Combinational Recommendation of Vaccinations, Mask-Wearing, and Home-Quarantine to Control Influenza in Megacities: An Agent-Based Modeling Study With Large-Scale Trajectory Data

The outbreak of COVID-19 stimulated a new round of discussion on how to deal with respiratory infectious diseases. Influenza viruses have led to several pandemics worldwide. The spatiotemporal characteristics of influenza transmission in modern cities, especially megacities, are not well-known, which increases the difficulty of influenza prevention and control for populous urban areas. For a long time, influenza prevention and control measures have focused on vaccination of the elderly and children, and school closure. Since the outbreak of COVID-19, the public's awareness of measures such as vaccinations, mask-wearing, and home-quarantine has generally increased in some regions of the world. To control the influenza epidemic and reduce the proportion of infected people with high mortality, the combination of these three measures needs quantitative evaluation based on the spatiotemporal transmission characteristics of influenza in megacities. Given that the agent-based model with both demographic attributes and fine-grained mobility is a key planning tool in deploying intervention strategies, this study proposes a spatially explicit agent-based influenza model for assessing and recommending the combinations of influenza control measures. This study considers Shenzhen city, China as the research area. First, a spatially explicit agent-based influenza transmission model was developed by integrating large-scale individual trajectory data and human response behavior. Then, the model was evaluated across multiple intra-urban spatial scales based on confirmed influenza cases. Finally, the model was used to evaluate the combined effects of the three interventions (V: vaccinations, M: mask-wearing, and Q: home-quarantining) under different compliance rates, and their optimal combinations for given control objectives were recommended. This study reveals that adults were a high-risk population with a low reporting rate, and children formed the lowest infected proportion and had the highest reporting rate in Shenzhen. In addition, this study systematically recommended different combinations of vaccinations, mask-wearing, and home-quarantine with different compliance rates for different control objectives to deal with the influenza epidemic. For example, the “V45%-M60%-Q20%” strategy can maintain the infection percentage below 5%, while the “V20%-M60%-Q20%” strategy can maintain the infection percentage below 15%. The model and policy recommendations from this study provide a tool and intervention reference for influenza epidemic management in the post-COVID-19 era.

The travel chains of individuals with mobile phones are constructed according to the mobile phone location data, and the travel chains of individuals without mobile phones are constructed according to the travel survey data. After cleaning the mobile phone location data, 5.8 million users with complete 24-hour records during the day were selected. For each mobile phone user, their home location and work location are identified based on the frequent places they stay during the day and at night. Then, taking the home address of mobile phone users as the matching object, a 24-hour activity chain is synthesized for each mobile phone user and given a workplace. The travel survey data includes individual travel records, home addresses, and workplaces. We match the home addresses of the synthesized individuals without mobile phones with the survey data and then synthesize their travel chains. Students' schools are allocated according to their family address.

Contact network
Taking the independent individuals in urban areas as agents, the agent-based mobility model we constructed has 11.16 million synthesized individuals (4.5 million households) and 230,000 work units in Shenzhen. Population age structure and household size in the model are shown in Fig. S2A and Fig. S2B, respectively. The activity chain of an agent is divided into 24 hours with an activity purpose and a series of active locations, and a typical activity chain is expressed in the form of "home-work-leisure-home". An individual has an activity location every hour, which is represented by building coordinates.
Based on the agent travel chain, with an hour as the time step, a contact network of 24 time series in a day is dynamically constructed. This study set individuals with the same activities at the same time and in the same building location as individuals with spatiotemporal co-occurrence. Individuals with spatiotemporal co-occurrence at home, work, and school are divided into multiple fixed contact groups. The individuals in the fixed contact group are regular encounters, and the other individuals outside the fixed contact group are random encounters, so as to simulate close contacts for acquaintances and casual contacts for strangers (Table S1). The comparison between the contact numbers of different ages in the model and the survey data of Shanghai (1) is shown in the Fig. S2C.

Effectiveness of three interventions
This study analyzed the effects of three interventions on adults (Fig. S3). Different vaccination rates for adults led to a gradual decline in the infection scale, with vaccination for 50% of adults reducing the percentage of infected individuals to 19.64% (Fig. S3A). During the influenza season, 70% of infected adults wear masks for all activities except living at home, reducing the percentage of infected people to 20.18% (Fig. S3B). The decreasing rate of infection size showed a linear trend with the increase in vaccine coverage rate (Fig. S3D), but the decreasing rate of infection size showed a non-linear trend with the increase in mask wearing rate (Fig. S3E). By improving the home-quarantine proportion of infected adults after onset (Fig. S3C), it is found that the decline rate of infection size shows a linear trend (Fig. S3F). The higher the home-quarantine proportion, the higher the decline rate of infection size (except when all infected adults are quarantined at home after onset).

Sensitivity to immune ratios
In order to understand the impact of the immunization ratio on the temporal and spatial diffusion characteristics of influenza, the model simulates two scenarios: a population immunization ratio of 0 (comparison scenario) and a population immunization ratio of 30% (baseline scenario). The two scenarios are simulated 100 times each, and the median value of the 100 simulation results is taken as the final simulation result. In this study, the effects of the initial immune ratio on the simulation results were analyzed from the aspects of effective reproduction number, cumulative infection size, peak time, infection location, age distribution of infected persons, and the infection size in each spatial unit at different spatial scales.
In the comparison scenario, 61.94% of the population in Shenzhen will be infected (Fig. S4A), and the peak time of the daily confirmed curve is 28 days ahead of the baseline scenario (Fig. S4B). The infection ratios of different age groups and the infection ratios of different activity types are basically consistent with the baseline scenario (Fig. S4C). At the scale of district, sub-district, and community, the infection sizes of the two scenarios have a strong linear relationship (Fig. S4D-F). The effective reproduction number in the baseline scenario is 1.59, while in the comparison scenario it is 2.39, which is greater than 2.0. It can be seen that different immune ratios have an impact on the infection size, peak time, and effective reproduction number, and they have a weak impact on the relative transmission intensity in each spatial unit, the infection ratio of different age groups, and the infection ratio of different activity locations in Shenzhen.