Extremes of baseline cognitive function determine the severity of delirium: a population study

• Published in: Brain (London, England : 1878) Vol. 146; no. 5; pp. 2132 - 2141
• Main Authors: Tsui, Alex, Yeo, Natalie, Searle, Samuel D, Bowden, Helen, Hoffmann, Katrin, Hornby, Joanne, Goslett, Arley, Weston-Clarke, Maryse, Lanham, David, Hogan, Patrick, Seeley, Anna, Rawle, Mark, Chaturvedi, Nish, Sampson, Elizabeth L, Rockwood, Kenneth, Cunningham, Colm, Ely, E Wesley, Richardson, Sarah J, Brayne, Carol, Terrera, Graciela Muniz, Tieges, Zoë, MacLullich, Alasdair M J, Davis, Daniel
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 02.05.2023
• Subjects: Cognition; Delirium - epidemiology; Humans; Original; Prospective Studies; Research Design; epidemiology; baseline cognitive function; delirium
• ISSN: 0006-8950; 1460-2156
• DOI: 10.1093/brain/awad062

Abstract Although delirium is a significant clinical and public health problem, little is understood about how specific vulnerabilities underlie the severity of its presentation. Our objective was to quantify the relationship between baseline cognition and subsequent delirium severity. We prospectively investigated a population-representative sample of 1510 individuals aged ≥70 years, of whom 209 (13.6%) were hospitalized across 371 episodes (1999 person-days assessment). Baseline cognitive function was assessed using the modified Telephone Interview for Cognitive Status, supplemented by verbal fluency measures. We estimated the relationship between baseline cognition and delirium severity [Memorial Delirium Assessment Scale (MDAS)] and abnormal arousal (Observational Scale of Level of Arousal), adjusted by age, sex, frailty and illness severity. We conducted further analyses examining presentations to specific hospital settings and common precipitating aetiologies. The median time from baseline cognitive assessment to admission was 289 days (interquartile range 130 to 47 days). In admitted patients, delirium was present on at least 1 day in 45% of admission episodes. The average number of days with delirium (consecutively positive assessments) was 3.9 days. Elective admissions accounted for 88 bed days (4.4%). In emergency (but not elective) admissions, we found a non-linear U-shaped relationship between baseline global cognition and delirium severity using restricted cubic splines. Participants with baseline cognition 2 standard deviations below average (z-score = −2) had a mean MDAS score of 14 points (95% CI 10 to 19). Similarly, those with baseline cognition z-score = + 2 had a mean MDAS score of 7.9 points (95% CI 4.9 to 11). Individuals with average baseline cognition had the lowest MDAS scores. The association between baseline cognition and abnormal arousal followed a comparable pattern. C-reactive protein ≥20 mg/l and serum sodium <125 mM/l were associated with more severe delirium. Baseline cognition is a critical determinant of the severity of delirium and associated changes in arousal. Emergency admissions with lowest and highest baseline cognition who develop delirium should receive enhanced clinical attention. In a prospective study of older adults with unscheduled hospital admissions, Tsui et al. observe a U-shaped relationship between baseline global cognition—assessed by previous telephone interview—and the severity of episodes of delirium. See Lennertz and Sanders (https://doi.org/10.1093/brain/awad115) for a scientific commentary on this article.