Skip to main navigationSkip to main content
The University of Southampton
Health SciencesOur research

Research project: Safer and more efficient vital signs monitoring to identify the deteriorating patient: An observational study towards deriving evidence-based protocols for patient surveillance on the general hospital ward

Currently Active: 

Taking measurements of heart rate, blood pressure, temperature and other vital signs is an important part of care for nearly all patients in hospital. Changes in observations are used to track recovery and can show when someone's condition is getting worse and needs urgent attention. There is now some evidence to support using the NEWS score to identify patients at increased risk of cardiac arrest, death or unexpected transfer to intensive care, so that staff can take preventative action. Although it is recommended that people with higher scores need more frequent observation, there is no direct evidence for this. We need to find out how often we need to monitor patients.

The frequency at which patients should have their vital signs (e.g. blood pressure, pulse, oxygen saturations) measured on general medical and surgical wards is currently unknown. Monitoring protocols in use at present are based on expert opinion, but supported by little empirical evidence.

A key conclusion of national audits of NHS hospitals is that inadequate patient surveillance contributes to avoidable harm through a failure to detect patient deterioration, and risks missing early signs of deterioration and delaying remedial treatment. In contrast, over-observation of stable patients wastes valuable nursing resources that could be better deployed to other essential aspects of patient care.

Using our unique database of vital signs, this study will be the first to guide hospitals on how best to target observations, irrespective of the system or early warning score used to record them. This is a retrospective study using two large databases of vital signs observations from patients on general medical and surgical wards in Portsmouth and Oxford. We will develop continuous time state-space models (e.g. Hidden Markov Models) of the patient population to estimate the risk of future deterioration, given a patient's current acuity.

To ensure our models are robust, we will use a range of early warning scores (such as the National Early Warning Score (NEWS)and the Centile Early Warning Score(CEWS)) as measures of clinically significant changes in physiology.

The results of these models will allow us to predict the expected time of onset of deterioration, given a set of vital sign measurements. Estimates from our models will then be used to design a range of safe, simple monitoring protocols, based on patient acuity, to target observations to patients at risk of deterioration.

In parallel, we will undertake observations of nursing care to estimate the time taken to obtain vital sign observations and (where necessary) initiate an appropriate clinical response, and use these to estimate the expected workload and costs generated by different approaches to patient surveillance. This will form part of an economic analysis of the costs and consequences of our novel monitoring protocols. The ultimate aim of the study is to derive the first evidenced-based protocol for patient surveillance, which will be both safe and achievable across all acute NHS hospitals, independent of the system to record vital signs.

Associated research themes

Health Workforce & Systems

Related research groups

Health Work and Systems
Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings