Presentation (CTI)

OBJECTIVES: A NICE-commissioned impact assessment (Biz et al. 2026) found the new UK EQ-5D-5L value set would affect cost-effectiveness unevenly, lowering ICERs in oncology but raising them substantially for non-oncology interventions without a survival gain. We hypothesised these same conditions are disproportionately those where EQ-5D struggles to capture quality of life, such that companies have often sought alternatives, but NICE has rarely accepted them, leaving these therapies doubly disadvantaged: a measure already considered limited, now valued less favourably. We examined how often appraisals depart from EQ-5D, their justifications, and committee responses.
METHODS: All NICE STA/MTA appraisals published since January 2022, when the current evaluation manual (PMG36) was introduced, were screened; Highly Specialised Technology appraisals were excluded. Appraisals whose base-case model used a primary utility source other than trial-collected EQ-5D were extracted, and committee papers reviewed to classify departure type, instrument, mapping approach, evidence, and decision.
RESULTS: 33 of 285 appraisals (11.6%) departed from EQ-5D: 10 oncology (30%) and 23 non-oncology (70%; 52% with survival gain, 48% without). Of these, 19 mapped a non-EQ-5D measure onto EQ-5D, while 14 used alternative valuations (vignettes, condition-specific preference measures, literature) that bypass EQ-5D entirely. In oncology, every departure occurred because EQ-5D was not collected in the trial, with consistent acceptance. In non-oncology, roughly half were similarly driven by EQ-5D absence; the remainder had EQ-5D available but argued it was not the most relevant source. Among nine explicit inappropriateness arguments, seven cited insensitivity or face-validity failures, two cited an omitted domain. Supporting evidence rarely met the manual's requirements. Domain-omission arguments were generally accepted; insensitivity arguments were variably accepted, rejected, or deferred.
CONCLUSIONS: Explicit challenges to EQ-5D concentrate in non-oncology conditions and are variably accepted. These are the conditions the impact assessment identifies as most disadvantaged by the new value set, suggesting a compounding effect warranting attention during adoption.