Abstract
BACKGROUND: Panama is a dengue endemic country which experienced a large outbreak in 2024 with over 32,000 reported cases and an incidence rate exceeding 700 cases per 100,000 inhabitants. Despite decades of circulation, the epidemiology of dengue and its heterogeneity in transmission intensity across Panama have not yet been characterised. METHODS: We used 25 years of dengue case notification and population data from across Panama's 16 health regions and 82 districts to characterise dengue epidemiology and transmission intensity in the country. The analytic dataset comprised 128,890 dengue cases, of whom 52% were female and 48% were male; the mean age was 32.4 years (range 0-108 years). Ethnicity data are not collected in Panama's national dengue surveillance system and were therefore unavailable for this analysis. We characterised spatial heterogeneities in delay distributions by fitting parametric probability distributions to epidemiological delays, and demographic differences in the incidence risk ratio of dengue, and of dengue attributable hospitalisations and deaths. We also implemented catalytic models to infer the time-constant dengue force-of-infection (FOI) (i.e. the long-term average annual per capita risk of infection for a susceptible individual) from the age-stratified case notification data reported across Panama during 2000-2024 and explored age- and sex-related differences in dengue case reporting in sensitivity analyses. FINDINGS: We observed spatial variation in delay distributions across health regions. The mean of the regional average time from symptoms onset to (i) reporting was 4.78 days (95% CI: 4.72-4.84 days), (ii) hospitalisation was 4.49 days (95% CI: 4.22-4.76), and (iii) recovery was 7.82 days (95% CI: 6.47-8.85 days). The dengue transmission intensity also showed spatial heterogeneity, with a mean regional per-serotype FOI of 0.008 (95% CrI: 0.004-0.015). The mean regional probability of detecting a secondary case was 0.415 (95% CrI: 0.233-0.871) and the probability of detecting a primary case relative to a secondary case was 0.131 (95% CrI: 0.049-0.348). We found evidence of age- and sex-related differences in dengue reporting. Given the estimated seroprevalence at nine years (obtained from the analysis of case-notification data) is below 40% across the whole country, dengue vaccination with TAK-003 is currently not recommended in Panama according to the WHO guidelines. In the future, age-stratified seroprevalence surveys would be useful to validate these estimates. INTERPRETATION: This analysis provides a characterisation of dengue epidemiology and estimates of dengue FOI and infection burden in Panama across the last 25 years. This work will inform policy decisions at the Ministry of Health of Panama, providing guidance on resource allocation to strengthen the local surveillance system and to decision making on the potential implementation of new interventions. FUNDING: SENACYT-IFARHU (Panamá), UK Medical Research Council and Wellcome Trust.