Is NICE Too Severe With Severity? Exploring How Well Its Severity Modifier Aligns With UK Preferences

OBJECTIVES: NICE, similar to other HTA bodies, has implemented a severity modifier that assigns greater value to health gains to patients with greater absolute (AS) or proportional (PS) health shortfalls. Given NICE’s commitment to accounting for societal preferences in its methods, this study sought to understand how well the current modifier aligns with UK preferences.

METHODS: We conducted a stated preference study with a representative sample of the public in England & Wales (n=997). We elicited preferences over: 1) thresholds for ‘severe’ and ‘very severe’ categories as defined by AS and PS, and 2) the relative value of a QALY gain to patients in different categories of severity using a person trade-off (PTO) approach.

RESULTS: The mean PS cutoffs for ‘severe’ and ‘very severe’ were approximately 50% and 65%, respectively, compared to NICE's thresholds of 85% and 95%. The PTO showed a steady increase in the value of treating more severe patient groups, even at relatively moderate levels of shortfall. These plateaued around 65% PS at a median weight of 1.75. The median weights from the PTO were higher than NICE’s weights across the range of severity, except at the highest shortfall tested (95% PS). Here, the public assigned less weight to 95% PS than the preceding 85% PS, and less than NICE’s current weight for 95% PS.

CONCLUSIONS: These results suggest that the public’s preference for prioritizing severity begins at a substantially lower threshold than currently applied by NICE. Likewise, the sample assigned greater weight to more severe groups at almost every level of severity. Our results suggest a concave value function, with the relative value of health gains peaking around 65% PS rather than at the most extreme shortfall. Together, these suggest NICE’s modifier may undervalue severity in terms of thresholds and weights relative to public preferences.