The August International Psychogeriatrics Article of the Month is entitled “Development of a smartphone application for the objective detection of attentional deficits in delirium” by Zoë Tieges, Antaine Stíobhairt, Katie Scott, Klaudia Suchorab, Alexander Weir, Stuart Parks, Susan Shenkin and Alasdair MacLullich.

Delirium is a sudden disruption in consciousness and cognition marked by an inability to focus and sustain attention. People with delirium can also experience psychotic symptoms such as terrifying hallucinations and delusions. Delirium is extremely common, affecting around one in eight hospital patients. It can occur at any age but is particularly common in older patients in hospital, where it is often mistaken for dementia. Delirium tends to develop after infection or surgery, or as a result of drug side-effects, and has serious consequences: it affects how long patients stay in hospital and future mental and physical well-being, and increases the risk for future dementia and death.

The first step towards improving treatment and subsequent health outcomes of patients with delirium is accurate detection of the syndrome. Yet up to three-quarters of delirium is undetected. Part of the problem is that delirium is difficult to recognise, with some features similar to those in other conditions such as dementia and depression. Most available delirium tests discriminate inadequately between delirium and dementia, rely on subjective judgements, and require trained personnel. It is evident that new ways of easily and accurately detecting delirium are needed.

Researchers at the Edinburgh Delirium Research Group and LLHW Centre for Cognitive Ageing and Epidemiology (CCACE), at the University of Edinburgh, have previously developed an electronic method for measuring delirium using simple visual attention tests, which were implemented on a purpose-built device known as ‘Delbox’. The rationale for this method is that patients with dementia will perform better because they are able to focus their attention for longer periods than patients with delirium. This is exactly what the researchers found.

The team have taken this approach one step further by developing a software application (‘DelApp’) for smartphones based on the Delbox, in collaboration with the Medical Devices Unit in Glasgow (NHS Greater Glasgow & Clyde). The DelApp incorporates a brief arousal assessment, a visual test and a graded counting task to measure inattention. The patient is asked to count a short sequence of circles shown on the smartphone screen. The DelApp takes a few minutes.

The researchers studied the feasibility and validity of the DelApp. First, 20 older hospital patients were tested with both Delbox and Delapp, showing comparable performance on both tests. Patient feedback was overall positive and suggested that the DelApp was acceptable to them. Secondly, the DelApp was pilot tested in 156 selected older patients split in three groups (delirium, dementia, and no delirium/no dementia). The delirium group performed more poorly compared to dementia and no delirium/no dementia groups. The DelApp not only showed good diagnostic performance, but also appeared to capture the severity of delirium as measured with a validated delirium severity scale.

In conclusion, this study showed the initial feasibility and validity of using a smartphone test for delirium assessment in hospital. The DelApp test appears to be objective and easy to use at the patient’s bedside. It requires little training, which may be particularly useful for non-experts.

The research team is developing the DelApp further and will confirm the clinical effectiveness of the test in larger trials with blinded raters.


The full paper “Development of a smartphone application for the objective detection of attentional deficits in delirium” is available free of charge for a limited time here.

The commentary paper “Bringing delirium into the 21st century: will physicians get the app out?” by Kenneth Rockwood is also available free of charge here.

Leave a reply

Your email address will not be published. Required fields are marked *