Clinical Manifestations

Neurocognitive symptoms are among the most debilitating features of long COVID (persistent, ongoing symptoms and conditions following SARS-CoV-2 infection). Traditional neurocognitive screening instruments have been found to be non-sensitive to post-COVID cognitive impairments, often described as 'brain fog', that disrupt performance of complex daily routines. Accurate screening and diagnosis of cognitive impairment is therefore a critical area of clinical research for people with long COVID. We examined whether self-reported or neuropsychological cognitive function tests were more strongly associated with long COVID.

In a sample from a community-engaged cross-sectional cohort study, we fitted a binary logistic regression model for the outcome of long COVID status (having history of COVID illness, plus a Post-COVID Functional Status Scale (PCFS) score >=2), classified by MoCA cut score weighted for race/ethnicity, and associated with the Patient-Reported Outcomes Measurement Information System (PROMIS) 8-item Cognitive Functioning t-scores, controlling for age.

We analyzed data from N = 174 participants (n = 67 cases, n = 107 controls). Our final model was fitted with 5-knot natural cubic splines for age and PROMIS Cognition scores. At a "Non-impaired" score on the MoCA, a PROMIS t-score at the lower quartile of 36.67 increased the odds of having long COVID by over seven times at age 25 (1.0,56.89, p = 0.01), over five times at ages 45 (1.4,18.6, p = 0.001) and 55 (1.4,20.4, p = 0.002), and six times at ages 65 (0.9,51.2, p = 0.02) and 75 (0.8,48.9, p = 0.03). By contrast, scores above and below race/ethnicity-weighted MoCA cut score did not significantly affect the odds of having long COVID at any age (p value ranges 0.22-0.25 and 0.19-0.22, respectively) adjusted for mean PROMIS score.

Across ages, low PROMIS Cognitive Function scores showed between five- and six-fold increases in odds of long COVID, adjusted for age and MoCA score. This self-rated cognitive function scale was the best predictor of functional status changes attributed to post-COVID-19 health changes in our sample.