Volatility and heterogeneity of vaccine sentiments means continuous monitoring is needed when measuring message effectiveness (preprint)

Background The success of vaccination programs often depends on the effectiveness of the vaccine messages, particularly during emergencies such as the COVID-19 pandemic. The current suboptimal uptake of COVID-19 vaccines across many parts of the world highlights the tremendous challenges in overcoming vaccine hesitancy and refusal even in the context of a world-devastating pandemic. Methods We conducted a randomized controlled trial in Hong Kong to evaluate the impact of seven vaccine messages on COVID-19 vaccine uptake (with the government slogan as the control). The participants included 127,000 individuals who googled COVID-19-related information during July-October 2021. Results The impact of vaccine messages on uptake varied substantially over time and among different groups of users. For example, the message that emphasized the indirect protection of vaccination on family members (i) increased overall uptake by 30% (6-59%) in July but had no effect afterwards for English language users; and (ii) had no effect on overall uptake for Chinese language users throughout the study. Such volatility and heterogeneity in message effectiveness highlight the limitations of one-size-fits-all and static vaccine communication. Conclusions Epidemic nowcasting should include real-time monitoring of vaccine hesitancy and message effectiveness, in order to adapt messaging appropriately. This dynamic dimension of surveillance has so far been underinvested. Trial registration The study was registered at ClinicalTrials.gov (NCT05499299).

Estimating the transmission dynamics of Omicron in Beijing, November to December 2022 (preprint)

We tracked the effective reproduction number Rt of Omicron BF.7 in Beijing in November - December 2022 by fitting a transmission dynamic model parameterized with real-time mobility data to (i) the daily number of new symptomatic cases on November 1-11 (when the zero-covid interventions were still strictly enforced) and (ii) the proportion of individuals who participated in online polls on December 10-14 and self-reported to have been previously test-positive. After the announcement of "20 measures", we estimated that Rt increased to 3.42 (95% CrI: 2.79 - 4.17) on November 18. Infection incidence peaked on December 10, and the cumulative infection attack rate was 42.5% (95% CrI: 20.3 - 63.9) on December 14. Surveillance programmes should be rapidly set up to monitor the evolving epidemiology and evolution of SARS-CoV-2 across China.

Modelling the adjustment of COVID-19 response and exit from dynamic zero-COVID in China (preprint)

Background Since the initial Wuhan outbreak, China has been containing COVID-19 outbreaks through its "dynamic zero-COVID" policy. Striking a balance between sustainability and cost-benefit, China has recently begun to adjust its COVID-19 response strategies, e.g. by announcing the "20 measures" on 11 November and further the "10 measures" on 7 December 2022. Strategies for safely exiting from dynamic zero-COVID (i.e. without catastrophically overburdening health systems and/or incurring unacceptably excessive morbidity and mortality) are urgently needed. Methods We use simulations to assess the respective and combined effectiveness of fourth-dose heterologous boosting, large-scale antiviral treatment and public health and social measures (PHSMs) that might allow China to further adjust COVID-19 response and exit from zero-COVID safely after 7 December 2022. We also assess whether local health systems can cope with the surge of COVID-19 cases posed by reopening, given that chunyun, a 40-day period with extremely high mobility across China associated with Spring Festival, will begin on 7 January 2023. Findings Reopening against Omicron transmission should be supported by the following interventions: 1) fourth-dose heterologous boosting 30-60 days before reopening by vaccinating 4-8% of the population per week with [≥]85% uptake across all ages; 2) timely antiviral treatment with [≥]60% coverage; 3) moderate PHSMs to reduce transmissibility by 47-69%. With fourth-dose vaccination coverage of 85% and antiviral coverage of 60%, the cumulative mortality burden would be reduced by 26-35% to 448-503 per million, compared with reopening without any of these interventions. Simultaneously reopening all provinces under current PHSMs would still lead to hospitalization demand that are 1.5-2.5 times of surge hospital capacity (2.2 per 10,000 population per day). Interpretation Although the surge of disease burden posed by reopening in December 2022-January 2023 would likely overload many local health systems across the country, the combined effect of vaccination, antiviral treatment and PHSMs could substantially reduce COVID-19 morbidity and mortality as China transits from dynamic-zero to normality. Planning for such a nationwide, coordinated reopening should be an urgent priority as part of the global exit from the acute phase of the COVID-19 pandemic.

