Replication Data for: Seroprevalence of Dengue, Chikungunya and Rift Valley fever virus IgG antibodies in Kenyan blood donors

<p>This is a replication dataset for the manuscript titled: "<strong><i>Seroprevalence of Dengue, Chikungunya and Rift Valley fever virus IgG antibodies in Kenyan blood donors</i></strong>."</p> <p>Aedes-borne arboviruses such as dengue (DENV), chikungunya (CHIKV), and Rift Valley fever virus (RVFV) are increasing globally due to climate change, urbanization, and increased human mobility, and continue to cause recurrent outbreaks in Kenya. However, national data on population exposure remain limited, with most studies being outdated or region-specific. Surveillance is further complicated by overlapping clinical symptoms, while molecular methods detect only short-term infections and serological approaches capture longer-term exposure. Conventional assays such as ELISAs and neutralization tests are often costly, time-consuming, or unsuitable for large-scale studies. To address this, we developed and validated a multiplex bead-based immunoassay (arbo-plex MIA) on the Luminex® platform for the simultaneous detection of IgG antibodies to DENV, CHIKV, and RVFV. Using the Foci Reduction Neutralization Test (FRNT) as the gold standard, we estimated the sensitivity and specificity of the assay, achieving a high proportion of correctly classified samples, and further compared its performance with commercial ELISAs to assess agreement. Assay optimization was conducted using multiple cohorts, including adult samples from coastal Kenya (n=147) for CHIKV and RVFV, and the KIPMAT (n=76) and CPGH (n=57) cohorts for DENV. Additional samples from Kilifi (n=795) and Nairobi (n=843) Health and Demographic Surveillance Systems (HDSS) were used for comparison with ELISAs, alongside a panel of pooled convalescent sera (n=10) from Kenya and Tunisia. The validated assay was then applied to a large national dataset of blood donor samples (n=11,420) to estimate seroprevalence. We used a Bayesian multilevel logistic regression framework to generate pathogen-specific estimates, accounting for age and sex differences in sampling. Assay performance was incorporated into the model through sensitivity and specificity parameters informed by validation data, with uncertainty propagated into the final estimates. Population-representative seroprevalence was obtained by post-stratifying model outputs using age and sex distributions from the 2019 Kenya Population and Housing Census.</p>