Fighting Omicron epidemic in China: Real-world big data from Fangcang shelter hospital during the outbreak in Shanghai 2022.

Dear editor,

A recent letter published in your journal by Yuan et al. summarized the measures being taken in China to effectively contain COVID-19 epidemics including the recent Omicron outbreak in Shanghai. Omicron has a reproduction number (R0)>10, and causes relatively mild symptoms, both of which aggravate the rapid spread of the virus in a “stealth mode”. Shanghai faces especially severe challenges to control the outbreak at minimal social and economic costs, considering it is a major metropolis with a population of 25 million. As of 30th June 2022, the cumulative diagnosed cases have reached over 600,000 and close to 600 cases have died with or from COVID-19, which poses significant challenges to the city's healthcare system. Fangcang shelter hospital is a type of basic medical facility converted from large public space which offers patients with adequate medical care and provides isolation from their local communities to cut down further spread of the virus, which has been adopted as a central strategy to achieve ‘dynamic zero’, i.e., to minimize COVID-19 cases as quickly as possible at minimal costs.

In line with the previous letter, this approach has been successfully adopted to rapidly subdue several waves of SARS-CoV-2 in China over the past 2 years. Since the Omicron outbreak in Shanghai, more than one hundred Fangcang shelter hospitals have been rapidly converted from large public spaces including schools, exhibition centers and other public facilities, and have made paramount contributions in providing proper care to patients with mild to moderate symptoms, and preventing further viral spreading in the community. On 1st June 2022, the total reopening of Shanghai marks the end of the city's stringent lockdown since late March and the strategic victory of the implemented public health measures and social services.

Here, we present for the first time, the largest cohort of 165,760 COVID-19 cases during the Omicron outbreak in Shanghai. This dataset was acquired at one of the largest Fangcang shelter hospitals in China that was converted from the Shanghai National Exhibition and Convention Center, the world's third-largest exhibition center with an indoor space of 400,000 m. To suppress this extremely contagious Omicron outbreak, Shanghai has employed a series of effective measures to have the majority of its 25 million residents screened by SARS-CoV-2 PCR tests on a very frequent basis. With efficient referral and transfer mechanisms in local communities, the majority of patients were admitted to Fangcang shelter hospitals within 1–2 days after they tested positive. All patients admitted to the Fangcang shelter hospitals were screened daily by PCR tests, and were discharged only when two consecutive PCR tests returned negative. Accordingly, the average length of hospital stay corresponds to approximate the negative conversion time for the infection. The data from this cohort showed an average length of hospital stay of 7.18 ± 3.05 days, as compared to the 16.08 ± 5.13 days for those who were treated at Fangcang shelter hospitals during the outbreak of the original alpha strain in Wuhan, in early March 2020. Additional analysis of the cumulative discharge incidence revealed that about 50% of patients were discharged on day 6 and 90% were discharged on day 10 (Fig. A ), indicating a much faster hospital bed turnover rate and a generally milder spectrum of symptoms for this wave of Omicron infection compared to that of alpha strain infection in Wuhan. The data is further supported by a recent study in Southern California by Lewnard et al., that symptomatic hospital admission due to Delta variant is roughly 2.5 times more than that of Omicron infection.

