Back

Chronic post-COVID neuropsychiatric symptoms (PCNPS) persisting beyond one year from being infected during the Omicron wave

Chau, S. W. H.; Chue, T. M.; Lam, T. C.; Lai, Y. L.; Chan, R. N. Y.; Wong, P. W.; Li, S. X.; Liu, Y.; Chan, J. W.; Chan, P. K.-s.; Lai, C. K.-C.; Leung, T. W.; Wing, Y. K.

2024-09-03 psychiatry and clinical psychology
10.1101/2024.09.01.24312691 medRxiv
Show abstract

The heterogeneity of chronic post-COVID neuropsychiatric symptoms (PCNPS), especially after infection by the Omicron strain, has not been adequately explored. Our pre-registered hypotheses are 1. chronic PCNPS in patients infected with SARS-CoV-2 over a year ago during the Omicron wave showed a similar clustering pattern with symptoms in patients infected with pre-Omicron strain; 2. these chronic PCNPS are associated with a) clinical risk factors, such as, severity of the acute infection; b) socioeconomic status e.g., level of deprivation; and c) pre-infection vaccination status. We assessed 1205 subjects using app-based questionnaires and cognitive tasks. Partial network analysis on chronic PCNPS in this cohort produced two major symptom clusters (cognitive complaint-fatigue cluster and anxiety-depression symptoms cluster) and a minor headache-dizziness symptoms cluster, like our pre-Omicron cohort. Subjects with high number of symptoms (4 or more) can be further grouped into two distinct phenotypes: a cognitive complaint-fatigue predominant phenotype (CF) and another with symptoms across multiple clusters (AD-CF). Multiple logistic regression showed that both phenotypes are predicted by the level of deprivation before infection (adjusted p-value for CF and AD-CF = 0.025 and 0.0054 respectively). While the severity of acute COVID (adjusted p-value = 0.023) and the number of pre-existing medical conditions predict only the CF phenotypes (adjusted p-value = 0.003), past suicidal ideas predict the AD-CF phenotype (adjusted p-value < 0.001). Pre-infection vaccination status did not predict either phenotype. Our finding suggests that we should recognize the heterogeneity under the umbrella of chronic PCNPS, and a holistic bio-psycho-social approach is essential in understanding them.

Matching journals

The top 6 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.