Assessing the risk of dengue virus transmission in a non-endemic city surrounded by endemic and hyperendemic areas.

OBJECTIVE: To assess the potential risk of dengue transmission in a non-endemic city using a spatial epidemiological approach.

METHODS: Past dengue exposure of the general population was examined by dengue virus (DENV) IgG testing of archived samples from voluntary blood donors. Vector intensities were determined by local ovitrap index (OI). Analyses were made in the context of population statistics at both the district and sub-district level.

RESULTS: The overall prevalence of DENV IgG was low at 2.25%. Positive donors were more likely to be older, non-Chinese, and female. Neither the OI nor the location of residence was associated with DENV serology. The sub-district level OI was clustered, but no correlation could be confirmed with the location of residence of positive blood donors.

CONCLUSIONS: The cumulative exposure of Hong Kong residents to dengue has so far been low. Coupled with the lack of a spatial relationship between exposed cases and vector intensities, a high risk of local transmission of DENV is not supported. The apparently higher exposure likelihood of females could be explained by past infection in workers from dengue endemic countries, while frequent travel could have exposed older adults to DENV. Continued surveillance, risk assessment, and intensive vector control remain essential to prevent the transformation of a non-endemic to an endemic city.