Population-based sero-epidemiological estimates of real-world vaccine effectiveness against Omicron infection in an infection-naive population, Hong Kong, January to July 2022 (preprint)

The SARS-CoV-2 Omicron variant has demonstrated enhanced transmissibility and escape of vaccine-derived immunity. While current vaccines remain effective against severe disease and death, robust evidence on vaccine effectiveness (VE) against all Omicron infections (i.e. irrespective of symptoms) remains sparse. We addressed this knowledge-gap using a community-wide serosurvey with 5,310 subjects by estimating how vaccination histories modulated risk of infection in Hong Kong (which was largely infection naive) during a large wave of Omicron epidemic during January-July 2022. We estimated that Omicron infected 45% (41-48%) of the Hong Kong population. Three and four doses of BNT162b2 or CoronaVac were effective against Omicron infection (VE of 47% (95% credible interval 34-68%) and 70% (43-99%) for three and four doses of BNT162b2 respectively; VE of 31% (1-73%) and 59% (10-99%) for three and four doses of CoronaVac respectively) seven days after vaccination, but protection waned with half-lives of 15 (3-47) weeks for BNT162b2 and 5 (1-37) weeks for CoronaVac. Our findings suggest that booster vaccination can temporarily enhance population immunity ahead of anticipated waves of infections.

Resurgence of Omicron BA.2 in SARS-CoV-2 infection-naive Hong Kong (preprint)

After keeping infections at bay for two years, Hong Kong experienced a surge of Omicron BA.2 infections in early 2022 that overwhelmed the health care system, isolation facilities, and contact tracing capacity, leading to one of the highest per-capita death rates of COVID-19 in early 2022. The outbreak occurred against a backdrop of a dense population with low immunity towards natural SARS-CoV-2 infection, high vaccine hesitancy in vulnerable populations, comprehensive disease surveillance and the capacity for stringent public health and social measures. Using genome sequences and epidemiological data from this time, we reconstruct the epidemic trajectory of the BA.2 wave, estimate transmission and incidence rates, and evaluate the effectiveness of policy changes. We identify an increase in the effective reproductive rate (Re) to 9.5 in mid-January 2022, which preceded real-time estimates of transmission (Rt), revealing that BA.2 community transmission was under-ascertained weeks before the epidemic appeared to surge in mid-February 2022. Due to this, public health measures were relaxed in early February (Spring Festival) while Re increased and remained > 1 throughout February. An independent estimation of point prevalence and incidence using phylodynamics also indicates extensive superspreading at this time, which likely contributed to the rapid expansion of the epidemic. This study demonstrates that relying on Rt estimation methods dependent on case reporting can misinform epidemic response planning, sometimes with substantial consequences. There is a need for future research and implementation of improved estimates of epidemic growth in near real-time that combine multiple disparate data sources to better inform outbreak response policy.

Insights into COVID-19 epidemiology and control from temporal changes in serial interval distributions in Hong Kong (preprint)

The serial interval distribution is used to approximate the generation time distribution, an essential parameter to predict the effective reproductive number "Rt", a measure of transmissibility. However, serial interval distributions may change as an epidemic progresses rather than remaining constant. Here we show that serial intervals in Hong Kong varied over time, closely associated with the temporal variation in COVID-19 case profiles and public health and social measures that were implemented in response to surges in community transmission. Quantification of the variation over time in serial intervals led to improved estimation of Rt, and provided additional insights into the impact of public health measures on transmission of infections. One-Sentence SummaryReal-time estimates of serial interval distributions can improve assessment of COVID-19 transmission dynamics and control.

Homologous and heterologous boosting with CoronaVac and BNT162b2: a randomized trial (the Cobovax study) (preprint)

