Predict - Predicting Emergency Department Incident Delirium with an Interactive Computer Tablet
Abstract
Background: Delirium is common and potentially lethal. Unfortunately, delirium recognition is poor- 75% of Emergency Department (ED) cases are missed. Objectives: We previously developed a “serious game” using participants’ game performance to evaluate delirium risk. Our goal is to validate this algorithm against the delirium severity index (DSI). Methods: This is multi-center prospective observational trial included an English and French version. We included patients ≥65 years of age and excluded those with critically illness, pain ≥6/10, or communication difficulties. Consenting participants played the serious game and then had DSI and Montreal Cognitive Assessment (MOCA) assessed by trained research assistants. We used logistic regression to assess the relationship between a game score based on > 800 data points and DSI ≥4, controlling for age and sex. Results: We enrolled 306 participants from 3 Canada provinces -92.3% of completed the game. Their average age was 75.9 and 48.7% were women. Their median MOCA score was 23. There were 24/306 patients with a DSI ≥4 out of a possible 21 points. Their mean game score was 0.70 (95% CI 0.63–0.78) vs. 0.61 (95% CI 0.60–0.63) for participants with a DSI<4.The odds of having a DSI≥4 increased 1.89 for each 0.1 increase in game score (95% CI for OR = 1.4–2.7, p=<0.001). Conclusions: The odds of having a DSI≥ 4 significantly increased with game scores. The vast majority of older ED patients could use our game. Given current poor clinical delirium recognition, our “serious game” may help improve delirium recognition.
Authors: J.S. Lee, Mark Chignell, Tiffany Tong, Marcel Émond, Marie‐Josée Sirois, Joshua Norkin Goldstein, Kenneth J. Rockwood, Mary Catherine Tierney
Published in: Innovation in Aging (2017)