Dissipating the fog: Cognitive trajectories and risk factors 1 year after COVID-19 hospitalization.

Abstract

Cognitive impairment is common after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. However, associations between post-hospital discharge risk factors and cognitive trajectories have not been explored. A total of 1105 adults (mean age ± SD 64.9 ± 9.9 years, 44% women, 63% White) with severe coronavirus disease 2019 (COVID-19) were evaluated for cognitive function 1 year after hospital discharge. Scores from cognitive tests were harmonized, and clusters of cognitive impairment were defined using sequential analysis. Three groups of cognitive trajectories were observed during the follow-up: no cognitive impairment, initial short-term cognitive impairment, and long-term cognitive impairment. Predictors of cognitive decline after COVID-19 were older age (β=-0.013, 95% CI =-0.023;-0.003), female sex (β=-0.230, 95% CI=-0.413;-0.047), previous dementia diagnosis or substantial memory complaints (β=-0.606, 95% CI =-0.877;-0.335), frailty before hospitalization (β=-0.191, 95% CI =-0.264;-0.119), higher platelet count (β=-0.101, 95% CI =-0.185;-0.018), and delirium (β=-0.483, 95% CI =-0.724;-0.244). Post-discharge predictors included hospital readmissions and frailty. Cognitive impairment was common and the patterns of cognitive trajectories depended on sociodemographic, in-hospital, and post-hospitalization predictors. Cognitive impairment after coronavirus disease 2019 (COVID-19) hospital discharge was associated with higher age, less education, delirium during hospitalization, a higher number of hospitalizations post discharge, and frailty before and after hospitalization. Frequent cognitive evaluations for 12-month post-COVID-19 hospitalization showed three possible cognitive trajectories: no cognitive impairment, initial short-term impairment, and long-term impairment. This study highlights the importance of frequent cognitive testing to determine patterns of COVID-19 cognitive impairment, given the high frequency of incident cognitive impairment 1 year after hospitalization.