Background: There are few trials comparing homologous and heterologous third doses of COVID-19 vaccination with inactivated vaccines and mRNA vaccines. Methods: We conducted an open-label randomized trial in adults >=18 years of age who received two doses of inactivated vaccine (CoronaVac) or mRNA vaccine (BNT162b2) >=6 months earlier, randomised in 1:1 ratio to receive a third dose of either vaccine. We compared the reactogenicity, immunogenicity and cell-mediated immune responses, and assessed vaccine efficacy against infections during follow-up. Results: We enrolled 219 adults who previously received two doses of CoronaVac and randomised to CoronaVac ("CC-C", n=101) or BNT162b2 ("CC-B", n=118) third dose; and 232 adults who previously received BNT162b2 and randomised to CoronaVac ("BB-C", n=118) or BNT162b2 ("BB-B", n=114). There were more frequent reports of mild reactions in recipients of third-dose BNT162b2, which generally subsided within 7 days. Third doses significantly increased neutralising PRNT50 antibody titers against ancestral virus and Omicron BA.1 variant in all four study arms, and against Omicron BA.2 in all arms except CC-C, with statistically significant improvements for recipients of a third dose of BNT162b2 over CoronaVac irrespective of prior vaccine type. Boosting of CD4+ T cells only occurred in CoronaVac-primed arms, but we did not identify overall differences between vaccine groups in CD4+ and CD8+ T cell responses. When Omicron BA.2 was circulating, we identified 58 infections with cumulative incidence of 15.3% and 15.4% in the CC-C and CC-B (p=0.93), and 16.7% and 14.0% in the BB-C and BB-B arms, respectively (p=0.56). Conclusions: Similar levels of incidence of infection in each arm suggest all third dose combinations may provide similar degrees of protection against prevalent Omicron BA.2 infection, despite very weak antibody responses to BA.2 in the recipients of a CoronaVac third dose. Further research is warranted to identify appropriate correlates of protection for inactivated COVID-19 vaccines.

Slow waning of antibodies following a third dose of BNT162b2 in adults who had previously received two doses of inactivated vaccine (preprint)

IntroductionThird doses of COVID-19 vaccination provide an important boost to immunity, reducing the risk of symptomatic infection and the risk of severe disease. Third doses have been particularly important for improving protection against variants. However, waning of clinical protection particularly against Omicron has been noted after receipt of third doses. MethodsWe administered BNT162b2 as a third dose to adults aged [≥]30 years who had previously received two doses of inactivated vaccination. We collected blood before the third dose and again after one month and six months, and tested sera using a spike receptor binding domain IgG enzyme-linked immunosorbent assay, a surrogate virus neutralization test, and live virus plaque reduction neutralization assay against ancestral virus and Omicron BA.2. ResultsWe administered BNT162b2 as a third dose to 314 adults. We found robust antibody responses to the ancestral strain at six months after receipt of BNT162b2. Antibody responses to Omicron BA.2 were weaker after the third dose and had declined to a low level by six months. From a small number of participants we observed that natural infection or a fourth dose of vaccination generated similar antibody levels against ancestral virus, but infection generated higher antibody level against Omicron BA.2 than vaccination, suggesting a potential advantage in the breadth of antibody response from hybrid immunity. ConclusionsWhile antibody levels against the ancestral strain remained robust at six months after the third dose, antibody levels against Omicron BA.2 had fallen to low levels suggesting the potential benefits of a fourth dose.

Within-hotel transmission of SARS-CoV-2 during on-arrival quarantine in Hong Kong (preprint)

Background: On-arrival quarantine has been one of the primary measures used to prevent the introduction of COVID-19 into Hong Kong since the start of the pandemic. Most on-arrival quarantines have been done in hotels, with the duration of quarantine and testing frequency during quarantine varying throughout the pandemic for various reasons. However, hotels are not necessarily designed with infection control in mind. We aimed to study the potential risk of transmission between persons in on-arrival quarantine. Methods: We examined data on each laboratory-confirmed COVID-19 case identified in on-arrival quarantine in a hotel in Hong Kong between 1 May 2020 and 31 January 2022. We sequenced the full genomes of viruses from cases that overlapped with other confirmed cases in terms of the hotel of stay, date of arrival and date of testing positive. A combination of epidemiological information and sequence information was then used to identify probable transmission events. Findings: Among 221 imported cases that overlapped with other quarantined cases, phylogenetic analysis identified eight suspected clusters comprising 20 cases in total. Only three of these clusters had been recognised as hotel transmission events. Interpretation: We have identified potential occurrences of COVID-19 transmission within hotel quarantine in Hong Kong demonstrating the underlying low but non-zero risk associated with sequestering arrivals within hotels. In future pandemics, on-arrival quarantine in hotels could be used to delay the introduction of infection, but the construction of purpose-built facilities for on-arrival quarantine might be necessary to minimize importation risk.

