Search

Agent-based models can be used to better understand the impacts of lifting restrictions or implementing interventions during a pandemic. However, agent-based models are computationally expensive, and running a model of a large population can result in a simulation taking too long to run for the model to be a useful analysis tool during a public health crisis. To reduce computing time and power while running a detailed agent-based model for the spread of COVID-19 in the Republic of Ireland, we introduce a scaling factor that equates 1 agent to 100 people in the population. We present the results from model validation and show that the scaling factor increases the variability in the model output, but the average model results are similar in scaled and un-scaled models of the same population, and the scaled model is able to accurately simulate the number of cases per day in Ireland during the autumn of 2020. We then test the usability of the model by using the model to explore the likely impacts of increasing community mixing when schools reopen after summer holidays. [ABSTRACT FROM AUTHOR]

Background: In order to be prepared for an infectious disease outbreak it is important to know what interventions will or will not have an impact on reducing the outbreak. While some interventions might have a greater effect in mitigating an outbreak, others might only have a minor effect but all interventions will have a cost in implementation. Estimating the effectiveness of an intervention can be done using computational modelling. In particular, comparing the results of model runs with an intervention in place to control runs where no interventions were used can help to determine what interventions will have the greatest effect on an outbreak.Methods: To test the effects of a school closure policy on the spread of an infectious disease (in this case measles) we run simulations closing schools based on either the proximity of the town to the initial outbreak or the centrality of the town within the network of towns in the simulation. To do this we use a hybrid model that combines an agent-based model with an equation-based model. In our analysis, we use three measures to compare the effects of different intervention strategies: the total number of model runs leading to an outbreak, the total number of infected agents, and the geographic spread of outbreaks.Results: Our results show that closing down the schools in the town where an outbreak begins and the town with the highest in degree centrality provides the largest reduction in percent of runs leading to an outbreak as well as a reduction in the geographic spread of the outbreak compared to only closing down the town where the outbreak begins. Although closing down schools in the town with the closest proximity to the town where the outbreak begins also provides a reduction in the chance of an outbreak, we do not find the reduction to be as large as when the schools in the high in degree centrality town are closed.Conclusions: Thus we believe that focusing on high in degree centrality towns during an outbreak is important in reducing the overall size of an outbreak. [ABSTRACT FROM AUTHOR]

Background: Escherichia coli (E. coli) comprise part of the normal vaginal microflora. Transfer from mother to neonate can occur during delivery resulting, sometimes, in neonatal bacterial disease. Here, we aim to report the first outbreak of CTX-M ESBL-producing E. coli with evidence of mother-to-neonate transmission in an Irish neonatal intensive care unit (NICU) followed by patient-to-patient transmission.Methods: Investigation including molecular typing was conducted. Infection was defined by clinical and laboratory criteria and requirement for antimicrobial therapy with or without positive blood cultures. Colonisation was determined by isolation without relevant symptoms or indicators of infection.Results: Index case was an 8-day-old baby born at 34 weeks gestation who developed ESBL-producing E. coli infections at multiple body sites. Screening confirmed their mother as colonised with ESBL-producing E. coli. Five other neonates, in the NICU simultaneously with the index case, also tested positive. Of these, four were colonised while one neonate developed sepsis, requiring antimicrobial therapy. The second infected neonate's mother was also colonised by ESBL-producing E. coli. Isolates from all eight positive patients (6 neonates, 2 mothers) were compared using pulsed-field gel electrophoresis (PFGE). Two distinct ESBL-producing strains were implicated, with evidence of transmission between mothers and neonates for both strains. All isolates were confirmed as CTX-M ESBL-producers. There were no deaths associated with the outbreak.Conclusions: Resources were directed towards control interventions focused on hand hygiene and antimicrobial stewardship, which ultimately proved successful. Since this incident, all neonates admitted to the NICU have been screened for ESBL-producers and expectant mothers are screened at their first antenatal appointment. To date, there have been no further outbreaks. [ABSTRACT FROM AUTHOR]

The dynamics that lead to the spread of an infectious disease through a population can be characterized as a complex system. One way to model such a system, in order to improve preparedness, and learn more about how an infectious disease, such as COVID-19, might spread through a population, is agent-based epidemiological modelling. When a pandemic is caused by an emerging disease, it takes time to develop a completely new model that captures the complexity of the system. In this paper, we discuss adapting an existing agent-based model for the spread of measles in Ireland to simulate the spread of COVID-19. The model already captures the population structure and commuting patterns of the Irish population, and therefore, once adapted to COVID-19, it can provide important insight on the pandemic, specifically in Ireland. We first investigate the different disease parameters that need to be adjusted to simulate the spread of COVID-19 instead of measles and then run a set of experiments initially comparing the model output for our original measles model with that from the adjusted COVID-19 model. We then report on experiments on how the different values of the basic reproductive number, R 0 , influence the simulated outbreaks, and find that our model behaves as expected: the higher the R 0 , the more agents are infected. Then, we demonstrate how different intervention strategies, such as vaccinations and school closures, influence the spread of measles and COVID-19 and how we can simulate real pandemic timings and interventions in our model. We show that with the same society, environment and transportation components among the different disease components lead to very different results for the two diseases, and that our COVID-19 model, when run for Leitrim County, Ireland, predicts a similar outbreak length to a real outbreak in Leitrim County, Ireland, but the model results in a higher number of infected agents compared to the real outbreak. This difference in cases is most likely due to identifying all cases of COVID-19 in the model opposed to only those tested. Once an agent-based model is created to simulate a specific complex system or society, the disease component can be adapted to simulate different infectious disease outbreaks. This makes agent-based models a powerful tool that can be used to help understand the spread of new and emerging infectious diseases. [ABSTRACT FROM AUTHOR]

Agent-based models are a tool that can be used to better understand the dynamics of an infectious disease outbreak. An infectious disease outbreak is influenced by many factors including vaccination or immunity levels, population density, and the age structure of the population. We hypothesize that these factors along with interactions of factors and the actions of individuals would lead to outbreaks of different size and severity even in two towns that appear similar on paper. Thus, it is necessary to implement a model that is able to capture these interactions and the actions of individuals. Using openly available data we create a data-driven agent-based model to simulate the spread of an airborne infectious disease in an Irish town. Agent-based models have been known to produce results that include the emergence of patterns and behaviours that are not directly programmed into the model. Our model is tested by simulating an outbreak of measles that occurred in Schull, Ireland in 2012. We simulate the same outbreak in 33 different towns and look at the correlations between the model results and the town characteristics (population, area, vaccination rates, age structure) to determine if the results of the model are affected by interactions of those town characteristics and the decisions on the agents in the model. As expected our results show that the outbreaks are not strongly correlated with any of the main characteristics of the towns and thus the model is most likely capturing such interactions and the agent-based model is successful in capturing the differences in the outbreaks., Competing Interests: The authors have declared that no competing interests exist.