PMU46 Impact of Assumptions Used in Cost-Effectiveness Analysis (CEA) on the Outcomes of CEA Across Disease Areas

CEA frameworks have been criticized for failing to accurately assess the value of innovative treatments in populations with high unmet need. We examined the impact of conventional CEA assumptions on treatments for diseases with specific attributes. We identified four key CEA assumptions likely to impact cost-effectiveness of treatments across disease areas: fixed discount rates for costs and benefits, inclusion of routine disease management costs of survival, preference for generic utility instruments, and fixed drug pricing following patent expiry. We developed three cohort Markov models to examine the hypothetical impact of these assumptions in disease archetypes characterized by selected attributes: life expectancy (age of onset, mortality), treatment impact on life expectancy, treatment impact on morbidity, and expected treatment duration. Disease archetypes combining different attributes were modeled and a targeted review of submissions to National Institute for Health and Care Excellence (NICE) 2015-2020 conducted to compare hypothetical results to real-world submissions for conditions representative of modeled disease archetypes, including Crohn’s disease, cystic fibrosis, and acute cancer. Models showed conventional assumptions mandating fixed discounting, no generic pricing, and inclusion of disease management costs resulted in incremental cost-effectiveness ratios (ICERs) 25%-75% higher than if assumptions were relaxed in conditions with long time horizons (long life expectancy, high treatment impact on mortality, long treatment duration), but made little difference to ICERs in diseases with short horizons. Use of generic utility instruments increased the ICER for long-term conditions with high treatment benefit on mortality and morbidity, but had little impact on short-term conditions. Review of NICE submissions confirmed and expanded on these findings. Conventional assumptions in CEA frameworks have a large adverse impact on ICERs in diseases with early onset, lifelong treatment, and high treatment benefits on life expectancy and morbidity compared to diseases without these attributes. Funding: Vertex; Support: Complete HealthVizion.