Back

Persistence of neuropsychiatric symptoms associated with SARS-CoV-2 positivity among a cohort of children and adolescents

Castro, V. M.; Gunning, F. M.; Perlis, R. H.

2021-09-29 psychiatry and clinical psychology
10.1101/2021.09.28.21264259 medRxiv
Show abstract

BackgroundPost-acute sequelae of COVID-19 are common among adults. The prevalence of such syndromes among community samples of children and adolescents remains less well characterized. MethodWe identified all individuals age 5-18 across 2 New England health systems who had a positive SARS-CoV-2 PCR test between 3/12/2020 and 4/18/2021 and at least 90 days of follow-up visits documented in electronic health records. We identified neuropsychiatric symptoms in intervals prior to, and following, this testing using a previously-derived set of ICD-10 codes and natural language processing terms. Primary analysis examined sociodemographic features associated with presence of at least one incident (i.e., new-onset) neuropsychiatric symptom between 90 and 150 days after an initial positive test for COVID-19. ResultsAmong 5058 children (50% female, 2.9% Asian, 6.3% Black, and 63% White; 30% Hispanic; mean age was 12.4 (IQR 8.9-15.6), 366 (7.2%) exhibited at least one new-onset neuropsychiatric symptom between 90 and 150 days following initial SARS-CoV-2 test positivity. The most common incident symptoms at 90-150 days were headache (2.4%), mood and anxiety symptoms (2.4%), cognitive symptoms (2.3%), and fatigue (1.1%). In regression models, older children, girls, those with Hispanic ethnicity, those with public versus private insurance, and those with greater overall burden of medical comorbidity were more likely to exhibit subsequent symptoms. ConclusionThe prevalence of neuropsychiatric symptoms between 3- and 5-months following SARS-CoV-2 test positivity is similar to that observed in the period prior to infection. Prospective controlled studies will be needed to further refine these estimates.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.