Real-world effectiveness of molnupiravir and nirmatrelvir/ritonavir among COVID-19 inpatients during Hong Kong's Omicron BA.2 wave: an observational study (preprint)

Background Real-world evidence on the effectiveness of oral antivirals in mild-to-moderate COVID-19 patients is urgently needed. This retrospective cohort study aims to evaluate the clinical and virologic outcomes associated with molnupiravir and nirmatrelvir/ritonavir use in COVID-19 patients during a pandemic wave dominated by the Omicron BA.2 variant. Methods We analyzed data from a territory-wide retrospective cohort of hospitalized patients with confirmed diagnosis of SARS-CoV-2 infection from 26th February 2022 to 26th April 2022 in Hong Kong. Oral antiviral users were matched with controls using propensity-score matching in a ratio of 1:4. Study outcomes were a composite outcome of disease progression (all-cause mortality, initiation of invasive mechanical ventilation [IMV], or intensive care unit admission) and their individual outcomes, and lower viral load of cycle threshold (Ct) value [≥]30 cycles. Hazard ratios (HR) of event outcomes were estimated using Cox regression models. Results Among 40,776 hospitalized patients with SARS-CoV-2 infection over a mean follow-up of 41.3 days with 925,713 person-days, 2,359 and 1,000 patients not initially requiring oxygen therapy were initiated with molnupiravir and nirmatrelvir/ritonavir, respectively. The crude incidence rates of all-cause mortality and IMV were 22.24 and 1.06 events per 10,000 person-days among molnupiravir users, 11.04 and 1.75 events per 10,000 person-days among nirmatrelvir/ritonavir users. Oral antiviral use was associated with a significantly lower risk of the composite outcome of disease progression (molnupiravir: HR=0.53, 95%CI=0.46-0.62, p<0.001; nirmatrelvir/ritonavir: HR=0.33, 95%CI=0.24-0.46, p<0.001) than non-use, which was consistently observed for all-cause mortality (molnupiravir: HR=0.55, 95%CI=0.47-0.63, p<0.001; nirmatrelvir/ritonavir: HR=0.32, 95%CI=0.23-0.45, p<0.001). Molnupiravir users had lower risks of IMV (HR=0.31, 95%CI=0.16-0.61, p<0.001). Time to achieving lower viral load was significantly shorter among oral antiviral users than matched controls (molnupiravir: HR=1.21, 95%CI=1.07-1.37, p=0.002; nirmatrelvir/ritonavir: HR=1.25, 95%CI=1.04-1.50, p=0.015). Amongst survivors, nirmatrelvir/ritonavir had shorter length of hospital stay (-0.70 days, 95%CI=-1.37 to -0.04, p=0.039) than matched controls. Head-to-head comparison of molnupiravir and nirmatrelvir/ritonavir reported higher risk of mortality (HR=1.53 95%CI=1.01-2.31, p=0.047) and longer length of hospital stay (0.83 days, 95%CI=0.07-1.58, p=0.032) for molnupiravir users. Conclusions Against Omicron BA.2, initiation of novel oral antiviral treatment in hospitalized patients not requiring any oxygen therapy was associated with lower risks of disease progression and all-cause mortality, in addition to achieving low viral load faster.

Vaccine effectiveness of two and three doses of BNT162b2 and CoronaVac against COVID-19 in Hong Kong (preprint)

Background: Hong Kong maintained extremely low circulation of SARS-CoV-2 until a major community epidemic of Omicron BA.2 starting in January 2022. Both mRNA BNT162b2 (BioNTech/Fosun Pharma) and inactivated CoronaVac (Sinovac) vaccines are widely available, however coverage has remained low in older adults. Vaccine effectiveness in this predominantly infection-naive population is unknown. Methods: We used individual-level case data on mild/moderate, severe/fatal and fatal hospitalized COVID-19 from December 31, 2021 to March 8, 2022, along with census information and coverage data of BNT162b2 and CoronaVac. We used a negative binomial model, adjusting for age and calendar day to estimate vaccine effectiveness of one, two and three dose schedules of both vaccines, and relative effectiveness by number of doses and vaccine type. Findings: A total of 12.7 million vaccine doses were administered in the 7.3 million population of Hong Kong, and we analyzed data from confirmed cases with mild/moderate (N=5,474), severe/fatal (N=5,294) and fatal (N=4,093) COVID-19. Two doses of either vaccine protected against severe disease and death, with higher effectiveness among adults [≥]60 years with BNT162b2 (VE: 88.2%, 95% confidence interval, CI: 84.4%, 91.1%) compared to CoronaVac (VE: 74.1%, 95% CI: 67.8%, 79.2%). Three doses of either vaccine offered very high levels of protection against severe outcomes (VE: 98.1%, 95% CI: 97.1%, 98.8%). Interpretation: Third doses of either BNT162b2 or CoronaVac provide substantial additional protection against severe COVID-19 and should be prioritized, particularly in older adults who received CoronaVac primary schedules. Longer follow-up is needed to assess persistence of different vaccine platforms and schedules.

