Swansea University

Academic Institute

911V2

This project will create new analytical methods, such as machine learning which uses statistical models that computer systems use to more effectively perform a specific task than a person would doing it manually and relies on patterns and learning instead. Machine learning algorithms build a mathematical model of sample data, known as 'training data', in order to make predictions or decisions without being explicitly programmed to perform the task, which will help to identify people suffering from a wide range of multiple health conditions that we may not have seen or known about before. It will then investigate the short and long-term health & wellbeing outcomes of mixtures of conditions, for example, in people who suffer from high blood pressure and also have asthma. We will study how having multiple conditions effects individuals’ quality of life, how long they live and patterns of treatment from general practice and hospitals. We will then investigate whether there are patterns of treatments that appear to have better or worse outcomes for people with different combinations of illnesses. Such people are hardly ever included in trials to assess the safety and effectiveness of drugs. Information derived from this work will help inform clinicians and the public on which treatments work best for which combinations. We may also identify new, unknown, combinations of diseases that suggest a common origin and would help scientists to target their efforts into why these conditions occur and speed up the development of new treatments. By looking back over the whole patient history and linking all laboratory results to the dates that symptoms and diagnoses are first made we may be able to identify conditions at an earlier time frame, when treatment can be more effective. The project will use data within the SAIL Databank to investigate the changes in the number of individuals with multiple health conditions over time, changes in the number of individuals with multiple health conditions in different age groups and genders, numbers and distributions of common health conditions in Wales, categorised by key criteria such as demographics (5 year age-groups and gender), geographical location at suitable levels (such as local authorities or local health board or higher). We are working with world-leading experts from the Alan Turing Institute to develop new analytical insights into disease clustering that will be of value to many other studies that use electronic health records. The development of this project was helped by members of the public from the Consumer Panel for Data Linkage. Members of the public will be involved in all aspects of the study.

The development of this project was helped by members of the public from the Consumer Panel for Data Linkage. Undoubtedly, that contributed to a very high ranking (9/10) from the Medical Research Council panel that assessed these bids and recommended funding (27th February 2019). We will recruit two members of the public who will be involved in all aspects of the study going forward and play leading roles in public engagement and dissemination of results. They will work closely with clinicians and colleagues from NHS Wales, NHS Wales Informatics Service (NWIS), who also contributed to the initial design of the study. We expect that results will feed back via multiple key stakeholder groups e.g. SAIL, Health Data Research UK, Medical Research Council, Academy of Medical Sciences and the Wellcome Trust ‘Understanding Patient Data’ initiatives and local and national events that aim to improve the public understanding of science.

28/03/2022

Anonymous

(e) processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller;

(j) processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

Not applicable

Not applicable

TRE