Are freshers’ weeks in September going to be banned?

That’s the conclusion from a new paper on the impact of “university reopenings” (yes, I know they never closed) on cases in Scotland featured in issue 80 of Covid Economics from the Centre for Economic Policy Research.

Even in the context of a successful vaccination programme in Scotland, the findings:

…invite a rethink of how close proximity activity in universities can be resumed, as the pre-existing mitigations barring severe curtailment of trading activity were not successful in keeping the spread in check.”

We have two big questions being addressed here. The first asks whether it was possible to see an increase in cases within locations containing halls that might be attributable to the migration event. The second addresses this potential spillover of Covid-19 from halls to the wider community and adjacent areas – “spillover analysis”.

To work all of this out researchers geolocated all of Scotland’s student halls of residence (both university owned and private PBSA provision) and matched that to neighbourhood level Covid-19 case data from Public Health Scotland. They also compared student neighbourhoods with their nearest neighbours.

First they contrasted daily cases in neighbourhoods with student halls with cases found in immediately adjacent neighbourhoods. A peak of 220 cases per 100k per day then fell following the big restrictions put in by the Scottish Government in October. Here’s what the graph looks like for the group of universities that started back on the 14th September:

Next, to evaluate potential spillover to more distant areas, researchers employed a “donut” approach, comparing cases occurring in areas adjacent to those neighbourhoods where halls are located (the inner ring of the donut) to contiguous areas (the outer ring donut areas), but excluding locations of the halls themselves (the donut hole).

It’s an interesting method – it accounts for things like incubation periods, testing delays, and multiple rounds of transmissions. The models used here also control for differences in baseline health and socio-demographic characteristics at neighbourhood level (including the proportion of over 50s, overcrowded housing and disability rates from the Scottish Indices of Multiple Deprivation interacted with month time trend) and a weekly measure of public transport mobility that varies at the local authority level.

The conclusion is clear. Locations nearest to student halls see an increase in cases (relative to their surrounding areas) approximately one week after the spike that was detected in student neighbourhoods. The increase in cases represents roughly an additional 20-80 daily cases per 100,000 and persist until the end of November. This represents a 20-50% increase in cases between the inner and the outer donut.

In many ways it’s not a particular surprise (and reinforces findings from Cambridge) to find that a strategy that focussed on reducing the capacity of campuses without reducing the capacity of halls of residence meant that we saw an infection and transmission problem centred on halls. The question is what we do about it in relation to this September.

Right now we have what looks like an exponentially growing third wave with a particular problem among the unvaccinated young. We also know that the variant that’s dominating that wave really requires both doses of the vaccine in order to be brought under control.

That gives policymakers a headache. The easy bit is controlling spread on campus – we tried that already last Autumn by basically putting the padlocks on and hoping for the best. But doing forced students studying away from home into their rooms, and generated huge mental health issues – and a significantly diminished student experience.

The current “plan” seems to assume that a mixture of students and staff being regularly tested on campus (using tests now so problematic that they’re being banned in the US), and vaccinating international students on arrival if they have not had a jab, will do the job – but it’s not at all clear that it will.

The bigger issue is that right now the four governments are doing their best to keep restrictions universal. But there are only so many more weeks when the (double) vaccinated will put up with not being able to board the plane to Benidorm. If (and it’s a big if) we continue to see an exponential rise in cases amongst young adults you end up with three choices between now and “everyone’s been double jabbed”.

You can:

1. lift all restrictions and let it spread amongst the young (with the moral hazard of rare “long Covid” cases for those who’ve not yet been able to get both jabs);
2. ask everyone to wait to until everyone’s been double jabbed, or;
3. implement/retain restrictions on the young while lifting/easing restrictions on the old (a version of which is implement/retain restrictions on those not double jabbed while lifting/easing restrictions on those that have been).

Few governments would want to be seen to be doing 1) or 2), and option 3) is fraught with problems – we’ve already seen vaccine passports all but abandoned as an idea, and it’s not as if you can start explicitly turning people away from hotels and nightclubs based on their youth.

That leaves you with internal and international (im)migration – which governments are perfectly happy to restrict to stop spread. The thing about this research is that it pinpoints that September 14th opening date that was common to many Scottish universities as the big issue. So if we can’t get Britain’s 18 year-olds double jabbed by then (and given the cross border flows of students, we do need a UK-wide perspective here), you can see how tempting it will become to ask everyone to pause the major migration event by a month until everyone has been.

We could (and probably should) be scenario-planning an online start and then a reading week that doubles as a Freshers week – with, as is now traditional, a failure to compensate students for accommodation they can’t use.

The final piece of the puzzle then is international students. You first have to work out how on earth you cope with the vast amount of inward migration that almost certainly would exceed the capacity of the current hotel quarantine schemes in use – which is why there’s intense lobbying around halls being an alternative. But then you have to figure out how to get to “double vaxxed” status quickly – either by relying on home countries’ efforts (which may be behind our schedule and/or use vaccines not yet approved in the UK) or by vaccinating on arrival (which would still need several weeks to get to double dosage). This feels like a head scratcher that needs more work.