions in ACEA and MCDA does require, however, that health and consumption are not substitutes – i.e. they are either complements or independent attributes. If they were substitutes, higher consumption would be associated with lower utility from health gains, and the change in value trade-offs would not be predictable. In other words, reconciling ACEA and MCDA decisions requires that good health allows one to enjoy consumption and vice versa, as Smith and Keeney put in their paper “Your Money or Your Life: A Prescriptive Model for Health, Safety, and Consumption Decisions”.

Consequently, this complementarity between health and consumption is a necessary condition for reconciling ACEA and MCDA. The MCDA multi-attribute utility function value trade-off then represents the same concept as expected utility WTP: the optimal cost-per QALY trade-off. Most MCDA models include several health and non-health domains, for example, the four criteria domains in the Advance Value Framework. Zamora et al. conclude that choice between MCDA and ACEA, when used as tools built on the cost-per-QALY metric, is a pragmatic question for decision-makers. They also note that, although their analysis uses a welfare economics framework, they would expect similar results in a budget-constrained health system where a supply-side opportunity cost threshold is relevant in the short run, such as the UK NHS or a U.S. State Medicaid programme.

Most decision-makers do not use, or want to use, algorithms to make decisions – be they ACEA (net monetary benefit) type formulas or a full quantitative MCDA. The ISPOR STF Report recognised the importance of deliberative processes, but proposed decision-makers consider supporting these with explicit frameworks such as ACEA and MCDA that emphasise structure and transparency. As the Baltussen et al. 2019 paper argues, “quantitative MCDA should always include a deliberative component allowing a committee to make a flexible interpretation of results.” Baltussen et al. also make the case for structured decision making, which they term MCDA with Decision Rules, where, for example, the threshold may vary if certain criteria are met, e.g. vary with severity, as in the case of the Netherlands. Arguably, ICER’s taking of votes on the relevance and importance of specific categories of potential non-QALY other benefit or disadvantage and other contextual considerations in an assessment falls into this category.

Debate will continue as to (i) which elements of value are relevant in decision making in addition to health gain and cost savings, (ii) how these elements should be evidenced and assessed, and (iii) how much discretion should be left to the appraisal committee. However, this paper shows that MCDA anchored on the QALY with an opportunity cost threshold is an equivalent method to the use of QALY-based ACEA.

Acknowledgement

This work was funded by the Pharmaceutical Research and Manufacturers of America Foundation (PhRMA Foundation) via a grant award to the University of Washington (P.I. Lou Garrison) for the project “Implementing Augmented Cost-Effectiveness Analysis: Challenges and Next Steps”.

Citation

Zamora, B., Garrison, L.P., Unuigbe, A., and Towse, A. (2021).  Reconciling ACEA and MCDA: is there a way forward for measuring cost-effectiveness in the U.S. healthcare setting? Journal of Cost Effectiveness and Resource Allocation. 19:13 https://doi.org/10.1186/s12962-021-00266-8

Related Research

Garrison, L.P. Jr, Neumann, P.J., Willke, R.J. et al. (2018). A health economics approach to U.S. value assessment frameworks—summary and recommendations of the ISPOR special task force report [7]. Value in Health; 21:161–5. https://www.sciencedirect.com/science/article/pii/S1098301517338949?via%3Dihub

Baltussen, R., Marsh K., Thokala P.,et al. (2019).  Multicriteria Decision Analysis to Support Health Technology Assessment Agencies: Benefits, Limitations, and the Way Forward.  Value in Health, 22 (11): 1283-1288. https://doi.org/10.1016/j.jval.2019.06.014

Garau, M. and Devlin, N. (2017). Using MCDA as a decision aid in Health Technology Appraisal for coverage decisions: opportunities, challenges and unresolved questions. In “Multi-Criteria Decision Analysis to Support Healthcare Decisions” (Eds.).Marsh, K., Goetghebeur, M., Thokala, P., Baltussen, R. (2017) Springer.  https://link.springer.com/chapter/10.1007%2F978-3-319-47540-0_14

Garau, M., Hampson, G., Devlin, N. et al. (2018). Applying a Multicriteria Decision Analysis (MCDA) Approach to Elicit Stakeholders’ Preferences in Italy: The Case of Obinutuzumab for Rituximab-Refractory Indolent Non-Hodgkin Lymphoma (iNHL). PharmacoEconomics Open 2, 153–163. https://doi.org/10.1007/s41669-017-0048-x

Marsh K., et al., (2016). Multiple Criteria Decision Analysis for Health Care Decision Making—Emerging Good Practices: Report 2 of the ISPOR MCDA Emerging Good Practices Task Force. Value in Health, 19, 125 – 137.

Thokala, et al., (2016). MCDA for Health Care Decision Making – An Introduction: Report 1 of the ISPOR MCDA Emerging Good Practices Task Force. Value in Health, 19, 1 – 13.

Sussex, J., Rollet, P., Garau, M., Schmitt, C., Kent, A. and Hutchings, A., (2013). A pilot study of multicriteria decision analysis for valuing orphan medicines. Value in Health, 16(8), pp.1163-1169.

Devlin N., Sussex J.: (2011).  Incorporating Multiple Criteria in HTA – Methods and Processes. Office of Health Economics.