Detection of Increased Scarlet Fever Incidence Using Digital Surveillance Data. (March 2022)

• Record Type: Journal Article
• Authors: Chua, H.; Zhou, W.; Cowling, B.J.; Lau, E.H.
• Is Part Of: International journal of infectious diseases. Volume 116(2022)Supplement
• Page Start: S99
• Page End: S100
• Publication Date: 2022-03
• Subjects: Communicable diseases -- Periodicals; Communicable Diseases -- Periodicals; Communicable diseases; Periodicals; Electronic journals; 616.9
• DOI: 10.1016/j.ijid.2021.12.235
• Languages: English
• ISSNs: 1201-9712

Abstract

Abstract : Purpose: Digital data could be used to monitor disease activity and to detect emerging infections, and is motivated by its abundance, affordability and accessibility. Studies show that Google Trend data alone may not accurately predict incidence of influenza-like-illnesses, but there is a potential to capture the epidemiological characteristics or trend of diseases qualitatively and to provide early warning of epidemics. In this study, we validated the use of Google queries of scarlet fever using routine surveillance data, and inferred disease trend of scarlet fever in other countries/regions where specific surveillance for scarlet fever is not available. Methods & Materials: We collected scarlet fever surveillance data from Hong Kong, England, Germany and Taiwan from where surges in scarlet fever incidence have been reported. We obtained language-specific Google queries of scarlet fever corresponding to the time period where surveillance data was available in each country/region. We constructed Poisson regression models including trends and seasonality and tested for their significance using likelihood ratio tests, and also tested the presence of change point, separately for scarlet fever surveillance data and Google queries. Similar methods were applied to characterize Google query data for scarlet fever in other countries/regions. Results: We identified change points in scarlet fever incidence resulting in long-term level changes from Hong Kong and England in 2011 and 2014, respectively. Although no change point was detected in Germany and Taiwan, both showed increasing trends. The overall characteristics identified from Google query data were consistent with those from existing surveillance data in each country/region, such as trend, seasonality and long-term change in disease incidence. Among 63 countries/regions studied, we identified level changes in Argentina, China, Egypt, Ireland, South Korea, Philippines, Serbia, Singapore and United Arab Emirates. 38/63 (60.3%) of the countries/regions showed increasing trends. Seasonality was frequently observed across countries/regions. Conclusion: We validated Google queries of scarlet fever using 4 countries/regions with existing surveillance data. Digital data suggested increased scarlet fever incidence globally especially in Europe and some parts of Asia. Our findings highlight the potential use of digital surveillance for detection of re-emerging and emerging diseases and may strengthen pandemic preparedness and response.