Long‐term cognitive impairment after probable delirium in long‐term care residents: A population‐based retrospective cohort study.

Background: The impact of delirium on cognition has not been well‐studied in long‐term care (LTC) residents. This study examined changes in cognition 1 year after a probable delirium episode among LTC residents, compared to LTC residents without probable delirium. We also evaluated whether the relationship between probable delirium and cognitive change differed according to a diagnosis of dementia. Methods: We conducted a population‐based retrospective cohort study using linked health administrative data. The study population included adults aged 65+ residing in LTC in Ontario, Canada and assessed via the Resident Assessment Instrument‐Minimum Dataset between January 1, 2016 and December 31, 2018. Probable delirium was ascertained via the delirium Clinical Assessment Protocol on the index assessment. Cognition was measured quarterly using the Cognitive Performance Scale (range 0–6, higher values indicate greater impairment). Cognitive decline up to 1 year after index was evaluated using multivariable proportional odds regression models. Results: Of 92,005 LTC residents, 2816 (3.1%) had probable delirium at index. Residents with probable delirium had an increased odds of cognitive decline compared to those without probable delirium, with adjusted odds ratios of 1.64 (95% confidence interval [CI] 1.35–1.99), 1.56 (95% CI 1.34–1.85), 1.57 (95% CI 1.32–1.86) and 1.50 (95% CI 1.25–1.80) after 1–3, 4–6, 7–9, and 10–12 months of follow‐up. Residents with probable delirium and a comorbid dementia diagnosis had the highest adjusted odds of cognitive decline (adjusted odds ratio 5.57, 95% CI 4.79–6.48) compared to those without probable delirium or dementia. Residents with probable delirium were also more likely to die within 1 year than those without probable delirium (52.5% vs. 23.4%). Conclusions: Probable delirium is associated with increased mortality and worsened cognition in LTC residents that is sustained months after the probable delirium episode. Efforts to prevent delirium in this population may help limit these adverse effects. See related editorial by Bellelli et al. in this issue. [ABSTRACT FROM AUTHOR]