Discussion

COVID-19 is the most extensive global pandemic to afflict humanity in the past decades. It is a serious crisis and a daunting challenge for the whole world. In the face of the unprecedented, sudden, and violent outbreak of disease, information disclosure is the best vaccine without special drugs. Shenzhen and Tianjin, 2 large cities in China, 1 south and 1 north, can timely disclose and publish case data, including: age, gender, current address, exposure source, onset date, treatment date, confirmed date, discharge date, clinical symptoms, clinical classification, epidemiology track after disease, and virus nucleic acid testing number. The governments of the 2 cities are in the forefront of comprehensive evaluation of the COVID-19 epidemic information. ^{Reference Zhou, Xu and Dacheng5} The public has real-time understanding of the cases in these 2 cities, ^{Reference Cao, Li and Liu7,Reference Wen, Wei and Li8} which has also triggered a lot of public discussion and private analysis. ⁹ Through this information, the public can quickly predict whether they are at risk of infection, and can make timely preventive actions. ^{Reference Chen, Wang and Li10}

The current study found that the completeness of the information released by the 2 cities in order of diagnosis was as follows: basic information (age, gender, current address) was better than exposure information (exposure source, onset date), than clinical diagnosis and treatment information (treatment date, confirmed date, discharge date, clinical symptoms, clinical classification, epidemiology track after disease, and virus nucleic acid testing number). This was also a common rule for the release of case information in other regions. The reason might be that: (1) the basic information of the case was more convenient to obtain; (2) patients might have recall bias or conceal contact history when answering questions from epidemiological investigators; and (3) the information released by the government was usually just after diagnosis, and the follow-up treatment would no longer be collected, resulting in most missing clinical diagnosis and treatment information. Therefore, the lack of information on the incubation period, symptoms at onset, clinical characteristics, and outcomes of the 2 cities resulted in limited role in the analysis of the clinical characteristics of new infectious diseases and the risk assessment of medical resource occupation.

The continuity and stability of the case information variables released by Shenzhen were better than those of Tianjin, while the comprehensiveness was slightly better of Tianjin than that of Shenzhen. The 2 cities have separate emphasis on the release of case information. Tianjin added nucleic acid detection information to the later published cases, indicating that the instability of information is bias caused by the influence of hotspots of public concern. Shenzhen paid more attention to the release of the date of onset and treatment, considering that the time interval between onset and consultation is an important indicator for evaluating the risk of epidemic spread. ^{Reference Lau, Zheng and Tsang11} On the whole, the continuity and stability of information variables were affected by multiple factors, such as public attention, scientific prevention, and control. However, the public demand for epidemic data has not been fully satisfied by the data supply. From the current situation of uneven release levels and lack of standardization, China still needs to further improve relevant laws and regulations, formulates more specific and operable implementation rules and data standards, so as to make the epidemic data opening work more legal, compliant, uniform, and orderly.

Under the situation of a similar population, the number of confirmed cases and incidence rate in Shenzhen is significantly higher than that in Tianjin. There may be 2 reasons for this finding: first, there is a large gap between the immigrant population and permanent residents in the 2 cities. The proportion of immigrant population in Shenzhen is 65%, and 31% in Tianjin. Second, there is a big gap between the floating populations of Hubei provinces (the COVID epidemic outbreak center) in the 2 cities. According to the big data of Tianyan on Baidu map, during the period from January 1 to 20, 2020, the daily emigrated population in Hubei province with Shenzhen as the destination was 1.73-2.92%, and that of Tianjin was 0.32-0.64%. To sum up, the incidence rate of Shenzhen was significantly higher than that of Tianjin in a similar time period. This may be the main reason for the 2 cities with a larger gap between local and imported cases.

The first cases were reported on January 20 and 21, and the first-level public health incident response were launched on January 23 and 24 in Shenzhen and Tianjin, respectively. The time interval curve from the onset to the time of medical treatment in 2 cities showed the same trend as SARS in 2003. ^{Reference Pan, Liu and He12} It took more than 40 d from the onset of the first case to the first time that there were no new cases in the region, which was consistent with the model prediction carried out by Sheng et al. ^{Reference Sheng, Wu and Xiao13} It has been reported that a 1-d delay in implementing social distancing resulted in a containment delay of 2.41 (95% CI, 0.97-3.86) d. ^{Reference Du, Xu and Wang14} Although at the early stage of the epidemic, due to the lack of awareness of the symptoms, most of the infected people chose to take medicine on their own, which caused the time interval between the onset and the visit to the clinic of the case uneven. However, due to the early response of the 2 cities, and with the effect of prevention and control measures, the range of average time interval between the onset and treatment was greatly reduced, and the epidemic basically subsided by the end of February.

At the same time, this study also has many limitations and shortcomings. The current study only relies on information published by the government; hence, there will be some bias from the actual epidemic situation. ^{Reference Liu, Liao and Qian15,Reference Dong, Li and Bai16} Whether there is a correlation between the severity of the epidemic situation and the level of the epidemic data release still needs further research. In addition, how to effectively control the disease and improve self-protection of hospital staff in the early stage of the epidemic, when the disease is not fully understood and when medical resources are not suitable, should be the focus of future public health emergencies for unknown reasons.

In conclusion, our research proposed a new strategy that, based on public information sources, some basic epidemiological analysis could be conducted on the local epidemic. The epidemiological characteristics and sources of Tianjin and Shenzhen had lots of both similarities and differences. It has certain guiding significance for prevention and control of epidemic situations in different cities.