Mass gatherings pose significant challenges to public health by facilitating the transmission of infectious diseases. The COVID-19 pandemic prompted governments worldwide to suspend gatherings to reduce social contact and virus transmission. However, limited research has investigated the contribution of mass gatherings to local COVID-19 transmission. In Hong Kong, the relaxation of group gathering restrictions in mid-2020 coincided with large-scale demonstrations against the National Security Law, raising concerns about the safety of mass gatherings during the pandemic. This study addresses this gap by examining the impacts of mass gatherings on the local transmission of COVID-19 in Hong Kong and evaluating the effectiveness of social distancing policies, specifically the Prohibition On Group Gathering Regulation (POGGR). The study uses Hong Kong as a case study due to the unique confluence of relaxed restrictions and large public demonstrations, providing a valuable opportunity to assess the impact of mass gatherings on COVID-19 transmission. Understanding this impact is crucial for informing future pandemic control strategies and adapting to a ‘new normal’ where restrictions may be relaxed.

Existing literature extensively documents the association between mass gatherings and increased transmission of various respiratory infectious diseases like influenza and measles. The WHO has highlighted the amplifying effect of mass gatherings on virus spread and their potential to hinder a country's pandemic response. While the role of smaller, indoor gatherings in COVID-19 transmission is relatively well understood, empirical evidence on the impact of large-scale public gatherings on local community transmission remains limited. Some studies have linked specific mass gathering events to COVID-19 outbreaks, such as the Sri Petaling Mosque event in Malaysia and a traditional festival in Borriana, Spain. Conversely, other research suggests that large public gatherings like the Black Lives Matter protests might have had an offsetting effect due to increased stay-at-home behavior by non-participants. This lack of conclusive empirical evidence necessitates further investigation to comprehensively assess the impact of large-scale gatherings on COVID-19 transmission in local communities.

This study employs a quantitative approach using the Synthetic Control Method (SCM) to assess the impact of mass gatherings on COVID-19 transmission in Hong Kong. SCM creates a counterfactual 'synthetic Hong Kong' by combining data from over 200 Chinese cities to simulate what the COVID-19 infection pattern might have looked like without the mass gatherings. This approach addresses the challenge of finding a suitable control city by systematically selecting a combination of comparison units that best represent Hong Kong's pre-intervention characteristics. The dataset integrates epidemiological data, city-level meteorological data, and socioeconomic data to control for confounding factors influencing COVID-19 transmission. The study uses daily new and total COVID-19 case numbers as the outcome variables. Predictor variables include demographic characteristics (population, population density), socioeconomic factors (GDP, per capita GDP, healthcare capacity), and meteorological factors (temperature, wind speed, humidity, air quality index). The SCM model creates a weighted average of the selected donor cities to simulate a synthetic Hong Kong, minimizing differences with actual Hong Kong before the intervention. By comparing the actual epidemiological trajectory of Hong Kong with the synthetic control's trajectory, the study estimates the impact of mass gatherings. To account for the incubation period and testing lags, the study considers various intervention points, examining the impact over both 10-day and 30-day periods after the commencement of the mass gatherings.

The SCM analysis, with June 20th, 2020 (following the relaxation of POGGR), as the intervention point, reveals a substantially higher infection rate in Hong Kong (2.63% daily growth rate of total cases) compared to the synthetic control (0.07%). Specifically, the study estimates that mass gatherings increased the number of new infections by 62 cases (87.58% of total new cases) over the first 10 days following the intervention and by 737 cases (97.23%) over the following 30 days. Robustness tests using different intervention points (June 21st-30th) yielded consistent results, albeit with some variation in the magnitude of the impact depending on the lag considered. Shorter lags (1-5 days) suggest an average increase of 57 new infections (75% of the total) over 10 days and 961 infections (97.66%) over 30 days. Longer lags (6-10 days) showed even more significant increases: 101 new cases (91.82%) and 1553 cases (99.36%) over 10 and 30 days, respectively. These findings strongly suggest a causal link between the mass gatherings and the substantial increase in COVID-19 cases during Hong Kong's third wave.

The findings strongly support the hypothesis that the relaxation of social distancing policies and the subsequent mass gatherings, particularly the demonstrations, contributed significantly to the resurgence of COVID-19 cases in Hong Kong. The significant difference in infection rates between Hong Kong and the synthetic control, consistently observed across different intervention points and lag periods, provides compelling evidence for the impact of mass gatherings. These results complement existing research focused on smaller, indoor gatherings by highlighting the risks associated with large-scale outdoor events. The study's findings have important implications for pandemic control strategies. They underscore the crucial role of stringent social distancing measures, including restrictions on mass gatherings, in limiting COVID-19 transmission, particularly during early phases of an outbreak when herd immunity is low. The Hong Kong experience demonstrates the potential for rapid resurgence of infections when restrictions are relaxed, even if precautions such as mask-wearing are observed. The study’s findings highlight the need for proactive risk assessment and timely adjustments to social distancing policies based on the evolving situation.

This study provides robust empirical evidence demonstrating the substantial impact of mass gatherings on COVID-19 transmission in Hong Kong during a period of relaxed social distancing restrictions. The findings strongly support the implementation of strict social distancing policies, particularly those limiting mass gatherings, to effectively control outbreaks. The methodology used—the synthetic control method—offers a powerful tool for evaluating the impact of policy interventions in complex situations. Future research could extend this methodology to other contexts and infectious diseases, exploring variations in transmission risks based on event type, location, and other contextual factors. Further investigation into the interplay between public health policies, social dynamics, and the response to large public events could inform improved pandemic preparedness strategies.

The study acknowledges limitations arising from potential disparities in the social and political environments between Hong Kong and the mainland Chinese cities used to construct the synthetic control. While the SCM model controls for observable socioeconomic and meteorological factors, unobservable time-invariant differences might affect the estimation. The study addresses this by using an extended pre-intervention period and carefully selecting comparison units. Furthermore, the study focuses on the initial COVID-19 variant, and the findings might not fully reflect the transmission dynamics of later, more transmissible variants or situations with substantial population immunity. Finally, compliance with government policies and public trust in government agencies, factors influencing behavior during mass gatherings, are not explicitly modeled and could have affected the results.