Background: Cardiovascular disease is a leading cause of death globally and medications to prevent cardiovascular outcomes are prescribed based on evidence from randomised controlled trials. However, generalisability of trial results to at-risk groups, who are often underrepresented in trials, is unknown. Methods: This thesis used trial replication methods applied to the ONTARGET trial to validate findings from electronic health record data before extending inferences to trial underrepresented and excluded groups. Results: Using a cohort of 137,155 patients in a propensity-score—weighted analysis conducted in the UK Clinical Practice Research Datalink (CPRD) GOLD I obtained comparable treatment effects to the ONTARGET trial for ARB compared to ACEi. After benchmarking findings to the ONTARGET trial results using a pre-specified validation criteria and aided with an increased sample size with more diverse characteristics, I extended findings to females, those aged 75 years and patients with chronic kidney disease and obtained consistent results. Consistent results were observed for Black and South Asian ethnic groups using CPRD Aurum. I observed a small increase in angioedema reported among Black individuals compared to White individuals. However, I observed ARBs were associated with a decreased risk of developing angioedema over a maximum follow-up of 5.5 years compared to ACEi in both Black and White ethnic groups despite clinical guidance recommending an ARB in preference of an ACEi in Black patients only. The replicability of the dual therapy comparison using an operational definition to capture dual users was explored and also led to comparable results. Conclusion: When studying the use of ARBs and ACEi in high-risk individuals for the prevention of cardiovascular outcomes, applying trial replication methods to electronic health record data can add confidence to findings and provide evidence on treatment effects and risk in key at-risk groups who are often underrepresented in trials.