Neurocognitive and Neuropsychiatric Trajectories in a Post-COVID Cohort: A Descriptive Longitudinal Study

Cognitive dysfunction ("brain fog") is a common manifestation of post-acute COVID-19 syndrome (PACS) and may persist long after the acute infection. While cross-sectional studies have described cognitive deficits, longitudinal evidence on recovery trajectories remains limited.

We conducted a longitudinal observational study of neurocognitive performance and neuropsychiatric symptoms in patients with PACS. Participants underwent assessment with 20 standardized tests covering five cognitive domains (memory, attention, language, executive functions, psychomotor processing speed); anxiety, depression, and sleep quality were assessed at three time points. Changes were analysed using the Friedman test.

Forty-two patients were included (median age 57 years; 35.7% female) from a predominantly hospitalized cohort (81% hospitalised; 66.7% requiring respiratory support). Patients who completed all three assessments (completers, n = 42) were compared with those who attended the first evaluation but did not complete follow-up (non-completers, n = 544); completers were more severely ill during the acute phase rather than healthier or more motivated. At the group level, statistically significant improvements over time were observed across the whole sample in verbal short-term learning, visuospatial memory, working memory, constructional praxis, phonological verbal fluency, and psychomotor processing speed (all p ≤ 0.05); after Benjamini-Hochberg adjustment across the twenty cognitive outcomes, visuospatial span forward and backward and psychomotor processing speed remained significant (all FDR-adjusted p ≤ 0.013), with the change confined to the first six months. Sleep quality also improved (p < 0.0001).

In this cohort, group-level performance improved in six of the twenty tests administered, of which three remained significant after correction for multiple comparisons, while 28 of 42 patients (66.7%) still scored in the impaired range on at least one test at 12 months, and 17 (40.5%) on two or more. These findings highlight the importance of long-term neuropsychological monitoring and integrated cognitive-psychiatric evaluation in post-COVID care. Given the small, predominantly hospitalized sample, improvements should be interpreted cautiously and confirmed in larger controlled studies, although the use of alternate forms for part of the battery makes task-specific learning an incomplete explanation.