The Univeristy of Melbourne The Royal Melbourne Hopspital

A joint venture between The University of Melbourne and The Royal Melbourne Hospital


08 Nov 2021

Work Package 2: Subpopulations – Schools

Keeping schools open and returning students to in-person learning has been identified as a national priority for all states and territories. This work assesses the effectiveness of a variety of school-based surveillance, contact tracing and quarantine strategies to maximise face-to-face teaching by reducing transmission in schools and preventing outbreaks in schools.

Key findings:

Early infection detection and high vaccination coverage amongst staff and eligible students markedly reduce outbreak risk.

  • Management of cases or outbreaks in schools: allowing ongoing school attendance for class contacts of a case through a ‘test to stay’ strategy achieves equivalent outbreak containment to home quarantine and enables maximal face-to-face learning.
    • Test to stay was modelled as daily asymptomatic testing of classroom contacts of a positive case for seven days using rapid antigen tests, with students allowed to continue face-to-face learning without quarantine provided they continue to test negative.
    • The effectiveness of test-to-stay requires compliance with testing, but even at partial (e.g. 50%) compliance, it is effective relative to no test-to-stay or no quarantine.
    • The high frequency of testing (daily for seven days for classroom contacts) compensates for the reduced sensitivity of rapid antigen tests.
    • These findings were true for primary and secondary schools.
  • Surveillance: regular screening of students in schools at risk of COVID-19 incursions can result in fewer infections and in-person teaching days lost.
    • By detecting cases faster, there are fewer infections present when the first diagnosis is made and a lower risk of larger outbreaks occurring.
    • Identifying and isolating cases earlier leads to fewer downstream cases requiring isolation.
    • This was true for primary and secondary schools.
    • The benefits of student surveillance testing for reducing infections and days of face-to-face teaching lost increase as incursion rates increase.
    • The more frequent the screening the greater the benefits (i.e. daily versus twice weekly versus weekly).
  • Surveillance: regular screening of teachers had minimal benefit for reducing outbreaks in schools.'
    • Teachers only comprise a small proportion of the school community.
    • However, this analysis only considered outcomes following an incursion in a school, and does not capture potential benefits that screening teachers may have on preventing of incursions.
  • Findings may vary depending on assumptions for the number of non-classroom contacts students have.
    • The quarantine and test-to-stay strategies modelled were focused on classroom contacts rather than close contacts as they are more practical to identify.
    • These strategies will be less effective if a greater proportion or risk comes from non-class contacts (i.e. if students have more contacts outside the classroom).

Outside the scope of this analysis/future work:

This analysis focuses on transmissions taking place within schools following a COVID-19 case in a school. It does not consider:

  • The benefits of community public health responses on reducing COVID-19 coming into schools.
  • The role of school transmission on the broader community.
  • Early Childhood Education and Care or specialised settings (e.g. specialist schools and boarding schools) or small rural schools.
  • Severe COVID-19 cases or hospitalisations associated with infections in schools (rather, the focus is strategies to reduce total infections).
  • Implementation of different strategies.
  • Cost or cost-effectiveness.
  • Heterogeneity between schools or inequities of implementation.
  • Indirect health impacts on students of prolonged home-based learning (rather, the focus is to maximise face-to-face teaching).


  • If rapid antigen tests need to be self-administered, then compliance may reduce (see sensitivity analysis).
  • Rapid antigen tests are assumed to have a sensitivity of 77.3%, meaning that when someone with COVID-19 takes a test it will say they are negative nearly a quarter of the time. This estimate of sensitivity was based on the lower bound (to account for inconsistent use) from Muhi et al. The Lancet Regional Health-Western Pacific, 2021;9:100115. If actual test sensitivity were lower, this would make a test-to-stay strategy less effective, in a similar way to lower compliance.
  • Contacts and mixing patterns in schools are uncertain: for example if students have more student-student contacts outside of classrooms per day than estimated (two per day for primary school students and five per day for secondary school students) then quarantine or test-to-stay strategies that focus on classroom contacts would be less effective, and surveillance strategies would be more effective.

Frequently Asked Questions:

How high does transmission need to be before surveillance should be considered?
Surveillance of students through twice weekly rapid antigen testing always reduces outbreak risk in the model, no matter what level of transmission is occurring in the community. However, surveillance testing has greater benefits as community transmission increases, and so if it were to be targeted to some areas then the greatest benefits would be in communities with high rates of transmission. Decisions about whether surveillance testing is used, and how widely, are also likely to depend on factors such as acceptability among students, parents and teachers, feasibility, logistic challenges and costs.

Why consider alternatives to quarantine for students who are exposed to COVID-19?
Current protocols that require that if an infectious person is detected in a school, their close contacts (e.g. classmates) need to quarantine for 7 or 14 days, in case they are infected but not yet detectable. The purpose of the quarantine is so that if they do become infectious, they will already be removed from the school community and so cannot further transmit the virus. However, the high number of incursions into schools means that students are needing to spend lots of time in quarantine, which is difficult for children, parents and caregivers. It is therefore worth considering if there could be alternate options that would be equally effective for prevention of transmission but could enable face-to-face teaching to continue.

What’s the advantage of test-to-stay in schools?
The test-to-stay strategy has the advantage that quarantine of classroom contacts is no longer required when a case is detected in a school. Instead, classmates of the positive case will be required to have a daily rapid antigen test before attending school and can only attend if they test negative in the morning. This will hugely increase face-to-face teaching. Very importantly, the model shows that the test-to-stay strategy is approximately equivalent for prevention infection and transmission in schools.

Will test-to-stay in schools compromise community safety?
No. The modelling shows that if classroom contacts of positive cases are tested daily with a rapid antigen test for seven-days following exposure, this is approximately equivalent to quarantine for preventing onwards transmission.

When will test-to-stay be rolled out in schools?
It is unknown if or when test-to-stay would be implemented. This is only a modelling study, and not a policy announcement. This study is a proof of concept showing that test-to-stay has potential to be effective in Australia, and the next steps will be to determine how programs could look in practice. This includes considerations of acceptability among students, parents and teachers, feasibility of implementing (e.g. recording test results, delivering tests), logistic constraints and costs. None of these have been designed or decided yet.

How long would test-to-stay in schools be needed?
It is unknown. This analysis has only considered what would be required under a public health strategy of minimising cases.

Could test-to-stay apply to workplaces as well?
These results are only for schools and cannot be extrapolated to workplaces or other settings. Schools are unique because children are less infectious and less susceptible to becoming infected than adults. Also, contact networks in schools are easily identifiable and are contained, whereas adults have very different mixing patterns in workplaces. Other studies would be required to determine if test-to-stay would be effective in workplaces or other settings.