Forty days after WHO called COVID-19 a Public Health Emergency of International Concern, interest in school closures is starting to emerge. On 6th March, the New York Times published an opinion piece calling for schools to be shut down before the outbreak becomes widespread in the US.
Today, a doctor from Ryde Hospital called for school closures, among other measures – a recommendation was picked up by SBS News and shared on the national TV bulletin. The Chief Medical Officer responded saying we are “not at that stage” and that,
“If we had more sustained community transmission, then we wouldn’t hesitate to make recommendations about public gatherings, schools and the like.”
But should schools be closed at all? First it’s important to distinguish between closing a single school temporarily for case investigations and the kind of pre-emptive and widespread school closures that the Ryde doctor was referring to. The former is a useful strategy to contain the spread of virus in a population known to be exposed.
Closing schools as a pre-emptive strategy needs some evidence about its effectiveness and the costs. For effectiveness, the best place to turn is infectious disease modellers (there are people who live and breathe this). They look at features of the virus and its behaviour in populations – things like infectiousness, death rates, incubation period, asymptomatic transmission, mixing of populations and many other things. Then they map out different scenarios using mathematic equations built from different assumptions.
Modelling from pandemic influenza (see here for example) shows school closures only work if they are done early, at scale and for a long period of time. Regardless, these models are often based on controlling pandemic influenza when buying time for a vaccine that was on the way within months. A COVID-19 vaccine likely 12-18 months away, at least. So how long would a closure remain for? What are the costs?
There are very large impacts from widespread sustained school closures, starting with:
- disruption to education of children
A review also noted the following impacts:
- large economic losses – one UK review estimated a cost of between £0·2 billion and £1·2 billion per week
- job insecurity and income loss for casualised workforce
- taking out big chunk of health workforce – about 30%
- children <13 yrs ‘self caring’, home alone
- differential impacts for those living in poverty or social exclusion.
In summary, school closures for early reduction of disease spread have major social and economic impacts. Less restrictive options can help minimise spread and reduce the peak of the epidemic. A useful paper published this week by three NSW public health experts sets out some options centred around more minor social distancing and hygiene measures.
For any strategies aimed at reducing the impact of COVID-19, decision makers need to carefully assess all the impacts, particularly for disadvantaged population groups. The best way for this to occur is for governments should have pre-established pathways for two-way communication with key groups. This links to Covello’s first cardinal rule of effective risk communication, to Accept and Involve the Public(s). Learning from people about the possible impacts of their intervention can enable government to anticipate the full scope of impacts, not just those imagined by decision makers. When combined with epidemiological and modelling evidence, it will result in higher quality plans.
The media will continue to report various calls for measures that aim to curb the spread of COVID-19. But decisions should be based on sound and careful analyses of evidence that also weigh social and economic impacts.
Human factors 