Strength and durability of antibody responses to BNT162b2 and CoronaVac (preprint)

We studied 2864 adults in Hong Kong who received CoronaVac inactivated virus vaccine (Sinovac) and BNT162b2 mRNA vaccine (Comirnaty, BioNTech/Fosun Pharma). We found stronger and more durable antibody responses to two doses of the mRNA vaccine, and slightly stronger initial antibody responses to each vaccine in younger adults and women.

BNT162b2 vaccine boosts neutralizing antibodies to ancestral SARS-CoV-2 & Omicron variant in adults who previously received two doses of inactivated vaccine (preprint)

BackgroundLimited data exist on antibody responses to mixed vaccination strategies involving inactivated COVID-19 vaccines, particularly in the context of emerging variants. MethodsWe conducted an open label trial of a third vaccine dose of an mRNA vaccine (BNT162b2, Fosun Pharma/BioNTech) in adults aged [≥]30 years who had previously received two doses of inactivated COVID-19 vaccine. We collected blood samples before administering the third dose and 28 days later, and tested for antibodies to the ancestral virus using a binding assay (ELISA), a surrogate virus neutralization test (sVNT) and a live virus plaque reduction neutralization test (PRNT). We also tested for antibodies against the Omicron variant using live-virus PRNT. ResultsIn 315 participants, a third dose of BNT162b2 substantially increased antibody titers on each assay. Mean ELISA levels increased from an optical density (OD) of 0.3 to 2.2 (p<0. 001), and mean sVNT levels increased from an inhibition of 17% to 96% (p<0.001). In a random subset of 20 participants, the geometric mean PRNT50 titers rose very substantially by at least 24 fold from Day 0 to Day 28 against the ancestral virus (p<0.001) and rose by at least 11 fold against the Omicron variant (p<0.001). In daily monitoring, post-vaccination reactions subsided within 7 days for over 99% of participants. ConclusionsA third dose of COVID-19 vaccination with an mRNA vaccine substantially improved antibody levels against the ancestral virus and the Omicron variant with well-tolerated safety profile, in adults who had received two doses of inactivated vaccine 6 months earlier. SummaryIn this open label trial of Chinese adults aged [≥]30 years who received two doses of inactivated COVID-19 vaccine 6 months earlier, third-dose mRNA vaccine substantially improved antibody levels against the ancestral virus and Omicron variant with well-tolerated safety profile.

SARS-CoV-2 under an elimination strategy in Hong Kong (preprint)

Hong Kong utilized an elimination strategy with intermittent use of public health and social measures and increasingly stringent travel regulations to control SARS-CoV-2 transmission. By analyzing >1700 genome sequences representing 17% of confirmed cases from 23-January-2020 to 26-January-2021, we reveal the effects of fluctuating control measures on the evolution and epidemiology of SARS-CoV-2 lineages in Hong Kong. Despite numerous importations, only three introductions were responsible for 90% of locally-acquired cases, two of which circulated cryptically for weeks while less stringent measures were in place. We found that SARS-CoV-2 within-host diversity was most similar among transmission pairs and epidemiological clusters due to a strong transmission bottleneck through which similar genetic background generates similar within-host diversity.

Pandemic fatigue impedes mitigation of COVID-19 in Hong Kong (preprint)

Community-wide social distancing has been a cornerstone of pandemic control prior to mass vaccinations. The extent to which pandemic fatigue is undermining adherence to such measures and accelerating transmission remains unclear. Using large-scale weekly telephone surveys and mobility data, we characterize the evolution of risk perception and protective behaviours in Hong Kong. We estimate a 1.5% to 5.5% reduction in population compliance with protective policies for the fourth wave (October 2020 to January 2021) versus the third wave (July to August 2020), inducing prolonged disease circulation with increased infections. Mathematical models incorporating population protective behaviours estimates that the fourth wave would have been 14% smaller if not for pandemic fatigue. Mitigating pandemic fatigue is essential in maintaining population protective behaviours for controlling COVID-19.

