Presentation (CTI)

OBJECTIVES: While multiple guidelines are available clarifying the analysis of survival (time-to-event) outcomes for cost-effectiveness models (CEMs), there is limited guidance on the use of disease activity outcomes (DAOs) in CEMs. We aim to explore the issues related to the analysis and use of DAOs and provide recommendations for a future framework.
METHODS: Critiques of CEMs based on DAOs from three HTA agencies (NICE, CDA, SMC) were assessed, using rheumatoid arthritis and systemic lupus erythematosus as examples. Identified challenges of analysing and using DAOs in CEMs served as the foundation of a framework and considerations to guide future guidance.
RESULTS: The challenges of utilizing DAOs are commonly noted in HTA reviews, including the multitude of available instruments with short-term assessments, the relapsing-remitting disease nature where changes in DAOs drive quality-of-life/costs, limited patient follow-up within these chronic conditions, variability of patient characteristics across geographies driving uncertainty in clinical generalizability. Predictive equations are often linear, focused on long-term trends (with annualized mean scores), covariate selection driven by limited data availability. Additionally, published equations are often developed to measure drivers of disease activity to inform acute clinical decisions, rather than informing extrapolation, presenting challenges for inclusion in CEMs.
CONCLUSIONS: A future framework should consider the aim to inform health resource allocation decisions, focusing on formal selection of prognostic factors, consistent selection of DAOs, and explicit assessment of the attributes of selected DAOs. Future CEMs should also seek to assess different functional forms, clinical/biological plausibility of extrapolations and incorporation of uncertainties, for the specific decision problem. The challenges and criticisms of analysing and implementing DAOs emphasizes the importance of developing detailed guidelines like those for survival analysis. A framework would increase transparency, contribute to more robust decisions informed by CEMs while transparently illustrating their limitations to assess their clinical plausibility.