Advanced mathematics for human benefit
1 November 2011
Operational research uses the techniques of mathematics to maximize the productive use of resources in large systems or organizations. At UCL’s Clinical Operational Research Unit, the focus is on the varied challenges that confront the delivery of healthcare. CORU’s director Professor Martin Utley came to the field from high-energy physics in order, he says, ‘to get a feeling of more frequently solving problems’.
Although its methods seem abstract, clinical operational research can deliver real human benefit. Early CORU work showed that historical data on leukemia patients could be used to model outcomes for the standard treatment, allowing all participants in several trials to receive experimental treatments, speeding up the treatment evaluation. Recently, Utley has been working with Great Ormond Street Hospital to plan critical care in the event of a pandemic. Previous studies had recommended introducing triage, in which severely affected patients are denied treatment in favour of others more likely to recover.
According to Dr Paula Lister, consultant paediatric intensivist at GOSH, Utley’s involvement ‘clarified thinking about triage to a huge degree’. His purely mathematical modelling of likely scenarios allowed the objectives and threshold for introducing triage to be defined more clearly so that, if used, triage would be more likely to prove of overall benefit.
Utley says that his models are often ‘to prompt debate rather than deliver “answers”. The key consideration is to provide teams with the tools to monitor their own outcomes. They take responsibility for it. We step back from the interpretation precisely because we know we’re looking at a partial view of the story.’
For all the mathematical sophistication, it’s important to remember that it is human beings who are at the heart such studies. For this reason, consultation with clinicians is an essential preamble for Utley, and likewise his results are presented in such a way that clinicians can use them in their planning. ‘If I didn’t talk to clinicians and get their clinical nous I wouldn’t be able to do my job. The value of what we do extends entirely from this collaboration,’ says Utley. The triage mathematics was shared – ‘not as Greek letters, but graphically to convey the insights’ – with consultants at the hospital and nationally as well as with other interested parties.
CORU has licensed some modelling software, but it remains free to clinical bodies. Thanks to UCL Enterprise, his department also undertakes an increasing amount of consultancy work. ‘There’s a vast untapped potential for this sort of analysis within healthcare,’ says Utley. Other industries make extensive use of operational research, for example in transport logistics, but in healthcare there remains some resistance because it seems to lack the human touch. ‘It’s helpful to clinicians if a person such as me can take a cold mathematical view.’
IMAGE: UCL's Clinical Operational Research Unit - selection bias