School closure, a necessary break for both teachers and students, is also associated with mitigating influenza outbreaks. Purposely closing schools during outbreaks of influenza is not officially recommended by the World Health Organisation; however, this non-pharmaceutical intervention is well regarded as an approach in reducing the epidemic impact. A well-known example of this is the closure of schools during the swine flu epidemic in the UK, 2009.
De Luca et al’s study considers the effect of regulated school holidays in interrupting influenza propagation by offering variety in social mixing and travel – two key determinants of influenza epidemic potency. The authors aimed to integrate social mixing and travel of children and adults during school/holiday periods into a spatial metapopulation that also utilised data based on demographics and infection dynamics. Focusing on the 2008/2009 influenza season in Belgium, the mathematical model was calibrated to the singular district of Brussels which showed accurate scalability to a national level.
The authors observed how the growth of the number of new infections significantly slowed during the Christmas holidays whereas the fall and Easter holidays had little effect.
This model was further parameterized with influenza-like-illness (ILI) data reported by the Belgian Scientific Institute of Public Health. Once the Belgian school calendar was applied, the authors tested a number of experimental scenarios. Different scheduled holidays were first removed and replaced with regular school/working weeks from the model to observe the effect on an epidemic’s timing and intensity.
The authors observed how the growth of the number of new infections significantly slowed during the Christmas holidays whereas the fall and Easter holidays had little effect. The benefit of weekends in periodically slowing down the epidemic curve was also observed. Altering the contact and behaviour among groups of people over weekends and holidays is thought to alter epidemic dynamics, causing a smaller epidemic with less incidences. Altered mixing behaviour during weekends and holidays was found more influential on the spread of an epidemic than travel was in this study.
Furthering the observation that Christmas holidays can play a large role on epidemic impact, extension methods were tested. Adding a week of holiday before Christmas had no impact on the epidemics timing or severity, however, whilst extending the Christmas holidays could not delay the epidemic, it could reduce the severity as the number of ILIs decreased. Additional breaks beyond Christmas may also prove to have a mitigating impact, but the whole calendar must be considered as holidays throughout the year all affect the epidemic’s evolution differently.
The analysis on a single season represents how the impact of regular school closures slow and mitigate influenza epidemics. However, the epidemic impact of school closure is heavily dependent on the closure timing and influenza season. The authors also noted that contacts and their changes throughout the calendar are likely to be country-specific. An improved understanding on the behavioural changes caused by an epidemic are required to further parameterize models for epidemic management which may be achieved by incorporating additional seasons and more specific age groups.
Understanding the impact that school closures can have on influenza may help manage future epidemics and attenuate pressure placed on public health services. These findings may be of relevant use to authorities considering reducing summer holidays to redistribute spare holidays throughout the year to manage the influenza season. The results from this study have helped shine light on previous studies performed throughout the world; therefore, it would be worthwhile to apply similar models in different countries.