A theoretical epidemiological investigation into the transmission of respiratory infectious diseases during group meals among military personnel based on an individual-based model

Abstract

Introduction: Within military settings, soldiers are arranged to eat together in a self-service manner for every meal. The process of food selection and consumption often leads to close contact amongst individuals, heightening the risk of respiratory infectious disease transmission. To comprehend the transmission dynamics during communal dining, we have conducted an in-depth epidemiological investigation.

Methods: The dining process was divided into two phases: lining up for food and dining at designated seats. Soldiers were randomly split into two queues and entered the food selection area from the same side. The movements of the soldiers dynamically altered both the queues and the contact duration and distance between susceptible individuals and infection sources. We utilized a random computer model using MATLAB software, with the individual as the unit of study, for simulating the food selection process. This model quantitatively analyzed the dynamic process of disease transmission within the queues due to the dispersion of small pathogen-carrying particles.

Results: Our findings indicate that close interactions between individuals during picking up food can result in the persistent transmission of airborne infectious diseases. Implementing measures such as discontinuing buffet-style meals, implementing staggered dining schedules, and mandating mask-wearing during food collection could help control disease transmission during an epidemic.

Discussion: This study demonstrates that the individual-based model can simulate the dynamic process of disease transmission through complex behavioral activities and is more suitable for conducting research on the dynamics of infectious diseases in small-scale populations. Since this is a simulation conducted in a virtual scenario, the results of the model still need to be verified through field investigations. Nevertheless, once robust outbreak investigation studies have yielded reliable model parameters, these parameters can be adapted to this and other similar situations to demonstrate the potential for transmission.

Keywords: buffet; individual-based model; military; queues; respiratory infectious diseases.

Figure 1

(a) Illustration depicting the locations of the two food-collection queues within the pick-up zone and the pathogen-infected area. (b) Diagram demonstrating the transition patterns of population infection status. S represents susceptible individuals; E_d and E_q denote latent and non-infectious infected individuals that emerge during food collection and dining, respectively; I_m represents cases with mild symptoms during the infectious period; I_h denotes infectious patients with severe symptoms needing time off for treatment or rest; and R signifies recoveries.

Figures 2

(a–c) The functional relationship between the contact infection probability q, and the factors of contact duration t and distance d. (d–f) This refers to the time distribution of individuals who are picking up food, waiting for food pickup, and those who have completed food pickup, respectively, in the two queues. (g–i) The time distribution of Rt during the epidemic period.

Figure 3

The fundamental structure of the model’s program design during each primary cycle. Every day comprises three meals: breakfast, lunch, and dinner. To prevent repetition, we only present the plan for one meal here. For detailed procedures, please refer to Appendix I.

Figure 4

Temporal distribution of the number of infected individuals and cases when the exposure index varies. Rows 1–4, respectively, display the temporal distributions of the number of infections occurring during food pickup and dining, the number of infections occurring specifically during food pickup, the number of cases, and the number of hospitalized patients. The solid blue line represents the median number of new infections or cases, measured by the left vertical axis; the dashed line represents the median cumulative number of infections or cases, measured by the right vertical axis; the blue shaded area indicates the 25–75% fluctuation interval after 50 simulations.

Figure 5

The time distribution of the current number of susceptible, latent infected, infectious and hospitalized individuals.

Figure 6

Results of sensitivity analyses. λ denotes the exposure index, p_h signifies the rate of severe cases, d_m depicts the infectious period for mild cases (from the onset of illness to recovery), d_h indicates the infectious period for severe cases (from onset to treatment), and p represents the ratio of susceptible individuals to the total population when t equals 0.