Universal community nucleic acid testing for COVID-19 in Hong Kong reveals insights into transmission dynamics (preprint)

Background: Testing of an entire community has been used as an approach to control COVID-19. In Hong Kong, a universal community testing programme (UCTP) was implemented at the fadeout phase of a community epidemic in July to September 2020, to determine the prevalence of unrecognised cases and limit any remaining transmission chains. We described the utility of the UCTP in finding unrecognised cases, and analysed data from the UCTP and other sources to characterise transmission dynamics.Methods: We described the characteristics of people participating in the UCTP, and compared the clinical and epidemiological characteristics of COVID-19 cases detected by the UCTP versus those detected by clinical diagnosis and public health surveillance. We developed a Bayesian model to estimate the age-specific incidence of infection and the proportion of cases detected by clinical diagnosis and public health surveillance.Findings: 1.77 million people, 24% of the Hong Kong population, participated in the UCTP from 1 to 14 September 2020. The UCTP identified 32 new infections (1.8 per 100,000 samples tested), consisting of 29% of all local cases reported during the two-week UCTP period. Compared with the existing clinical diagnosis and public health surveillance, the UCTP detected a higher proportion of sporadic cases (62% versus 27%, p <0.01) and identified 6 (out of 18) additional transmission chains during that period. We estimated that 27% (95% credible interval: 22%, 34%) of all infections were detected by the existing clinical diagnosis and public health surveillance in the third wave.Interpretation: We reported empirical evidence of the utility of population-wide COVID-19 testing in detecting unrecognised infections and transmission chains. Around three quarters of infections have not been identified through existing surveillance approaches including contact tracing.

Early empirical assessment of the N501Y mutant strains of SARS-CoV-2 in the United Kingdom, October to November 2020 (preprint)

Two new SARS-CoV-2 lineages with the N501Y mutation in the receptor binding domain of the spike protein have rapidly become prevalent in the UK. We estimated that the earlier 501Y lineage without amino acid deletion {Delta}69/{Delta}70 circulating mainly between early September to mid-November was 10% (6-13%) more transmissible than the 501N lineage, and the currently dominant 501Y lineage with amino acid deletion {Delta}69/{Delta}70 circulating since late September was 75% (70-80%) more transmissible than the 501N lineage.

Real-time tracking and prediction of COVID-19 infection using digital proxies of population mobility and mixing (preprint)

Digital proxies of human mobility and physical mixing have been used to monitor viral transmissibility and effectiveness of social distancing interventions in the ongoing COVID-19 pandemic. We developed a new framework that parameterizes disease transmission models with age-specific digital mobility data. By fitting the model to case data in Hong Kong, we were able to accurately track the local effective reproduction number of COVID-19 in near real time (i.e. no longer constrained by the delay of around 9 days between infection and reporting of cases) which is essential for quick assessment of the effectiveness of interventions on reducing transmissibility. Our findings showed that accurate nowcast and forecast of COVID-19 epidemics can be obtained by integrating valid digital proxies of physical mixing into conventional epidemic models.

Empirical transmission advantage of the D614G mutant strain of SARS-CoV-2 (preprint)

The SARS-CoV-2 lineage carrying the amino acid change D614G has become the dominant variant in the global COVID-19 pandemic. The rapid spread of the G614 mutant suggests that it may have a transmission advantage over the D614 wildtype. Using our previous epidemiological framework to analyze COVID-19 surveillance and sequence data, we estimated that the G614 mutant is 31% (28-34%) more transmissible than the D614 wildtype. As such, interventions that were previously effective in containing or mitigating the D614 wildtype (e.g. in China, Vietnam, Thailand, etc.) might be less effective against the G614 mutant. Our framework can be readily integrated into current COVID-19 surveillance to monitor the emergence and fitness of mutant strains, such that pandemic surveillance, disease control and development of treatment and vaccines can be adjusted dynamically.

A global assessment of the impact of school closure in reducing COVID-19 spread (preprint)

Prolonged school closure has been adopted worldwide to control COVID-19. Such preemptive implementation was predicated on the premise that school children are a core group for COVID-19 transmission. Using surveillance data from the Chinese cities of Shenzhen and Anqing, we inferred that children aged 18 or below are only around half as susceptible to COVID-19 infection as adults. Using transmission models parameterized with synthetic contact matrices for 152 jurisdictions around the world, we showed that the lower susceptibility of school children substantially limited the effectiveness of school closure in reducing COVID-19 transmissibility. Our results, together with recent findings that clinical severity of COVID-19 in children is lower, suggest that school closure may not be ideal as a sustained, primary intervention for controlling COVID-19.