Hidden transmission and detection delays of dengue and chikungunya in Europe, 2007–2024

Background

Autochthonous outbreaks of dengue (DENV) and chikungunya (CHIKV) are occurring with increasing frequency in Europe. This study aimed to quantify undetected transmission prior to outbreak recognition and to estimate the proportion of cases potentially preventable under earlier detection scenarios.

Methods

We compiled a dataset of European DENV and CHIKV local transmission events (LTEs) from 2007 to 2024. Detection delay was rescaled using virus-specific generation times (18 days for DENV, 12 days for CHIKV), expressing it in generation-time equivalents. A branching process model, parameterised with outbreak-specific effective reproduction numbers, estimated cases occurring during the undetected phase and under counterfactual scenarios with detection advanced by 7, 14, and 21 days. A negative binomial regression model explored the association between detection delay and LTE size.

Results

Thirty LTEs were identified; 23 were included in the modelling (832 confirmed autochthonous cases). The median delay from symptom onset to index case diagnosis was 13 days (range: 4–64). Calendar-day delay was not significantly associated with LTE size (IRR 1.002; 95% CI 0.966–1.048; p = 0.887). In contrast, biologically scaled delay revealed substantial hidden transmission: several Italian outbreaks accumulated 4–6 generation-time equivalents before detection, whereas most French LTEs remained below 2. Advancing detection by 7, 14, and 21 days was associated with case reductions of 28.9%, 49.3%, and 62.8%, respectively.

Conclusions

The number of transmission generations occurring before recognition is a critical factor. Expressing delay in biologically meaningful transmission units more effectively captures hidden transmission. These findings underscore the need for enhanced clinical suspicion, rapid diagnostics, and strengthened seasonal surveillance systems across Europe.

Key messages

• In European DENV and CHIKV outbreaks, detection delay expressed in transmission generations — not calendar days — is the true predictor of preventable burden and should guide surveillance systems.

• Advancing outbreak detection by just one dengue generation time could prevent nearly half of all cases, highlighting the urgency of stronger clinical suspicion and rapid diagnostics across Europe.