Switching From EQ-5D-3L to EQ-5D-5L in England: The Impact in NICE Technology Appraisals

To estimate the impact of using EQ-5D 5-level (5L) in a sample of National Institute for Health and Care Excellence Technology Appraisals originally undertaken using the EQ-5D 3-level (3L).

For Technology Appraisals case studies published between 2016 and 2024, we extracted utility values used in the cost-effectiveness models and mapped them to 5L using the eq5dmap model with a new value set for UK. Results using 3L and mapped 5L values were compared for oncology, non-oncology with life extension, and non-oncology without life-extension interventions.

We selected 39 appraisal decisions. In almost all cases, EQ-5D values increased after being mapped to 5L, with the magnitude of the increase being greater for more severe initial 3L health values. The impact of using mapped 5L values to model results differed according to the intervention group: absolute incremental quality-adjusted life-years increased (mean = 12.5%) and incremental cost-effectiveness ratios decreased (mean = 11.1%) in the oncology group (n = 17), whereas the opposite and larger effect was observed in the non-oncology without a survival gain group (n = 11, decrease in incremental quality-adjusted life-years = 37.5%, increase in incremental cost-effectiveness ratio = 61.4%). Results were mixed for the non-oncology interventions with a survival gain (n = 11).

Mapping from 3L to 5L increases utility values, with magnitude of change being greater for more severe health states. The impact on estimates of cost-effectiveness varies according to the extent of quality-of-life versus length-of-life improvement but can be very substantial.