The UK’s National Institute for Health and Care Excellence (NICE) has published an update to its methods and processes for health technology assessment. The update follows a year-long review and includes a number of significant changes to how innovative medicines will be evaluated for pricing and reimbursement in the UK.
One key change is the introduction of a “severity modifier”. Previously, all quality-adjusted life years (QALYs) were considered of equal value in NICE decision-making, with the exception of “end-of-life” therapies and treatments for ultra-rare diseases, assessed via the highly specialized technologies track. NICE is now introducing a severity modifier, which will place increased value on QALY gains in severe diseases. For conditions considered more severe, QALY gains will be upweighted by a multiplier of 1.2 or 1.7. Eligibility for the multiplier will be determined by lifetime QALY shortfall, absolute or proportional, for the eligible patient population compared to the general population.
The proposal to replace the current end-of-life modifier with a new severity modifier will provide a broader definition of severity that considers quality as well as quantity of life, acknowledges high burden in chronic, non-fatal disease, and better reflects evidence on what society values. The new severity modifier will benefit patients with a wider range of conditions, for example, musculoskeletal, inflammatory, and mental health, in addition to cancer, which the previous end-of-life modifier focused on. It will support access to medicines that will treat very severe diseases.

How might granular real-world evidence support demonstration of severity?
This revision of the NICE evaluation process will make demonstration of burden of disease and evidence on disease natural history much higher priorities for conditions that are potentially eligible for the severity modifier. Real-world evidence (RWE) studies can be a powerful means of capturing this type of data – here are some study design implications for RWE studies developed with this purpose in mind:
-
Recruitment: speed is likely to be important, so recruiting a broad cohort very quickly may be vital, for which community recruitment could be well suited. Site-based/clinician-led recruitment could be run simultaneously and used to produce confirmatory data where required.
-
Inclusion/exclusion criteria: criteria should be broad to enable representation of the widest cohort, for example, age, severity of disease, etc.
-
Definition of subgroups: this would be based on the therapeutic indication and any relevant subgroups, particularly where NICE may consider representativeness in clinical data to be weak (including certain age groups or disease severities).
-
Duration: cross-sectional data capture would be the selected research methodology, although speed of recruitment may be a key driver of how long recruitment remains open.
-
Data capture: EQ-5D would be considered as a default instrument for data capture as it is the preferred method of NICE, along with adult, youth, and proxy versions which align with broad tranches of ages and severities. However, consistency with disease-specificity or other studies may argue for different or additional instruments. Other patient-reported outcome instruments may also enable quantification of aspects of burden for which EQ-5D is less well suited.