Back

Similarities between long COVID and cognitive impairment and potential implications; Results from the 2022 BRFSS

Adams, M.; Grandpre, J.

2024-04-05 epidemiology
10.1101/2024.04.02.24305215 medRxiv
Show abstract

BackgroundCOVID has been linked to cognitive issues with brain fog a common complaint among adults reporting long COVID (symptoms lasting 3 or more months). ObjectiveTo study similarities and differences between cognitive impairment (CI) (the cognitive disability measure) and long COVID. MethodsUsing 2022 BRFSS data from 50 states and 169,894 respondents in 29 states with COVID vaccine data, respondents with CI and long COVID were compared in unadjusted analysis and logistic regression. Apparent vaccine effectiveness was compared in the 29 states. ResultsPrevalence of long COVID was 7.4% (95% CI 7.3-7.6) and CI was 13.4% (13.2-13.6) with both rates higher among women, ages 18-64 years, Hispanics, American Indians, ever smokers, those with depression, e-cigarette users, and those with more of the co-morbidities of diabetes, asthma, COPD, and obesity. The strong association between long COVID and CI was confirmed. Apparent vaccine effectiveness of 3 or more doses vs <3 was 38% for long COVID and 35% for CI, in both cases reducing rates for 3 or more doses to those comparable to adults with 0 comorbidities and showing dose response gradients. For CI, apparent vaccine effectiveness was similar for respondents with or without long COVID. Logistic regression confirmed most results except the magnitude of vaccine effectiveness on CI was reduced in some models while vaccine effectiveness for long COVID was confirmed. ConclusionsMore research is needed to understand the apparent effectiveness of COVID vaccines on CI but, if confirmed, results could expand the list of non-infectious outcomes for which mRNA vaccines can be effective.

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.