Back

Ecological drivers of chikungunya virus transmission: baseline results from a geographically structured, longitudinal study in southern Thailand

Rapheal, E.; Buddhari, D.; Meyer, A.; Hunsawong, T.; Mendoza Guerrero, S.; Fernandez, S.; Farmer, A.; Anderson, K.; Dean, N.; Perkins, T. A.; Chusri, S.; Stoddard, S. T.

2026-02-06 epidemiology
10.64898/2026.02.04.26345533 medRxiv
Show abstract

BackgroundChikungunya virus (CHIKV) infection can cause significant and long-term morbidity. CHIKV typically appears in explosive outbreaks then vanishes for decades, but evidence from longitudinal studies suggests that it may persist in some populations through low levels of subclinical infection. These epidemiologic dynamics complicate prediction of CHIKV outbreaks and intervention trial planning. Songkhla province in southern Thailand is a promising location for vaccine trials due to its recent history of CHIKV outbreaks (2008 and 2018) and emerging data suggesting low levels of interepidemic transmission. Methodology and Principal FindingsHere, we describe baseline findings from a longitudinal cohort study (2022-2024, n = 5000) of CHIKV transmission in Songkhla. We used serocatalytic models to estimate CHIKV force of infection (FOI), adjusting for risk factors including developed land use surrounding the household, access to garbage collection and clean water, and others. Baseline CHIKV seropositivity in the cohort was 34.6%. Our crude catalytic model estimated 0.0160 (95% CI: 0.0153-0.0168) annual FOI. In adjusted FOI models, a higher proportion of developed land use was associated with an increase in risk of CHIK seropositivity among participants under age 12 (OR: 1.36; 1.25-1.48), who would have been exposed only to the 2018 outbreak, but associated with a decrease in risk (OR: 0.96; 0.94-0.99) among those ages 12 and up, who would have been exposed to two or more outbreaks. ConclusionsThese findings suggest that CHIKV infection risk is highly spatially variable, with prediction complicated by historical differences in virus strain and vector. Distinct ecological patterns of exposure are consistent with the 2018-20 outbreak affecting largely urban areas, with little to no exposures in rural areas, while the 2008-09 outbreak was concentrated more heavily in rural areas. Understanding the ecological drivers of this variation has important implications for identifying regions of highest risk for a future CHIKV outbreak.

Matching journals

The top 4 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.