OBJECTIVES: We evaluated the role of patient-reported outcome (PRO) data and patient testimonial evidence (e.g. patient advocacy) in reimbursement decisions in 10 countries.

METHODS: Websites of national/regional health technology assessment (HTA) agencies for the UK (England, Wales, Scotland), Spain, France, Germany, Italy, Canada, Japan, and Brazil were searched for decision documents that included PRO data and/or evidence of patient involvement in a pre-specified list of treatments for three diseases (diabetes, obesity, and haemophilia). HTA guidance for formal patient involvement processes was also identified and national clinical guidelines were reviewed for information regarding patient involvement. Data were compared to identify common themes and inter-country differences.

RESULTS: In total, 155 HTA decisions were reviewed for PRO data and patient involvement. Only 28% (44) reported use of PRO data in decision making. PRO data were referenced more frequently in HTA decisions from countries that use a cost-utility and/or cost-effectiveness approach to economic evaluation (UK, Canada, and Italy), for which detailed guidance on PRO data expectations exists. Reasons for PRO data not being included were: poor trial design (e.g. no head-to-head data or data from open-label studies), limited generalisability to the population, and lack of clinically or statistically significant differences between interventions. Processes for patient involvement were identified with substantial inter-country variability, from formal patient submissions/consultations and committee involvement (UK, Canada) to limited/unclear patient involvement (France, Japan). Clinical guidelines were largely clinically-led, although patients were formally involved in guideline development in the UK, Spain, and Italy.

CONCLUSIONS: Patient involvement had limited influence on HTA decision making in the diseases evaluated, suggesting that formal processes for patient involvement are not fully maximised. PRO data influenced few decisions, potentially due to being insufficiently robust or compelling. Valid, reliable, responsive PROs in well-designed comparative clinical studies are needed to generate data for reimbursement decision making.