The impact of statistical properties of incremental monetary net benefit and incremental cost-effectiveness ratio on health economic modeling choices

ABSTRACT

Introduction

There is controversy on whether to use incremental monetary net benefit (INMB) or incremental cost-effectiveness ratio (ICER) in health economic evaluations alongside randomized controlled trials. We studied the impact of restricted mean survival time (RMST) on the long-term projection of INMB and ICER.

Methods

We analyzed the unbiasedness and efficiency of ICER and INMB by (1) deriving the metrics' expected values and variances based on theoretical probability distributions, (2) simulating their 15-year post-trial projections based on between-arm-RMST-gained through a 2 × 4 × 2 factorial experiment of Markov 2-state microsimulations. Simulations and comparison were run on the data from the Cardiovascular Outcomes for People Using Anticoagulation Strategies Study (COMPASS).

Results

Our simulation findings using RMST showed that ICER was more efficient than INMB, regardless of disease populations, time horizon, modeling choices, and underlying probability distributions of incremental mean cost and effect. ICER had a small variance and thus showed its robustness to the choices of models.

Conclusion

INMB is unbiased, while ICER is biased-but-remediable. INMB’s variance varies with a willingness-to-pay (WTP) threshold value quadratically while ICER’s variance varies with incremental-mean-cost quadratically. A simple and naïve model can sufficiently estimate ICER. Future metrics are expected to be health-economic-meaningful, unambiguous, unbiased, efficient, and statistical-inference-friendly.

KEYWORDS:

• Restricted mean survival time (RMST)
• a two-state Markov micro-simulation model
• incremental cost-effect ratio (ICER)
• incremental net monetary benefit (INMB)
• unbiasedness and efficiency

Acknowledgments

The authors want to acknowledge the investigators of the Cardiovascular Outcomes for People Using Anticoagulation Strategies Study (COMPASS) study for providing access to the data for this investigation. The authors thank Dr. Janet Wittes for the nice discussion and suggestion. The authors also express deep appreciation to reviewers for their wonderful comments and suggestions that expand the width and depth of the article.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

F Yuan (i) originally proposed two theorems and provided proofs about the statistical properties and their mutual relationship of INMB and ICER; (ii) designed, implemented and analyzed Markov 2-state simulation models through the 2 × 4 × 2 full factorial experiment; and (iii) wrote the paper. S Bangdiwala critically read the proofs and contributed to the discussion and manuscript writing. W Tong and A Lamy proposed to use a Markov 2-state modeling approach in simulations and provided the data of unit costs. All authors read and approve the final manuscript for publication.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14737167.2023.2144838

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This paper was not funded.