OBJECTIVES
Recognition of delirium in the emergency department (ED) is poor. Our objectives were to assess: (1) the diagnostic accuracy of the Predicting Emergency department Delirium with an Interactive Computer Tablet (PrEDICT) “serious game” to identify older ED patients with delirium compared to clinical recognition and (2) the feasibility of the PrEDICT application compared to existing tests of attention.
DESIGN
Prospective observational study.
SETTING
ED of a Canadian tertiary care center.
PARTICIPANTS
We included ED patients, aged 70 years and older, with a minimum 4‐hour stay. We excluded anyone with critical illness, communication barriers, and visual impairment or those unable to use a computer tablet. None had prevalent delirium by ED clinicians' routine clinical assessment.
MEASUREMENTS
Participants were asked to tap targets on a tablet at four difficulty levels. Time and accuracy were automatically recorded. Other measures included the Confusion Assessment Method, the Delirium Severity Index, the Digit Vigilance Test (DVT), and the Choice Reaction Test (CRT).
RESULTS
We enrolled 203 patients. Their average age was 80.6 years, 49.8% were female, and their average ED length of stay was 15.9 hours. Sixteen subjects had clinically unrecognized delirium, and 14 of them completed the PrEDICT game (87.5%). We developed a threshold score with 100% sensitivity (95% confidence interval [CI] = 76.8%‐100.0%) and 59.7% specificity (95% CI = 52.3%‐66.6%) to identify patients with clinically unrecognized delirium. The area under the curve was 0.86 (95% CI = 0.77‐0.94). Completion rates were 196/203 (96.6%) for the PrEDICT serious game compared to 128/203 (63.1%) for the CRT and 51/203 (25.1%) for the DVT.
CONCLUSION
Older ED patients were able to use our serious game, including 87.5% of those with clinically unrecognized delirium. The PrEDICT application has potential to act as a sensitive screening tool to identify older ED patients with clinically unrecognized delirium. J Am Geriatr Soc 67:2370–2375, 2019