Seasonal winter peaks in infectious diseases put significant pressures on national health services, including hospitals and general practitioners. Respiratory Syncytial Virus (RSV) is a common virus which for the majority of people causes mild symptoms like a common cold, but it can be serious for very young infants and older adults. Currently, RSV is a significant contributor to pressures in paediatric intensive care.

Having an early warning of when and how many people will need care for infectious diseases such as RSV, can help public health practitioners and policy makers make sure NHS services can cope with the extra demand for care. Equally important to understanding when pressures start is knowing when they will end, so efforts can be focussed on planning for services getting back to normal when extra services are no longer needed.

Getting the necessary information about infectious diseases involves the use of near-real-time feeds of ‘big data’ collected from the NHS (such as anonymised information about NHS 111 calls and attendances at emergency departments across the whole of England). This information helps us understand how changes in the health care seeking behaviour of patients relates to changes in virus activity while it’s happening. For example, increases in patients attending emergency departments with influenza-like illness symptoms is a sensitive indicator that influenza activity is increasing in the community.

This project will use ‘machine learning’ techniques (a branch of computer science which uses sets of rules (algorithms) to ‘learn’ from data) to predict when the peaks in RSV-related healthcare demand will happen, how big those peaks will be and how quickly they will then reduce. This knowledge will enable effective and targeted planning, and public health messaging about the usage of different healthcare options.

This project is led by Dr Dan Todkill of Warwick Medical School, University of Warwick, coordinated and supported by Health Data Research UK with funding from the National Institute for Health and Care Research (NIHR). who said ‘We are excited to be looking in more detail at how we can use machine learning techniques to improve healthcare and patient outcomes. These techniques will give us more accurate information and advanced knowledge of the impact of RSV related healthcare issues so we can identify any trends and ultimately allocate the right healthcare resources in the right places at the right time. This means resource and funding can be directed to focus on the most urgent healthcare needs at this time of year’.

 

ENDS

Dr Dan Todkill

Warwick Medical School