Defining a long COVID ‘expotype’ within the P4O2 COVID-19 study

<bold>Introduction:</bold> Long COVID is estimated to affect at least 10% of COVID patients, with fatigue being a common complaint. The external exposome has been associated with COVID severity, however its association with long COVID remains underexplored. This study aims to identify possible exposome phenotypes (‘expotypes’) related to long COVID. Method The P4O2 COVID-19 study consists of long COVID patients in the Netherlands and assessed a range of factors and symptoms at 3-6 months post-infection. Fatigue, assessed by the Fatigue Severity Scale (FSS), was used as an indicator of long COVID severity with scores of 4 or more being considered as “positive” for fatigue. The external exposome included factors related to for example air pollution, the food environment, and green space. We performed univariate logistic regression models and a multivariable penalized model (Lasso). Preliminary results Of the 95 participants, only complete cases were included (n=78). Of these, 58 had a FSS score of ≥4. All models were adjusted for age, BMI, and sex. Univariable logistic regression found none of the variables to improve model fit after adjusting for confounders, determined by AIC values, but walkability provided the best estimation (odds ratio (OR)[95%confidence interval(95%CI)1.12[−0.10;0.35]). Multivariable models showed no improved fit either, with a model including PM2.5 (2.06[−0.24;1.77]), NO2(0.86[−0.32;0.00]), and walkability (1.22[−0.05;0.50]) providing the best fit. None of the variables were selected in LASSO regression models. Discussion The external exposome could be associated with fatigue severity in long COVID patients. Prevention strategies and urban planning could take these associations into account top optimize health.