OBJECTIVES: The use of patient-reported outcomes (PROs) in cost-effectiveness models (CEMs) is not well understood. Our objective is to review the use of PROs as key drivers in the CEMs submitted to the United Kingdom National Institute for Health and Care Excellence (NICE).
METHODS: We searched NICE health technology assessments (HTAs) from January 2016 through August 2022; three reviewers independently screened HTAs. HTAs were included if CEMs submitted by manufacturers contained a PRO as a key driver via the model structure (e.g., health state definition), treatment-effect parameters, treatment-stopping rules, or condition-specific utility measures from key trials. HTAs were excluded if they were terminated or were reviews. Oncology indications routinely rely on non-patient-reported endpoints and were excluded.
RESULTS: Of 428 HTAs reviewed, 26 (6.1%) met the eligibility criteria. Indications included diseases of the musculoskeletal system and connective tissue (n=10, 38%), nervous system (n=6, 23%), respiratory system (n=3, 12%), skin and subcutaneous tissue (n=3, 12%), and others (n=4, 15%). All indications but one (heart failure) have been recognized as PRO-dependent diseases, which mainly use PRO endpoints for regulatory review of treatment benefit. PROs were included in the CEMs submitted to NICE via treatment-effect parameters (n=25, 96%), the model structure (n=24, 92%), treatment-stopping rules (n=21, 81%), and condition-specific utility measures from key trials (n=15, 58%). For 16 HTAs (62%), key PRO drivers in the CEMs were primary trial endpoints.
CONCLUSIONS: Several recent NICE HTAs have used PROs as key drivers in CEMs. Most of these HTAs were for PRO-dependent diseases. When used, PROs were more likely to be incorporated in CEMs via treatment-effect parameters, the model structure, and treatment-stopping rules. Key PRO drivers in the CEMs were primary trial endpoints in most of the HTAs. Including PROs as key CEM drivers may be important in more fully reflecting the patient voice in HTA.
