Picture an intensive care unit in a major hospital. An oasis of calm, where seriously ill patients are nursed attentively back to health? Well, actually, no. Most ICUs are a whirlwind of noisy activity, with people coming and going, conversations carrying on and alarms sounding as machines signal their need for attention or a patient breaches safety thresholds. Add to that multiple computers with noisy fans plus a background of squeaky trolley wheels, clanging bins and banging doors. OCC has recently been working on the SILENCE Programme with Julie Darbyshire, Critical Care Research Programme Manager at the John Radcliffe Hospital in Oxford, to investigate whether alerting staff to high or rising noise levels can lead to their reduction, benefitting patients and staff alike.
“Noise levels in the ICU are much higher than they should be. The World Health Organisation recommends a maximum level of around 35 dBA, whereas we know our operating level is more like 52–59 dBA – and that’s pretty typical of any ICU,” explains Julie. “Given that the decibel is a logarithmic scale, we have a very much louder noise level than we should and it’s certainly not a relaxing environment for recuperation, especially if you’re there 24 hours a day.”
Research has established that patients generally experience very fragmented sleep patterns in the ICU, with sometimes only a few minutes of sleep at a time. Around 25% of patients experience delirium, becoming very confused about where they are and frightened to find themselves attached to tubes and wires and unable to move about. Sometimes they can become agitated or violent. “It’s quite possible that poor sleep is linked to delirium and that reducing noise levels would enhance sleep – that’s something we’re keen to research,” says Julie.
To that end, Julie was already collecting noise level data in one of the ICU’s 6-bed bays, using an array of 72 microphones set up by Southampton University’s Institute of Sound and Vibration Research to pinpoint on a floor plan where sound was coming from and to monitor the levels at each individual bed. She enlisted OCC’s help to turn that information into a display that would heighten staff awareness of noise, with a view to encouraging them to take action to dampen things down when possible. It was also important to record how noise levels fluctuated throughout the day in a particular location and thus identify any ‘hot spots’.
Understanding how the prospective user would interact with a ‘noise awareness’ system was an important part of the design process. OCC UX designer Andrew Wilson undertook a period of desk research into the role and responsibilities of an ICU nurse, who is typically continuously looking after a single patient. He then talked at length to nurses and doctors about what it’s really like working in the ICU, what are the necessarily noisy parts of the day (generally shift handovers and new patient admissions), what triggers various noises, where their attention is focused, and how they would want to be alerted to rising noise levels. Constructing the user persona of ‘Suzanne’, a typical nurse, allowed a user journey to be workshopped and thoroughly examined with clinical staff.
“It became clear fairly early on that we were talking about how to achieve behaviour change,” says Julie. “What simple images should we be showing busy nurses and doctors to quickly convey information about rising noise levels so that they respond appropriately?”
A range of visual prototypes was tested with users to investigate their interpretation of, for example, circles of different sizes and colours. Does this bigger/darker circle mean that a noise was louder or that it lasted longer? The clear message was that users wanted to see immediately how the sound levels being shown compared with ‘acceptable’ levels.
The pilot system currently being assessed in the ICU comprises a screen which flips between two displays, each using a traffic-light system to signify how a sound level compares with acceptable loudness. One shows a floor plan of the bay, with sound hotspots displayed. The other shows sound levels in three circles – one for the sound level now, one averaged over the past hour, and one averaged from the start of the shift. Clicking on these brings up a detailed graph of sound variation with time.
“The pilot is attracting considerable interest amongst ICU staff – we can clearly see peaks where new patients were admitted and regular increases in sound during handovers, and the shift information seems to be bringing out a competitive spirit among the nurses,” says Julie. “We needed to approach the problem of noise in the ICU in a new and creative way to make people think about noise differently. It was enormously beneficial that OCC was comfortable working in the hospital environment and helped us realise a design that was guided by the needs of the clinical staff.”