Quantify cognitive deficits in patients with postacute sequelae of COVID-19 (PASC) and identify key variables related to cognitive impairment in PASC.

Patients with polymerase chain reaction-confirmed COVID-19 underwent a comprehensive neuropsychological evaluation. The comparison group included patients without neurological disorders determined by the neuropsychologist to be cognitively intact. Cognitive impairment was defined as impairment (Composite T ≤35) in 1 of 6 cognitive domains. The PASC group was split into impaired or intact based on the above criteria. Multivariable logistic regression models assessed predictors including demographics, COVID-19 severity, clinical characteristics, and mood.

There were 210 patients with PASC, predominantly female (73.3%, P < .001), without other demographic differences when compared with 369 normal controls. Patients with PASC were more likely to have cognitive impairment (odds ratio 3.61; 95% confidence interval, 2.36-5.54; P < .001) compared with controls, with significantly lower scores in domains of memory, language, processing speed, visuospatial function, executive function (P < .001), and higher depressive (P = .004) and anxiety symptoms (P = .003). Patients with PASC who demonstrated cognitive impairment (n = 93) had higher body mass index compared with those with PASC without cognitive impairment (n = 117), without differences in other predictors.

Patients with PASC are almost 4 times more likely to evidence cognitive dysfunction compared with normal controls. Forty-four percent of patients with PASC demonstrated cognitive deficits about 7 months from infection. Estimated premorbid intelligence significantly correlated with impairment. Higher body mass index was the only metric shown to differentiate those with PASC and cognitive impairment from those with PASC who were cognitively intact.