Publication - Advice and guidance

Coronavirus (COVID-19): framework for decision making - further information

Sets out further information about the challenges Scotland faces and provides illustrative examples of the steps that might form part of initial changes to the current lockdown restrictions, when it is safe to do so.

27 page PDF

1.0 MB

27 page PDF

1.0 MB

Coronavirus (COVID-19): framework for decision making - further information
3. Evidence

27 page PDF

1.0 MB

3. Evidence

This section provides an update on evidence concerning the virus and its direct health impacts. We will shortly publish a supporting evidence paper that provides further data on this harm and the three other harms (wider health, societal and economic) that will inform our review of the regulations and broader restrictions.

The spread of the virus in Scotland: our lockdown rules are working

As of 5th May, 12,437 people in Scotland have been confirmed to have been infected with the virus, and as of Sunday 26th April 2,272 people have been recorded as having sadly lost their lives.[2]

The virus continues to pose a serious threat to public health in Scotland. The Scottish Government will not change restrictions until it is safe to do so. A second surge in infection would cause further harm to our health, society and economy.

To judge whether and when restrictions can be changed, we will consider a range of evidence on the progress of the pandemic in Scotland, including the number of people in our hospitals with COVID-19, the number of those who are in intensive care, and the number of deaths. Community surveillance is underway with surveillance for antibody testing due to start soon. The charts and the numbers are available online, and are regularly updated.[3] A range of these data is shown in the charts below.

The charts show the rate of change of key measures of the pandemic in Scotland: new cases; hospitalisations; numbers requiring ICU care; and deaths. The number of new cases is a measure of those who have tested positive - and so, to some extent, reflects changes in testing policy (for example, increased testing in care homes). Similarly, the reporting of deaths is subject to time delays from actual date of death, as it reflects the date of registration of deaths. The charts show the changes in these key measures since lockdown was introduced. The inset pictures show close-ups of the situation in the last seven days.

The line in each chart shows the ratio of each measure in the last seven days compared to the previous seven days. Where the ratio is one, that indicates no change - i.e. the numbers of each measure in the last seven days is the same as the previous seven days.

Overall, the charts show the effect of the lockdown in slowing the transmission of the virus. In more recent weeks, there are signs this rate of change has stabilised, or plateaued. Sustained continued evidence of ‚Äčthe ratio staying below one will be required to be more confident that transmission is falling, and that numbers of new cases per day are falling, to the extent that is required for key elements of pandemic response - including Test, Trace, Isolate and Support - to be effective.

We can estimate, from these and other data, the total number of people in Scotland currently likely to have the virus, whether or not showing symptoms, and the current reproduction rate, or R, of the virus: the number of people, on average, who are catching the virus from each person who already has it. R must be well below 1 for a sustained period in order to suppress the virus.

Figure 1: New Cases Ratio
The graph shows the number of new cases in the last 7 days, divided by the number in the 7 days before. When this ratio is one, the number of new cases is not growing but it is also not decreasing. The line starts around 2 and a half on the 23rd of March, rising to a peak around 3 by March 30th. It then declines steadily to around 1 by 13th April, and from the 27th of April is definitely below 1. This is confirmed by an inset graph focussing on the last 7 days, showing that the confidence interval does not include one in the last 7 days. This means the growth of new cases is falling.
Figure 2: New Deaths Ratio
The graph shows the number of new deaths in the last 7 days, divided by the number in the 7 days before. It shows that the growth of new deaths has plateaued around 1 since April 15th. This means that the number of new deaths each week is about the same as new deaths the week before.
Figure 3: Hospital Occupancy Ratio
The graph shows the ratio of hospital occupancy in the last 7 days to the 7 days before. It shows that this ratio fell from around 3 at the end of March, to around 1 by the 15th of April. This means that the number of people in hospital is not growing, but neither is it decreasing.
Figure 4: ICU Occupancy Ratio
The graph shows the ratio of ICU/HDU occupancy in the last 7 days to the 7 days before. It shows that this ratio fell consistently to the end of March, to around 1 by the 15th of April. Using the confidence intervals, it has been statistically lower than 1, but only just, since April 25th. This means that the number of people in intensive care is decreasing slowly over time.

Graphics provided by the University of Edinburgh.[4]

Limited headroom to change restrictions

The following chart indicates how little, if any, room to manoeuvre we have in Scotland in terms of changing restrictions at the current time. It shows how quickly our COVID bed capacity (4,250) in Scotland risks being overwhelmed if the R number increases from its current level. The fewer infectious cases there are, the less likely it would be for a slight increase in R to overwhelm hospital capacity.

