Modelling Resilient Hospital Systems

Especially with an imminent election, few issues rank higher with politicians and journalists than the problems with the NHS. A Health and Social Care Committee report in July 2022 declared “The National Health Service and the social care sector are facing the greatest workforce crisis in their history” with the situation having considerably worsened since then. Providing affordable healthcare which is free at the point of delivery is one of the greatest social challenges and fundamental to the postwar identify of the UK as a welfare state. This challenge needs to be addressed at a time of pressure on public finances, looming recession and healthcare sector labour shortages, so the NHS must become more resilient and robust with the current resources it has or something very similar. 

It is commonly claimed that the NHS needs to be run more efficiently but there is much less consensus about what this might amount to in practice with possible solutions frequently being linked to political preferences. Currently the NHS is trapped in a situation of trying to optimise individual processes to save resources but these optimisation attempts may be harmful to the overall healthcare system. A common example of this is the problem of ambulances delivering patients to Accident and Emergency who cannot be admitted into wards because beds are blocked by people who are actually fit to go home but cannot be discharged because no appropriate home care has been organised. Could there be ways of thinking about healthcare which are less dependent on political and ideological preferences and make good use of knowledge about optimising resources in complex systems involving people and technology?

The starting point for our research was to understand why, despite historical and current levels of funding, it seems so difficult to make significant improvements in patient care and cut waiting lists. A key aspect of this is bottlenecks in the system where the efficient care of patients is slowed by a blockage in the system at some point (such as a lack of available operating theatres) and gaps in staffing which have to be covered by expensive bank staff and locums. We decided to take one crucial part of the NHS system - hospital care - and see if we could represent it in ways which were illuminating. 

Instead of a traditional individual skill set approach, which requires an individual to hold fixed sets of skills in order to carry out particular roles, the modelling approach took the perspective of a collective approach to the knowledge and skills individuals hold and technology required to utilise these. The vision of the research is to use system engineering modelling techniques developed for complex engineering systems (such as aircraft or engine design) and apply them to system modelling at the hospital level to maximise the impact of shifting resources. 

A key element of this is employing change prediction models which analyse how changes to one part of a system affects other parts (such as patient discharge procedures in relation to patients being brought in by ambulances) and design margin modelling, which analyses where buffers can be deployed to break propagation through chains of events with margins being the level of over capacity in the system at various points. This will enable hierarchical approaches to modelling margins to analyse where changes can be absorbed and where effective investment will free up the greatest capacity with the innovative use of a branch of mathematics known as category theory. As well as better insight into these issues the research aims to develop a prototype visualisation tool kit for use with NHS managers. We are working with a variety of collaborators including universities, research institutes and NHS trusts with a valuable visit to the National Institute of Standards and Technology in Washington last year.

Category Theory diagram

Image: The modelling team at NIST.

About the authors

Mark Addis is Associate Dean Knowledge Exchange at the Open University and a Research Associate at the Centre for Philosophy of Natural and Social Science at the London School of Economics. His research interests include epistemology. 

Claudia Eckert is Professor of Design at the Open University. Her research focuses on empirical studies of development processes of complex products and the development of tools, methods, and concept to support them.