Facilitating bias in cost-effectiveness analysis: CHEERS 2022 and the creation of assumption-driven imaginary value claims in health technology assessment

The current standards for health technology cost-effectiveness assessment rest on the creation of lifetime assumption-driven modeled simulations for imaginary pricing and consequent patient access recommendations. A recent BMJ paper reports a detailed assessment of 8,192 cost-effectiveness analyses, concluding that industry-sponsored modeled claims were more likely to publish incremental cost-effectiveness ratios (ICERs) below a USD 50,000 threshold than non-industry sponsored studies, supporting the claim that the product was cost-effective. This is unsurprising; indeed, the opposite can occur with a modeled claim deliberately resulting in ICER is excess of USD 50,000. This methodology is well entrenched with the recently published Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 guidance for creating imaginary cost-effectiveness modeled claims ensuring the opportunity for deliberately manipulated cost-effectiveness claims. This overlooks limitations imposed by fundamental measurement, rendering assumption-driven simulations redundant analytical exercises. Manipulation of ICERs and claims for cost-effectiveness are mathematically impossible; including cost-utility thresholds, because the preference or utilities supporting creation of quality-of-life years (QALYs) are ordinal scores. Nevertheless, with the promotion of CHEERS 2022, the belief in imaginary modeled value claims is both facilitated and reinforced. With CHEERS 2022, there is a concerted effort, largely in self-interest, to perpetuate the current belief system. This is a self-defeating strategy. Rather than admitting to the deficiencies of assumption-driven simulated imaginary claims, leaders are maintaining that health system decision makers can ignore standards of normal science and fundamental measurement in value claims for pharmaceutical products. This disregard of standards that are commonplace in the sciences and mainstream social sciences perpetuates the opportunity for self-serving modeled claims; where models are a marketing vehicle leading to sponsored systematic bias in formulary submissions. This supports the need for a NEW START paradigm for health technology assessment, focusing on evaluable single-attribute value claims, meeting the required standards for normal science and fundamental measurement.