Transforming clinical evidence: causal inference in electronic healthcare records to improve inclusivity in management of MI with multimorbidity

Randomised controlled trials (RCTs) are the gold standard for clinical guidelines but often exclude people with multimorbidity. The resulting paucity of evidence of the comparative effectiveness of treatment for specific clusters of multimorbidity is associated with significantly worse outcomes and a major contributor to increasing health inequality. A fundamental shift in generation of evidence which accounts for the full complexities of human health and disease is needed. This fellowship aims to generate new treatment evidence capturing the full complexity of multimorbidity phenotypes using healthcare data spanning the UK population. Whilst target trial emulations are capable of mimicking RCTs in observational data, their successful implementation relies on advanced causal inference which remains under- utilised especially at full population scale. Therefore, I will develop a rigorous methodological pipeline capable of producing clinical guideline quality evidence at scale. In the UK, there is a hospital admission for myocardial infarction (MI) once every 5 minutes. At least 60% of these individuals have complex disease co-occurrence. Through implementation of an extensive series of target trial emulations – this fellowship aims to produce treatment evidence for MI extended to the full complexity of multimorbidity for more inclusive management meeting the health and care needs of the population.