Towards a Neurocognitive Profile of Post-Acute COVID-19: A Longitudinal Study Integrating Subjective Complaints and Computerized Cognitive Testing

Abstract

Highlights What are the main findings? Attention/psychomotor impairment was the most persistent objective cognitive deficit, affecting approximately two-thirds of participants at both assessments. Longitudinal change appeared domain-specific: selected learning/working-memory measures improved nominally, although these changes did not survive correction for multiple comparisons, and baseline subjective complaints did not predict cognitive performance at follow-up. What are the implications of the main findings? Subjective symptom screening alone may not adequately identify persistent objective cognitive impairment in post-COVID patients. Objective, repeatable computerized cognitive assessment may improve the longitudinal evaluation of post-COVID cognitive dysfunction. Abstract Background/Objectives: The relationship between subjective cognitive complaints and objective performance in post-acute COVID syndrome (PACS), and the persistence of objective impairment over extended follow-up, remain unclear. This prospective longitudinal study characterized the neurocognitive profile of PACS by integrating computerized cognitive testing with structured evaluation of subjective complaints across an extended post-infection interval. Methods: A total of 102 adults with a history of COVID-19 infection (3–35 months post-infection; mean age 40.2 ± 10.95 years; 70.6% female) underwent baseline (V1) CogState Battery testing (seven subtests) and structured assessment of persistent post-COVID complaints; 67 participants (65.7%) constituted the matched longitudinal sample and completed the follow-up (V2) after a mean inter-visit interval of 6.69 ± 1.00 months. Domain-specific composites (Attention/Psychomotor function, Learning/Working memory, and two Global indices) were derived from age-adjusted z-scores, with impairment defined as z ≤ −1.5. Results: Persistent complaints were reported by 63.7% of participants at baseline. In the matched longitudinal sample (n = 67) Attention/Psychomotor impairment was the most prevalent and stable finding in the paired sample (65.7% at V1 and 67.2% at V2; McNemar p = 1.000), whereas ONB-defined working-memory impairment declined nominally in the paired sample (39.4% to 24.2%; p = 0.006), although this change did not remain significant after correction for multiple comparisons (BH-FDR q = 0.063). In the baseline sample persistent complaints were associated with poorer objective performance at V1 across all composite indices with small-to-moderate effect sizes (all BH-FDR-adjusted q = 0.042–0.047). These associations were not observed in the follow-up sample at V2. Age was the most robust demographic correlate of impairment, particularly for visuospatial associative memory (CPAL; β = −0.349, p < 0.001); clinical and demographic predictors explained 9.4–13.9% of the variance (R2 = 0.094–0.139) in significant models at Visit 1. Conclusions: Attention/psychomotor dysfunction constituted a persistent objective deficit after COVID-19. Subjective complaints at baseline were associated with objective impairments but did not predict cognitive performance at follow-up. These findings support the use of objective computerized cognitive testing alongside symptom screening in the clinical assessment of post-COVID patients. Interpretation is limited by the absence of pre-COVID cognitive data and of a demographically matched healthy, uninfected control group.