Approaches for Creating Clinical Evidence of treatment effects in routine Populations excluded from Trials (ACCEPT) (360G-Wellcome-214588_Z_18_Z)

Clinical trials are the gold standard for estimating treatments effects but exclude very large proportions of the clinical population to which the evidence is applied. Extrapolation of trial evidence assumes trial-eligible and ineligible populations are similar in terms of: (1) their baseline risk of health outcomes; (2) the relative and absolute effect of treatments; and (3) the incidence of competing risk events. For several clinical trials I will use observational data to examine whether and how: \- Baseline risk of health outcomes varies among trial-eligible and ineligible populations \- Relative and absolute treatment effects vary among trial-eligible and ineligible populations \- Failure to account for competing risk events leads to biased estimates of treatment effect. I will define trial-eligible and ineligible real-world populations to measure baseline risk of health outcomes. I will use propensity score matching to build cohorts of treated and untreated individuals to examine heterogeneity in treatment effects among trial-eligible and ineligible populations. I will examine the impact of accounting for competing risk among trial-eligible and ineligible populations by fitting Fine and Gray models. My overarching aim is therefore to generate evidence to inform a balanced accounting of benefit and harm for decision-making relevant across the entire clinical population.