March 8, 2021

expert reaction to preprint using a simulation to look at potential outcomes of regular entire-school lateral flow testing

A preprint, an unpublished non-peer reviewed study, from the MRC Biostatistics Unit at the University of Cambridge uses a simulation to assess the potential contributions of LFD (lateral flow device) tests for school reopening in England.

Dr Louise Dyson, Assistant Professor in Epidemiology, University of Warwick, said:

“This work is very similar to prior work by Leng et al. (2021), which considered testing strategies in secondary schools, while Kunzmann et al. take a model of a primary school.  The main differences lie in the size of bubbles considered (since primary schools generally have smaller bubbles), and the range of scenarios tested.  In Leng et al. we find that weekly testing combined with daily testing of the contacts of cases can reduce infection levels while also reducing school absences compared to the pre-Christmas strategy of isolating contacts of cases.  However this combined strategy requires a large number of tests.  Kunzmann et al. do not study the combined strategy, and find that mass testing with isolation of the contacts of cases found reduces infection levels compared to the pre-Christmas strategy.

“The small bubbles used in this work reduces the number of school days missed, perhaps indicating that whether testing of contacts is needed to reduce absences depends on whether the school can successfully isolate smaller bubbles within the school.”

Reference: Leng, T, Hill, E.M, Thompson, R.N., Tildesley, M.J., Keeling, M.J. and Dyson, L. (2021) “Assessing the impact of secondary school reopening strategies on within-school COVID-19 transmission and absences: a modelling study” medRxiv https://doi.org/10.1101/2021.02.11.21251587

Dr Alexander Edwards, Associate Professor in Biomedical Technology, Reading School of Pharmacy, University of Reading, said:

“This is a fascinating study that will add to our understanding and ability to model the use of regular testing to reduce the spread of infections specifically through schools.  The biggest message from this preprint is that intuitive expectations of what might work cannot be relied upon to work out the benefits (and costs) of testing in schools.  We still need more models and better data, but the current best understanding suggests that regular school testing can be very helpful.

“There has been a lot of discussion recently about the purpose of, and accuracy of diagnostic lateral flow rapid tests.  This study – alongside other similar studies using different models and methods – illustrate something that “feels” surprising: diagnostic tests don’t have to be perfectly accurate to be useful.  One reason for this is that virus spread is typically exponential – so the more case there are, the faster the virus spreads – but this means even a small improvement in spotting cases early can have a significant benefit.

“Some clear messages remain about lateral flow tests.  They can’t pick up all infected individuals, and so even if you receive a negative test result, you must still follow current COVID control guidance.  Tests cannot ‘prove’ anyone is ‘safe’, because people can become infectious extremely quickly and not everyone who is infected will test positive.  However, just because they don’t have perfect sensitivity, doesn’t mean they are pointless, ineffective or unsafe.  The more people that embrace these community testing schemes, the more likely they are to succeed.  The aim of these testing programs is to allow schools to re-open and stay open, without rapid spread through communities leading to further restrictions in the future.  The other joint aim is to spot case within communities as early as possible, rather than waiting for spread and symptoms to appear, by which time self-isolation is harder and less effective.

“Finally, we are re-learning something that has been known for years, that it is vital but hard to work out the best way to use clinical diagnostics – especially in community screening programs.  The more we can invest in and advance our understanding of “real world” use of diagnostic tests, the better we can use testing to tackle health problems.”

Preprint (not a paper): ‘The `how’ matters: A simulation-based assessment of the potential contributions of LFD tests for school reopening in England’ by Kevin Kunzmann et al.  This work is not peer-reviewed.

https://arxiv.org/abs/2103.02035

Declared interests

Dr Alexander Edwards: “No conflicts although I am co-investigator for a research project understanding how instruction design affects usability performance of rapid diagnostic tests in the community.”

None others received