Clinical risk prediction models in COPD exacerbations

Research Question: Can COPD exacerbations outcomes be improved by clinical prediction tools? Background: The annual cost to the NHS budget of COPD is £2 billion, of which £1 billion is spent on exacerbations, defined as periods of symptom worsening. Exacerbation patient outcomes have remained bleak and unchanged for decades. The reasons for this are several-fold; i) exacerbations are heterogeneous in nature with different inflammatory, physiological, and symptomatic profiles; ii) there is overlap with comorbidities such as an increased risk of a cardiovascular events within 30 days of an exacerbation; iii) treatment with systemic steroids and antibiotics for exacerbations of COPD have not changed for decades, and in 40% of patients, the treatments are not effective and can lead to harm; and iv) we apply little precision medicine approaches to treatment. Evaluation of this heterogeneity and risk is currently absent in clinical practice in the management of an exacerbation of COPD. Aims and objectives: The COPD exacerbation James Lind Alliance priority setting partnership has set ten research questions that are important to answer. This proposal is focussed on answering three of the top 5 ranked questions: i) to best inform the cause of an exacerbation; ii) best identify treatment for an exacerbation; and iii) understand better the effect of cardiovascular comorbidity at the time of an exacerbation. To do this, the aims of this proposal are: Develop a COPD exacerbation endotype tool (COPD-PEARL). Develop a COPD exacerbation CVD risk tool (COPD-CARDIAC). Apply for further funding and progress findings into clinical practice. Methods: Using a wide range of clinical, physiological, and biological data from already collected clinical trials, the coPd ExAceRbation tooL (COPD-PEARL) and Copd heARt Disease rIsk cAlCulator (COPD-CARDIAC) will be created. Models will be developed from multiple data points from three studies with over 2000 patients, with sampling at the time of an exacerbation. For each model, data will be split into training, validation, and replication cohorts. COPD-PEARL will a) objectively define exacerbation endotypes; b) identify optimal treatment for exacerbations; and c) describe COPD exacerbation resolution. The COPD-CARDIAC tool will determine the risk of major adverse cardiovascular events, such as myocardial infarction, cerebrovascular accident, or death in the early (within 5 days), medium (within 30 days) and late (within 90 days) risk period after an exacerbation of COPD. Anticipated dissemination and impact: The goal of this award is to improve outcomes for patients with a COPD exacerbation by changing clinical practice. Firstly, it will lead to developement of COPD PEARL as a way to detect the cause of a COPD exacerbation and can inform the event trajectory. Secondly, it will develop CODP-CARDIAC, a COPD exacerbation cardiovascular predictor risk tool. Finally, it will lead to further funding applications to test the utility of COPD-PEARL in a feasibility trial and further validate COPD-CARDIAC from real-world primary care data. These prediction tools will change clinical pathways globally.