PMU108 ASSESSING THE IMPACT OF PATIENT-REPORTED OUTCOMES IN HEALTH-TECHNOLOGY ASSESSMENT

Abstract

There is growing emphasis on incorporating the patient perspective in assessing the value of new health technologies. The extent to which health technology assessment (HTA) bodies are using patient-reported outcome (PRO) data in their assessments is evolving and poorly understood. This analysis aimed to evaluate the role of PROs in decisions made by HTA bodies. We reviewed HTA appraisals conducted between January 2015 and June 2019 by CADTH, pCODR, G-BA, HAS, NICE, and PBAC for 3 diseases representing oncology (multiple myeloma), neurology (Parkinson’s disease), and immune-inflammatory (rheumatoid arthritis) therapies. Relevant HTA appraisals were identified using the Context Matters Market Access Platform (MAP). HTA appraisals for which complete data were not yet available were excluded. PROs were categorized as accepted/rejected/unknown, and reasons for rejection were documented. Overall, 96% of HTA appraisals (238/249) were available in the MAP. 29% of indications (69/238) reviewed by HTA bodies involved the use of one or more PROs (36% for multiple myeloma, 40% for Parkinson’s disease, and 24% for rheumatoid arthritis). A total of 165 PROs were included, and overall, 64% of PROs (105/165) were accepted. PRO acceptance rates were higher for rheumatoid arthritis (68%) and Parkinson’s disease (67%) than for multiple myeloma indications (56%). Acceptance rates also varied across HTA bodies. Common reasons for rejection of PROs were related to the statistical analysis methods used (e.g., hierarchical analysis of endpoints), open-label study design, the lack of specification (or validity) of minimally important difference thresholds/responder definitions, and low completion rates. Approximately two-thirds of PROs included in recent HTA appraisals were accepted. The use and acceptance of PROs in HTA appraisals varied across diseases and HTA bodies. Using validated PRO measures and validated minimally important difference thresholds in well-designed clinical studies are important for the PRO data to be considered in HTA decisions.