Bidirectional associations between chronic diseases and Long COVID: a population-based cohort and interrupted time-series analysis.

BACKGROUND

Long COVID has been associated with an increased incidence of chronic diseases, including type 2 diabetes (T2D), which has been suggested to have a bidirectional relationship with COVID-19. We examined these associations in a population-based cohort.

METHODS

Chronic non-communicable diseases (NCDs) diagnosed before (2010-2019) or after (2021-2023) the COVID-19 pandemic onset were tested for associations with Long COVID among Estonian Biobank participants (N = 2,639 with Long COVID, N = 199,171 controls) using logistic regression. Interrupted time-series analysis using a SARIMA model assessed changes in monthly T2D incidence after March 2020. A polygenic risk score (PRS) for severe COVID-19 was examined for association with new-onset T2D.

RESULTS

Pre-existing NCDs were associated with Long COVID, and individuals with Long COVID had increased odds of first-recorded NCD diagnoses after 2020. Long COVID was associated with COPD (OR = 2.33), asthma (OR = 2.18), and chronic kidney disease (OR = 2.15). It was also associated with higher odds of incident asthma (OR = 5.12), COPD (OR = 4.08), and sleep apnea (OR = 3.05), while the association with T2D did not surpass Bonferroni correction. However, T2D incidence increased after March 2020 beyond expected trends (p = 1.84 × 10-80). Individuals in the highest PRS quantile for severe COVID-19 had increased odds of T2D (OR = 1.19).

CONCLUSION

The findings support bidirectional associations between COVID-19 and NCDs, including T2D. Pre-existing T2D was associated with increased odds of Long COVID, while the pandemic period showed increasing trends in T2D incidence.