Half of students in (some) halls got Covid

Some in the sector were sharing the PHE press release within seconds of its publication, but it deserves some scrutiny. What we have here is a “seroprevalence study” carried out between 2 and 11 December 2020 at five universities – you can draw your own conclusions on the representativeness of Leeds Beckett University, Newcastle University, University of Manchester, Oxford Brookes University and Reading University. All the usual caveats about pre-prints apply here, by the way.

3,423 students who volunteered to take part had usable samples, and the study found that 17.8% of 17 to 25 year old students in that sample (2,905 of them) had antibodies from a prior Covid-19 infection. This, says PHE, suggests there was not widespread transmission of the virus across the wider student population. Of course it also suggests that a substantial proportion of students were still susceptible to Covid-19 infection in December 2020. And 17.8% is still higher than general seroprevalence reported by the Office for National Statistics (ONS) in England in December 2020 of 12.1%.

There are lots of potential problems with the overall figure – two of the five universities recruited from people volunteering for LF testing before Christmas, which will have had its own biases – 62.6 per cent were female and just 1.2 per cent were Black.

But the distribution of the 518 seropositives may give us important clues as to what led to that infection. Students aged 17 to 19 years had 4.1 times greater odds of being seropositive than 23 to 25 year olds. That’s likely to be about the relationship between age, stage and living situation.

And surprise, surprise – the study found that 49 per cent of students living in halls of residence were found to have antibodies – at least, those halls where there was a problem (ie that reported infection rates of more than >8%). Overall, those living in halls had 2.9 times greater odds of testing seropositive than those living in other accommodation types.

Meanwhile research carried out for the Scientific Advisory Group for Emergencies (SAGE) by the higher education working group at the Isaac Newton Institute has a good stab at both understanding and learning from the Autumn term, and carries out some modelling of the future return of students. This time there’s more data, using data on outbreaks and infections and setting it in the context of Higher Education Statistics Agency (HESA) data on home and term-time postcodes for the 2018/2019 cohort of students.

There’s a lot in there, but the headlines matter for what we’re about to do next. First – the overall distribution of outbreaks were consistent with expected importations from student intake from the wider community – universities reflect community disease prevalence at the start of term. “Staggering” future returns appears to be of “somewhat mixed and limited value” – and may act mainly to delay the outbreak to later in the term.

It says that the level of student adherence to testing and isolation is likely to have a far larger effect than subtleties between different staggered return regimes. It also concludes that while asymptomatic testing programs likely did help to prevent large outbreaks, “extremely frequent testing” (all students, every 3 days) would be needed now to prevent a major outbreak under plausible parameters for the currently dominant variant in the UK.

We’re close to offering LF testing every 3 days (twice a week isn’t far off) but nobody I’ve spoked to thinks we’ll get even close to “full” participation from students – and if you compare the current numbers getting tested with the number self-reporting that they’re “back” in university towns and cities, there’s a significant discrepancy.

But it’s halls where, again, we have the biggest problem. I’ve been banging on about artificial “households” all year – and here the study concludes that larger and more traditional halls of residence pose higher risks for larger attack rates – segmentation into smaller households within halls is “unlikely to be able to mitigate this”.

Last summer there were several warnings about hotel-style halls being like landlocked cruise ships. On one level, running a campus at 30% of normal capacity but halls at 100% (and then making students spend more time imn there than they were designed for and than usual) tells its own story – but the study quotes some of the prior research, and we’ve written about some of the warnings given by SAGE to government before. Why were halls of this type allowed to fill up – with what turned out to be entirely predictable results?

It’s been abundantly clear all year that artificial households in halls don’t work – but remember, according to DfE, “instances where transmission did occur were associated with guidance not being followed”. Not DfE’s fault, see – students’ fault. Then when it went wrong we closed kitchens and common rooms, cooped them into their rooms and collected the rent. And in some cases we’ve continued to collect the rent even though the setting doesn’t look workable if they’d all been there.

The question now is what happens next. I’ve heard some people say “well Jim, halls are half full now and there’s been no outbreaks”. That may be right, but the results in the Newton work suggest that only partially filling halls could significantly reduce transmission risk, especially if this is coordinated to reduce shared spaces. It’s what happens if they now get full this Spring that we should worry about.

I’d certainly be very worried about running hotel-style halls in their usual format and letting mature students or anyone with an underlying health condition move in. And once you’re there, are we really sure that we should be running a herd-immunity experiment (with potentially terrifying long-Covid risks) on the young, all in pursuit of their “mental health”?