Something slightly surprising has been happening across Scotland in recent weeks. At the same time as public health measures are eased, as set out in the Scottish government’s strategic framework, the number of people testing positive for Covid-19 has started to fall.

This is the opposite of what some expected, with concerns that opening up would lead to a sustained period of rising infections. The obvious answer is that Covid-19 vaccines are working. But it’s not as simple as that. Now that a similar trend is being observed in England and Wales, even the public health community is slightly puzzled.

Quite a bit has been written about this in recent days in England but it’s worth a closer look at data from Scotland. Earlier in the summer, the weekly Modelling The Epidemic report produced for Scottish government suggested cases could reach up to 100,000 per week by mid July under a worst case scenario. Instead the most recent peak was reached at the end of June with a seven-day average of just under 3,500 per day. Since then the fall has been steady and fairly consistent.

Is this due to fewer people coming forward for testing? The Office for National Statistics infection survey shows a different picture to case numbers. It tests a random sample of people in the community so isn’t dependent on people tested with symptoms. It still shows more positive tests each week starting from when restrictions were eased in mid May, reaching around one in 80 people in the week up to 17th July. The mismatch between ONS and case numbers may suggest fewer people have been taking up testing.

This isn’t a new phenomenon – surveys have suggested at various points in the pandemic that uptake of symptomatic testing can be poor. A recent pre-print from the CORSAIR study of just over 3,600 adults in England and Scotland during June reported just one third of people who reported that they had developed a cough, high temperature and or loss of taste or smell indicated that they’d taken a test to confirm whether they had Covid-19.

In Scotland, there are fewer tests being done now than earlier in July but this has fluctuated and in recent days test positivity has declined. This suggests a drop in testing is not the main explanation. There are also some signs that Covid-19 levels in wastewater are declining at least in some sites. This supports a genuine fall in virus levels.

A further bit of evidence that indicates the decline is real is that there has been a small but encouraging drop in hospital admissions since the middle of this month. We know admissions are significantly lower than during earlier waves of the pandemic because amongst those fully vaccinated, the risk of admission is reduced by over 90%, but not completely and when cases were rising in recent weeks we still had people in their 30s, 40s and 50s with no or partial protection.

There is also speculation that Scotland has reached ‘herd immunity’ when most of the population is immune to the virus from vaccines or prior infection. Putting a figure on the proportion of people who need to be vaccinated in order to reach population immunity is very difficult, and has been made worse by the more transmissible Delta variant. Some argue that without vaccinating children, we may not reach it.

But the vaccines are making a big difference. The more people receive their second doses the more vaccines' ability to reduce transmission (as well as illness and death) will play a part.

A further point is whether the weather is helping. Other coronaviruses that affect humans are more common in winter. But globally there isn’t much evidence for this as warmer countries have experienced significant waves of Covid-19.

At least two factors aid us in the summer. The first is that we spend more time outdoors meaning the risk of infection is much lower. The good weather we’ve had in recent weeks has helped. There is also some evidence that ultraviolet radiation in sunlight reduces how long the virus can survive, for example on surfaces, although picking up the virus from surfaces is not the main route of transmission.

Others have pointed the finger at schools as a driver of infections and that once they close for the holidays that has a dramatic effect on cases. This will be contributing, as the activities that take place inside schools, transport to and from school and children and parents gathering involves households mixing. The fact that cases dropped soon after schools closed for the holidays in Scotland lends credence to this, but a closer look at cases by age shows they have been declining across the age ranges – there wasn’t a sudden drop among school-aged children when term ended.

A further consideration is whether infections rose more rapidly because of mass events, specifically the Euro 2020 football tournament. The striking rise in cases among younger men (who make up the bulk of fans) associated with the tournament supports this, but we can’t blame the football for Scotland’s third ‘wave’ – there have been many factors at play.

Finally, is our behaviour playing a role? Data on the average number of contacts we have with people we don’t live with – which is also reported in the weekly Modelling The Epidemic report – shows that these remain at lower levels for most groups than they were last summer. They’ve risen since May but not dramatically, suggesting a lot of people are remaining cautious.

The most likely conclusion is that a combination of some or all of these factors accounts for where we are now. What happens next depends on many different things. There is still a risk that cases could rise again in the coming weeks. But direct health harms from the virus are diminishing and we can look ahead with more confidence.