Better information to reduce treatment variation
Within the NHS it can take many years for the latest research evidence to have an impact on clinical practice. As a result, the forms and standards of care that patients receive often do not reflect evidence-based best-practice. One key problem is that doctors and nurses gather large amounts of information and have to make lots of decisions based on it with very limited thinking time. So much data is held in these systems, that it rarely becomes available to clinicians in a form that could help them to improve their performance.
Solving this problem involves understanding why clinicians make the decisions they do and changing their behaviour. To achieve this, Dr Bourdeaux and colleagues at University Hospitals Bristol, have been testing so-called ‘nudge’ interventions in the hospital’s intensive care unit (ICU). Such interventions alter the environment to reduce and simplify thinking needed to reach a decision. One such intervention uses a large-screen display to order information in a way that alerts clinicians to physiological data collected as part of patients’ electronic health record (including minute to minute physiological data and details of care processes such as ventilation). This system has resulted in a significant and sustained improvement in the use of evidence-based ventilation practice and reduced variation in patients’ tidal volume (the air shifted between inhalation and exhalation). As a result, patients now spend less time in the ICU, yielding both medical and financial benefits.
Research for Health Award
Until now, analysis of this intervention had been relatively basic, and the influence of environmental factors on the strength of the observed behavioural change remained uncertain. Professor Gilchrist and colleagues drew on a Research for Health Award and a wealth of electronic health record data to carry out a more sophisticated investigation. Their focus was on the display system’s key variables and how they interact (e.g. time of day, time series, patient profile, display layout, business of the units, how long it takes for displayed warnings to be heeded).
They found that doctors responded to warnings about excessive tidal volumes and such warnings affected behaviour in the long term, even when the levels were below accepted warning levels. In addition, they showed the effect on tidal volume of the default setting on the ventilator. The results suggest that by subtly shaping the context in which decisions are made, health professionals can be nudged towards making better decisions.
The demand for intensive care is growing and in February 2013 the NHS already had 3,770 adult intensive care beds, each with an average cost of approximately £1,500 per day. Projects like this one, which help to improve the reliability of ICU interventions will reduce length of stay, reduce cost, improve outcomes for patients and could be rolled out to the rest of the health care system. The project has already helped to inform several journal articles and applications for additional funding in this area.