The chart shows that, on current estimates (and these estimates do change as the data are updated on an ongoing basis), there are approximately 26,000 infectious people in Scotland (vertical axis). This number remains much too high at present to consider the virus under control.

While precision on the R number is difficult, it is likely to lie between 0.7 and 1.0 (horizontal axis). This remains too high to be confident that case numbers will continue to fall. With the number of infectious people estimated at 26,000, this means that we are still at risk of a resurgence of the virus that would overwhelm NHS capacity. Moreover, this number is an average for all of Scotland. The R number for community transmission in Scotland is estimated to be below the R number in hospitals and care homes. This is a matter of critical concern.

Figure 5: Time taken for increases in R to overwhelm NHS capacity for different numbers of infectious people (assumes capacity of Covid Beds of 4,250 in Scotland) [5]
This chart shows the effect on NHS capacity for different values of R and numbers of infectious people. Where R is low (the green zone) there is no problem with NHS capacity. As R increases, so does the impact on the NHS until the black zone where if R was this high, NHS hospital capacity would be exceeded within a month.

(Source: Scottish Government)

Our current lockdown is pushing the number of infectious cases down. This means that the longer we leave restrictions in place, the more latitude we will have to change restrictions without risking a resurgence of the virus. But at present, given the continued high number of new cases and the imprecision around the R number for Scotland, which remains too close to one, the evidence is indicating that any change to the restrictions could easily lead to a resurgence of cases that would risk overwhelming NHS capacity in Scotland.

The risks to premature re-opening of schools and nurseries

The following chart indicates, for just one potential option that we are reviewing, the risk to re-opening schools and nurseries too early, even in a phased or redesigned way. It is based on very recent international experience and the data are still relatively limited, so our estimates of the impacts will change and become more robust over time.

This chart shows that, in the 'better' scenario, primary schools and nurseries in Scotland would be able to re-open without causing a resurgence in the virus. However, given the current level of infections and R, the chart also shows that, in the 'most likely' scenario, full re-opening would cause a resurgence in the virus such that hospital capacity in Scotland would be overwhelmed in less than two months. A 'worse' scenario would further exacerbate the situation. These scenarios illustrate the risks we face in considering different options and the merit in delaying a decision to re-open until transmission of the virus is much reduced from the current level.

Figure 6: Modelled estimates of the impacts on hospital capacity of fully re-opening primary schools and nurseries in Scotland in May, based on International experience [6]
This chart shows the likely impact on NHS hospital capacity over the coming months if primary schools and nurseries are opened in May. The vertical axis represents the number of hospital beds required: the most likely (dark blue line) and worse (green line) scenarios show demand is likely to outstrip supply of beds (the dotted red line)

(Source: Scottish Government)

Current assessment of COVID-19 evidence

Our current best estimate is that about 26,000 people in Scotland are currently infective and R is likely to lie between 0.7 and 1.0. While precision is difficult with these numbers, they remain too high to conclude that the virus has been suppressed and we must therefore continue to proceed with extreme caution. We consider that it would be unsafe to make anything other than marginal changes to restrictions until the estimated number of new cases is significantly lower and R is sufficiently below 1 to give confidence that easing the current restrictions will not increase it beyond 1.

An increase beyond 1 would risk exponential growth in the number of new cases, hospitalisations and deaths, causing very significant harm to Scotland's health, society and economy. Some restrictions and guidance are therefore likely to remain in place for the foreseeable future. And any signs of a resurgence in the number of cases, whenever it occurs, may necessitate the introduction or reintroduction of further restrictions.

The data so far show some encouraging signs of progress, but it is too early to say that is safe to ease restrictions. We will continue to monitor and publish the data and our assessment of what they show. At this stage, our view remains that there is very little room for manoeuvre in introducing changes to the restrictions currently in place.

The longer we leave the lockdown in place, the fewer people will be infected by the virus. This means that we will have more latitude in the future to change restrictions without risking a resurgence of the virus. In the meantime we are working with partners to ensure that any change to restrictions can happen safely when the time is right and to put in place the public health capacity and capability to ensure that, after we have successfully suppressed the virus, any future outbreaks can be effectively contained.

There is some evidence that the currently R number in Scotland is slightly above that elsewhere in the UK though comparative estimates depend on models used and are subject to a significant degree of imprecision and variation over time as new data become available. Any meaningful variation in the R estimates among the four nations could be a significant factor in co-ordinating decision making across the UK.

On our assessment of the current evidence for Scotland, it is almost certain that no significant change will be possible on 7 May. This fundamental assessment informs our consideration of the options set out in the following section.