In this outbreak, the earliest occurrence C1 was only an index case. Through the trajectory big data analysis means, deep search, isolation and control of close contacts, we set up a link between the various transmission chains, and ultimately traced the source of the infected person A1 returning from Indonesia. A1 in Jakarta to Fuzhou flight, next to a COVID-19 case, and then infected. Later, the virus samples of the confirmed case were analyzed by gene sequence analysis, and the results were compared with the virus gene sequences of the infected persons in this epidemic situation. It was confirmed that all of them belonged to the Branch II (North American branch) of the European L genotype, and compared with the virus carried by A1, the nucleotide sequence similarity was 99.99%. A1 conducted 14 days of centralized medical isolation observation in Fuzhou, during which 5 nucleic acid tests and 2 serological tests were carried out, might be due to too small viral load, the results were negative, which was also consistent with A1 as an asymptomatic patient. This showed that even under strict epidemic prevention and control measures, there were still various uncertain factors.
The analysis of the transmission chain showed that D3, D4 and C3 were colleagues, although C3 might be infected earlier by his wife B2, then C3 subsequently infected D3 and D4 at work, further spreading within two families, that was almost impossible. Because D3 and D4 both developed symptoms later than their wives C4 and C5. And Fig. 2 showed that D3 and D4 were also more likely to be the next generation cases infected by their wives, while other colleagues developed later than them, might be the same generation.
In this outbreak, the average incubation period was 4 days, consistented with the average incubation period of about 5 days of novel coronavirus [3–7], longer than the incubation period of influenza (1–3 days) [8]. Research results shown that people infected with novel coronavirus were highly infectious at the end of the incubation period and beginning of onset. The infectious characteristics were similar to influenza A (H1N1) or seasonal influenza [9], but SARS patients were not infectious before and at the beginning of the disease [10–11]. Therefore, people who had been exposed during the incubation and symptom periods of novel coronavirus, naturally had a higher risk of infection. In addition, no protection such as family exposure, diet exposure, and small space exposure such as within a vehicle, had a greater risk of infection than other exposures. Finally, the overall infection rate of this epidemic was only 0.56% (20/3581), which might be related to the exposure of most people wearing masks, indicating that personal protection was effective in reducing the risk of virus infection.
The experience of this epidemic investigation: firstly, can not be “first impressions are strongest”, take it for granted that the first case found as the initial case, such as, in this epidemic the first case C1 was found only an index case; secondly, once the infected person was found, it was necessary to quickly search for close contacts, even secondary close contacts, and expand the search time range if necessary, such as the establishment of the transmission chain between A1 and B2, because of tracing B2's onset time forward for 1 week and checking her trajectory within a week; thirdly, the isolation observation and nucleic acid detection should be carried out as soon as possible, and the community where the case was located should implement personnel control immediately, limited social contact between community residents and outside, to prevent the spread of the epidemic situation, and carried out nucleic acid testing for all community residents as soon as possible. The outbreak involved only 21 infected persons, but 3581 close contacts were identified, as well as a larger number of secondary close contacts observed in the community, and nucleic acid tests were carried out on all persons. These measures effectively ensured that infected persons could be identified and controlled in the first place; finally, when the traditional epidemiological investigation methods exist that the respondents can not recall clearly, intentionally or unintentionally conceal the facts to cause difficulties in traceability, the use of trajectory big data and other modern means can effectively cooperate with the traditional epidemiological investigation to guide the traceability of the epidemic situation.