select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
The Office for National Statistics (ONS), London School of Hygiene & Tropical Medicine (LSHTM), and Public Health England (PHE) have released the fifth round of results for their COVID-19 Schools Infection Survey, providing initial estimates of staff and pupils testing positive for COVID-19.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“The results from the latest round of Covid-19 Schools Survey are patchy, and can’t provide a clear guide of what might be going on with Covid-19 infections in English schools generally. The testing for Round 5 took place between 5 and 21 May, so more than a month ago now, which means it really can’t tell us much about what’s currently going on in schools. It might have provided some information on the effect of policies on mass testing and other precautions in schools in past months, but important gaps in the data make this difficult or impossible. Also, it can’t tell us about the overall (average) position in schools across England, because the sample of local authority areas included in the survey was never intended to be representative of local authorities in general. The ONS bulletin makes this very clear. When the local authorities where the survey was to be done were chosen in the early autumn last year, more local authorities where infection rates were high at the time were chosen, presumably to allow for more accurate surveillance there. This makes it impossible to use the results to provide national estimates – and also, the areas that had high or low prevalence back in September didn’t necessarily have high or low prevalence as the study goes on. Statistical adjustments (weighting) are made, so that as far as possible, the results for each local authority are reasonably representative of the position across all schools in that local authority, but that can’t fix the issue that the local authorities aren’t representative of all local authorities.
“Before I get to the estimates that have also appeared in previous rounds of the survey, I’ll mention one very interesting piece of information that we haven’t had before in these reports. It comes from a short questionnaire that parents were asked to complete between 12 April and 18 May, about their views on vaccination for their children. For parents of primary pupils, 43% said they would definitely be willing to have their child vaccinated, with another 46% saying they weren’t sure but probably would say yes. Only 4% said they definitely would not be willing, with another 7% saying they weren’t sure but would probably say no. For parents of secondary pupils aged under 16, the estimates were 53% definitely willing, 35% unsure but probably yes, 9% unsure but probably no, and just 3% definitely no. Again this applies only in the local authorities included in the survey, and can’t be taken to represent the national position across England. It’s also possible (though there’s no direct evidence of this) that parents who were unhappy with vaccination would simply have declined to answer the questionnaire, rather than answering and saying no. Also, I can’t find information on the response rate or sample size for this specific questionnaire, and the statistical error bounds are quite wide. So there’s some uncertainty about exactly what the results are telling us – but they do seem to indicate a very low level of vaccine hesitancy among parents.
“The Round 5 results include estimates of the percentages of primary school pupils and of secondary school pupils who would test positive for a current infection with the virus that can cause Covid-19. They do not provide estimates for staff with a current infection – those numbers are too low and therefore cannot be published because of potential privacy and confidentiality issues. The estimated numbers of infected pupils are very low – only 3.6% (under 4 in every 1,000 pupils) in primary schools, and even lower (0.04% or 4 in every 10,000 pupils) in secondary schools. The numbers in secondary schools are considerably lower than the previous round in March, when the rate was 0.33%. That comparison can’t be made for primary schools, because the infection rates weren’t published from the March round because the numbers were too small. The statistical margins of error around all these estimates are wide. For round 5, for primary pupils the results are compatible with a true infection rate between 8 in 10,000 and 101 in 10,000, and for secondary pupils, it’s from essentially no pupils infected in every 10,000 to 23 in 10,000. Of the 57 primary schools where the survey was run, only 6 had any infected pupils at all (and none had more than one infected pupil). Of the 85 secondary schools involved, only 5 had any positive tests. On the face of it, it is gratifying that infection levels are so low. But we can’t take this to be any measure of the overall level of infection in school pupils, or even of the infections levels in the local authorities that are included. That’s because only pupils who were actually in school at the time could be tested. Pupils who had Covid-19 symptoms should not have been in school, and nor should pupils self-isolating because they had been in contact with an infected person, so they wouldn’t have tested positive in this survey because they wouldn’t have been tested at all in the survey.
“The other data in this release are on antibody and vaccination levels in staff. (There are no data on antibody levels in pupils, and vaccination does not come into the picture for them.) Unlike many other sources of data on antibody positivity, the antibody test used in this survey only reacts to antibodies from an infection, and not to antibodies due to vaccination. About 27% of primary school staff would test positive for antibodies, across the local authorities involved, and about 25% of secondary school staff (though again there are statistical margins of error around these estimates). Both figures are up slightly for round 5 (from May) compared to round 4 (for March), when the antibody positivity estimates were 25% for primary staff and 23% for secondary staff. These small increases could well be because some staff had a new infection between March and May, but actually they are so small that it remains possible that there isn’t a real increase and the change is simply due to statistical variability. I wouldn’t have expected a large rise in antibody positivity from infection, given that national infection levels were not very high during that period. The rise in antibody positivity in staff is much smaller than the rise between round 2 last December and round 4 in March – but again that’s not at all surprising because infection levels were so high early this year. But again, I’ll remind you that these data do not necessarily reflect the average position across England. For the antibody data, this release gives information for the individual local authorities concerned, and the levels of antibodies and the change between round 4 and 5 do differ quite a lot between local authorities. For instance, for round 5 the rate of antibody positivity in primary school staff ranges from 12% in Bournemouth, Christchurch and Poole (and was probably even lower in Norfolk, where the data are suppressed because of low numbers), to 60% in Liverpool, where the rate went up from 49% in round 4.
“On vaccinations, given the success of the roll-out, it’s not surprising that the percentage of staff (primary and secondary taken together) who have been vaccinated has increased between the end of March and the end of May. By 31 March, 63% had had at least one dose – by 31 May this had gone up to 86%. Over the same period, the percentage who had had both doses had gone up from 1% to 43%. There’s a breakdown by the age of staff members, and of course it shows higher vaccination rates in older staff member. Details are again given for the overall staff vaccination rates in each of the local authorities involved, and the percentage vaccinated does vary quite a lot between authorities. Again across primary and secondary staff taken together, by 31 May, the percentage who had had at least one dose ranged from 78% in Salford to 92% in Leicester.”
https://www.ons.gov.uk/releases/covid19schoolsinfectionsurveyround5englandmay2021
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19
Declared interests
Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee. I am also a member of the Public Data Advisory Group, which provides expert advice to the Cabinet Office on aspects of public understanding of data during the pandemic. My quote above is in my capacity as an independent professional statistician.”