OPTIMising therapies, disease trajectories, and AI assisted clinical management for patients Living with complex multimorbidity (OPTIMAL study)

Background Complex multimorbidity (cMM, patients with four or more long term diseases) is associated with polypharmacy, treatment burden, and complex clinical decision making. Problem At present multimorbidity research has focussed on clusters of disease with a simple binary classification; not taking into account disease trajectories, the spectrum of disease severity, disease accumulation, or patients and clinicians attitudes. Understanding disease trajectories within cMM clusters, taking into account sub-clinical disease and severity of disease, is important for early identification of populations at high risk of developing further diseases. Treatment strategies in cMM are based on single diseases leading to polypharmacy and complex decision making. Furthermore we do not know the effect of prescribed medications given for one component disease in cMM, on trajectories of other component diseases. We need a deeper understanding of the interaction between diseases and prescribed medications to: 1) optimise treatment and prevent harm; 2) identify most appropriate medication when there are more than one option; and 3) discover pleotropic medicines that could be repurposed for multimorbidity prevention. Methodology We will aim to solve these problems by integrating multimodal primary care, secondary care, qualitative, and prescribing data to develop an artificial intelligence tool. The tool will optimise clinical decision making in patients with cMM and polypharmacy, by using a patient similarity approach. Firstly, we will develop a risk prediction algorithm to identify the likely next disease and the next best treatment option in patients with cMM. We will identify disease clusters and trajectories using biomarkers and physiological measurements of disease severity, and then add in prescription data to model how they interact with trajectories of diseases. To do so we will adopt a well-founded AI modelling approach that has recently become popularised for the characterisation of observational electronic medical record data. Secondly we will develop a prototype of an automatic Advanced Patient Similarity Tool. This will be developed building on our expertise of developing the DExtER platform, an automated platform for epidemiological studies, one of its kind in the UK. Throughout the project we will embed the voice of patients and health professionals, by exploring their attitudes to AI directed decision making. Patients and clinicians will also input into the initial algorithm to validate and co-design the software tool. Team We are a consortium with expertise in multimorbidity, ageing, artificial intelligence, health data science and qualitative methodologies. We will work in partnership with the RSF using our distinctive features to become a national exemplar of reproducible, secure and interoperable research in practice and demonstrating the utility of accessible research-ready data made available through rigorous engineering and technical standards. Significance This project will answer significant knowledge gaps in our understanding of multimorbidity, by using cutting edge AI techniques and data science. The developed software would provide tangible and immediate guidance to clinicians managing people with CMM. The tool will inform clinical guideline development, as well as providing the basis for future clinical trials of personalised interventions in cMM.