Delirium is prevalent in older hospital inpatients and associated with adverse outcomes: results of a prospective multi-centre study on World Delirium Awareness Day

Abstract

Background: Delirium is a common severe neuropsychiatric condition secondary to physical illness, which predominantly affects older adults in hospital. Prior to this study, the UK point prevalence of delirium was unknown. We set out to ascertain the point prevalence of delirium across UK hospitals and how this relates to adverse outcomes.

Methods: We conducted a prospective observational study across 45 UK acute care hospitals. Older adults aged 65 years and older were screened and assessed for evidence of delirium on World Delirium Awareness Day (14th March 2018). We included patients admitted within the previous 48 h, excluding critical care admissions.

Results: The point prevalence of Diagnostic and Statistical Manual on Mental Disorders, Fifth Edition (DSM-5) delirium diagnosis was 14.7% (222/1507). Delirium presence was associated with higher Clinical Frailty Scale (CFS): CFS 4-6 (frail) (OR 4.80, CI 2.63-8.74), 7-9 (very frail) (OR 9.33, CI 4.79-18.17), compared to 1-3 (fit). However, higher CFS was associated with reduced delirium recognition (7-9 compared to 1-3; OR 0.16, CI 0.04-0.77). In multivariable analyses, delirium was associated with increased length of stay (+ 3.45 days, CI 1.75-5.07) and increased mortality (OR 2.43, CI 1.44-4.09) at 1 month. Screening for delirium was associated with an increased chance of recognition (OR 5.47, CI 2.67-11.21).

Conclusions: Delirium is prevalent in older adults in UK hospitals but remains under-recognised. Frailty is strongly associated with the development of delirium, but delirium is less likely to be recognised in frail patients. The presence of delirium is associated with increased mortality and length of stay at one month. A national programme to increase screening has the potential to improve recognition.

Keywords: Collaboration; Delirium; Frailty; Older adults.

Fig. 1

Flow chart demonstrating participation in delirium day study. On 14th March 2018, 45 hospitals participated in World Delirium Day study. Two thousand three hundred and eighty-five individuals met the study criteria of admission between 08:00 on 12th March 2018 and 07:59 on 14th March 2018. Thirty-seven individuals were excluded as they were judged to be imminently dying. Seven hundred and nineteen individuals were excluded for logistical reasons. Seventy-seven were excluded as they had not yet had their initial assessment. Forty-five were excluded because the assessment was deemed to cause undue distress. One thousand five hundred and seven individuals were screened with 4AT. Of these, 366 had a score equal to or greater than four and underwent further assessment of delirium using DSM-5 criteria. Of those who were 4AT positive, 222 were proven to have DSM-5 delirium

Fig. 2

a Prevalence of recognised and unrecognised delirium by specialty. The total of each bar represents the overall prevalence of delirium within each specialty; standard error bars show the 95% confidence intervals of prevalence by specialty. The yellow portion of each bar represents recognised delirium, and the red portion of each bar represents unrecognised delirium. Prevalence differed between specialties; however, after controlling for other confounders (e.g. age), specialty was not predictive of delirium prevalence. There were reduced odds of recognition of delirium in patients admitted to general, other, or orthopaedic surgery as compared to acute medicine. b Screening of delirium by specialty. Each bar represents the total percentage of patients who were screened for delirium by the usual care team prior to assessment as part of this study within each specialty; the standard error bars show the 95% confidence intervals of percentage screened. Reduced odds of screening for delirium were exhibited in patients admitted under general or other surgery as compared to acute medicine