Figure A

A. The cumulative incidence functions for the time it took for patient discharge. The 50% and 90% cumulative incidence of discharge were labeled in the graph (n=165,760). B. Average length of hospital stay for different age groups. Results were presented as violin plots. One-way ANOVA with Dunnett's multiple comparison tests were performed for any age group compared to the age group 20-29 (control), **** P<0.0001 (n=165,760). C. Average length of hospital stay for patients with or without chronic diseases. Results of patients with (+) or without (-) hypertension or diabetes were analyzed. Unpaired t-tests, **** P<0.0001 (Hypertension n=2,707, Diabetes n=1,096, Total n=37,825). D. Impact of patient source on the average length of hospital stays. Results of patients discovered at their local community or in the fever clinic were shown. Unpaired t-tests, **** P<0.0001 (Community n=24,479, Fever Clinic n=13,346, Total n=37,825). E. Impact of vaccination status and vaccine production company on the average length of hospital stays. Left: results of unvaccinated (-) or vaccinated (+) patients of different doses were shown. Right: results of different vaccine companies including SINOVAC BIOTECH, Beijing Institute of Biological products, CanSino Biologics Inc, Wuhan Institute of Biological Products, Zhifei Longcom Biopharmaceutical. Unpaired t-tests (left) and one-way ANOVA (right), (ns= not significant). (Unvaccinated n=10,984, Vaccinated n=26,841; SINOVAC BIOTECH n = 17,878, Beijing Institute of Biological products n=6,531, CanSino Biologics Inc n=601, Wuhan Institute of Biological Products n=1,405, Zhifei Longcom Biopharmaceutical n=426).

A detailed examination of the age cohorts revealed young adults (age 20–29) were the fastest to recover (6.67 ± 2.75 days), while children (age <10) and the elderly population (age >80) needed the longest length of stay to turn negative, with an average of 7.47 ± 3.01 days and 8.42 ± 3.99 days, respectively. There was also a clear trend that as adults age they gradually recover slower from the infections (Fig. B). Out of the total 165,760 cases admitted at the Fangcang shelter hospital, 37,825 cases had detailed information about the origin of the patients, their accompanying medical conditions, and vaccination history and enabled further analysis. Notably, patients with chronic illnesses such as diabetes and hypertension took remarkably longer to recover compared to the unaffected population (Fig. C). This emphasized the importance to protect the chronically ill populations, as reported before. Interestingly, the patients who were discovered at the fever clinic were discharged much quicker than patients who were screened from the community (Fig. D). This may be explained by a theory that the latter was often asymptomatic and therefore at the preliminary stage of infection, while the former was frequently symptomatic and therefore at a later stage of disease of the natural history of COVID-19 infection. Furthermore, we examined the impact of vaccination on hospital stay, yet found out that neither vaccination nor types of vaccine made remarkable differences (Fig. E). The little difference could be partially explained by the relatively mild nature of Omicron infection as the majority (∼90%) of patients recovered within 10 days. The immune evasion due to epitope drifting is another possible explanation as the Omicron variant has mutated significantly compared to the original strain used for the generation of the approved vaccines in China. The impact of vaccination to reduce the incidence of severe cases or death rates was not examined in this cohort as Fangcang shelter hospitals only handle asymptomatic cases or patients with mild to moderate symptoms.

To conclude, we have provided the largest dataset of COVID-19 naturally recovered cases at the Fangcang shelter hospital which may serve as an excellent reference for public sectors, policymakers and healthcare professionals to assess the pandemic control strategies and better understand the disease characteristics of mild or asymptomatic COVID-19 patients. The Fangcang shelter hospital model in Shanghai has once again demonstrated its significance as a critical measure for patient care and pandemic control and has fulfilled the goal of China's “Dynamic Zero” policy. All in all, while the dawn of victory lies ahead for this wave of the Omicron epidemic in Shanghai, the tug of war between mankind and COVID-19 or other types of infectious disease continues. The Shanghai experience shared here may advance people's understanding of the SARS-CoV-2 Omicron variant as well as the use of Fangcang shelter hospitals. This may help prepare society for the epidemics to come.

Author contributions

L.Y performed data curation, formal analysis, and wrote the manuscript. W.F.L performed data validation and wrote the manuscript. J.S performed data curation and analysis. Both Z.X and J.J helped in data retrieval and project administration. H.X conceptualized and supervised the study as well as edited the manuscript. All authors contributed to the manuscript revision and read and approved the submitted version.

Funding

H Xu is supported by the National Natural Science Foundation of China (Grant No. 31821003), National Key Research and Development Project (Grant No. 2018AAA0100302), Shanghai Municipal Key Clinical Specialty (shslczdzk02602).

Declaration of Competing Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.