Background: In early 2013, a new type of avian influenza, H7N9, emerged in China. It quickly became an issue of great public concern and a widely discussed topic on the Internet. A considerable volume of relevant information was made publicly available on the Internet through various sources.
Objective: This study aimed to describe the outbreak of H7N9 in China based on data openly available on the Internet and to validate our investigation by comparing our findings with a well-conducted conventional field epidemiologic study.
Methods: We searched publicly accessible Internet data on the H7N9 outbreak primarily from government and major mass media websites in China up to February 10, 2014. Two researchers independently extracted, compared, and confirmed the information of each confirmed H7N9 case using a self-designed data extraction form. We summarized the epidemiological and clinical characteristics of confirmed H7N9 cases and compared them with those from the field study.
Results: According to our data updated until February 10, 2014, 334 confirmed H7N9 cases were identified. The median age was 58 years and 67.0% (219/327) were males. Cases were reported in 15 regions in China. Five family clusters were found. Of the 16.8% (56/334) of the cases with relevant data, 69.6% (39/56) reported a history of exposure to animals. Of the 1751 persons with a close contact with a confirmed case, 0.6% (11/1751) of them developed respiratory symptoms during the 7-day surveillance period. In the 97.9% (327/334) of the cases with relevant data, 21.7% (71/327) died, 20.8% (68/327) were discharged from a hospital, and 57.5% (188/327) were of uncertain status. We compared our findings before February 10, 2014 and those before December 1, 2013 with those from the conventional field study, which had the latter cutoff date of ours in data collection. Our study showed most epidemiological and clinical characteristics were similar to those in the field study, except for case fatality (71/327, 21.7% for our data before February 10; 45/138, 32.6% for our data before December 1; 47/139, 33.8% for the field study), time from illness onset to first medical care (4 days, 3 days, and 1 day), and time from illness onset to death (16.5 days, 17 days, and 21 days).
Conclusions: Findings from our Internet-based investigation were similar to those from the conventional field study in most epidemiological and clinical aspects of the outbreak. Importantly, publicly available Internet data are open to any interested researchers and can thus greatly facilitate the investigation and control of such outbreaks. With improved efforts for Internet data provision, Internet-based investigation has a great potential to become a quick, economical, novel approach to investigating sudden issues of great public concern that involve a relatively small number of cases like this H7N9 outbreak.
Keywords: Internet; big data; disease outbreaks; epidemiology; influenza A virus, H7N9 subtype.