NHS Lothian

NHS Lothian Charity Health Innovation South East Scotland

DL_2023_010

Too many young people in the UK die from asthma. Almost all of these deaths can be prevented if young people and their families know about their illness and are enabled to take the treatment that prevents attacks. Currently, children and young people are reviewed routinely either in primary or secondary care, but this may not be at a time when they are unwell. The clinic system currently doesn’t allow the flexibility to see patients when needed. We wish to use clinical data to pick out those who need to be seen, rather than waiting until they are very unwell. It is only by accessing data from many different sources that we can get a whole picture of any individual patient and then tailor their care to what they need. Accessing Dataloch provides the opportunity to use high quality data that already exists to design and develop a tool that can eventually be used by the NHS that can be used to improve care for individual patients.

Asthma deaths amongst young people in the UK are too high. In the UK, a child is admitted with an asthma attack every 20 minutes and in Scotland there are 1-2 deaths of young people with asthma each year. Currently, we don't know which individuals are at high risk of an attack until after the event. Asthma management seeks to control symptoms and manage attacks with treatment and education about risk reduction to avoid attacks. It is a common condition, affecting around 10% of children in the UK. Without knowing which patients are suffering increased symptoms at any one time, we rely on timed routine reviews in primary care or hospital clinics. These appointments may or may not be at a time of poor health, but there is no way of seeing patients who are at risk when they need to be seen. Academic literature has identified factors that increase individual risk, including previous attacks, reduced preventor inhaler use (poor adherence) and increased reliever inhaler use (symptoms out of control). These factors can be used together with other routinely collected EHR data to train machine learning based risk classification models. Healthcare data, routinely collected, together with a risk stratification algorithm, could identify individuals who are at risk and present these risk scores to clinical teams to support a proactive approach to managing these high risk patients with a view to preventing emergency presentations at ED.  In addition to reducing emergency care utilisation, clinical decisions support tools would improve care for young people with asthma through reduction in routine clinic appointments, reduced unnecessary travel to hospital, improved symptom control by optimising treatment adherence.

Public Health Research

30/05/2024

De-Personalised

TRE