COVID-19 lessons from Scotland

Edinburgh / Scotland – March 23 2020: Edinburgh’s much loved Greyfriars Bobby statue protected with face mask during the COVD-19 pandemic

I have recently been looking at the age-structured COVID-19 data from Scotland, particularly analysing the latest wave. I think the last couple of months – and especially the last week or so – teach us a lot about the virus spread in highly vaccinated populations.

The size of the outbreak in school-age children is staggering. Possibly there was some under-reporting back in the first or second wave in these age groups, but the latest outbreak is definitely fueled by cases among young people, with some spill-over to the parent – and grant-parent – generation. Compare the peaks this September with the first and the second wave in the graph below:

The proportion of cases and deaths in different age categories in Scotland: black/blue: cases; red: deaths. Points: raw data, lines: 7 days average

At the same time, vaccines are clearly working – the death rate, particularly in 60+ age classes, is much smaller relative to what we did see in the pre-vaccination time. Again, compare the red points in the graph above, in the first and the second wave versus the current, third, wave – a similar number of cases resulting in much fewer deaths.

This is clearly the result of a fantastic roll-out of vaccines, in all – except the youngest – age groups, as shown below:

The proportion of people in age classes with one or more vaccine doses (blue line) and two vaccine doses (red line).

Two things are concerning. The delay in vaccinating young people – and the lack of proper control measures in schools like masks and ventilation – clearly allowed a massive epidemic in these age groups. The deaths were only very few, which is why this epidemic is not getting enough attention in the media, but the consequences in terms of long term complications – “long COVID” – might be massive. In addition, the vaccination progress in the youngest groups is slowing down and reaching a plateau at 70-80% – way too low to make an impact!

Secondly, as seen below, the rapid decline in cases which we saw in the last couple of weeks is slowing down. The virus seems to be bouncing back as the cases actually go up in some age classes (particularly older). This could be because the immunity is waning and we see more breakout cases, but more likely this is a reaction of the epidemic to relaxation of the rules.

The proportion of cases and deaths in different age categories in Scotland: black/blue: cases; red: deaths. Points: raw data, lines: 7 days average

The graph below captures this recent change using the 7-days ratio between cases (a rough estimate of the reproductive number). The value below 1 means the case numbers are declining, and the value above 1 means the case numbers are accelerating.

The ratio of cases delayed by a week for different ages. Points: data, line: 7 weeks running average. The horizontal line at 1. The vertical line at 16th August when schools started.