Lock-down: did it work, and did we do it at the right time?

Lots of people are asking these questions. Many people are struggling with lock-down. It is hurting us financially, physically and mentally. So it is only natural to ask the question “is it right?”.

I want to set some expectations. The answer is, of course, we don’t know. Not yet anyway. But let me explain why we don’t know.

Before doing that, however, we need to be clear about what “right” actually means, and to do that, we have to be clear about the objective of lockdown.

The ultimate objective is to reduce the total number of deaths. Notice I do not say ‘total number of deaths due to Covid-19’. I mean total deaths. This distinction is important for a couple of reasons:

• it would be counter-productive if achieving fewer deaths due to Covid-19 we ended up causing a bigger spike in some other cause of death – whether that be cancer, suicide, seasonal flu or whatever (which is why excess deaths are also an important measure); and
• damage to our economy is an inevitable consequence of lockdown, and the likely medium and long term impact of that on poverty will also lead to deaths.

That said, reducing the deaths due to Covid-19 appears to be the dominant contributor to total deaths at this time, so let’s look at how the lockdown has helped that.

The graph above shows data from six countries on a time series based on days since each implemented their respective lockdown. The vertical axis shows the number of daily deaths reported per million of population. This ‘per million of population’ allows us to factor-out to some extent the fact that a bigger population is obviously going to have more deaths, and thereby compare country to country a little more reliably.

The death toll is counted using a standard global measure (based on deaths with Covid-19 recorded in hospital). It is not all deaths, but this method makes it as easy as possible for all countries to count the same way, and again, this contributes to our ability to share, compare and learn form the data.

However, despite the relatively good provenance of this data, it doesn’t help us answer our question: the grey block indicates the currently estimate range of when people who contracted the virus on day 0 (lockdown day) would die. Until this period has passed, we will still be counting deaths that are due to infection before lockdown, and therefore will not have a clear picture of whether lockdown has been effective or not.

Unfortunately, in the UK at the time of writing we are on day 19. We need to get to day 56 before the daily data is free from pre-lockdown effects, and then have a reasonable period (say to at least day 70) before we can start to infer anything from the data.

So if we can’t use death data yet, is there anything else we could look at that might be a reasonable proxy?

Well, yes – maybe. How about cases? This is of course what everyone reports on, and that is because a case can be identified much faster than a death, and so the data is more timely.

This graph is similar to the first except that instead of new deaths it shows new cases.

The grey bars indicate possible thresholds for pre-lockdown effects on case numbers. The left-most bar reflects the 5-6 day incubation period (before symptoms appear), and the other bar reflects an assumed 7 days between symptoms appearing and the case being recorded.

This latter 7 days may be too long. according to Dr Sarah Jarvis MBE (How quickly do Covid-19 symptoms develop and how long do they last?, patient.info, reviewed by Dr Colin Tidy), hospital admission for severe cases is likely to be 6 days after symptoms begin, and given the test results come back in 24-48 hours, that would give recorded cases on days 7 – 8.

The other problem with using case numbers is that they aren’t as objective as deaths. The biggest factor here is the amount of testing that each country is doing, and this varies quite dramatically. On the chart, the UK and Germany have had reasonably similar new case numbers for the last couple of weeks. However, Germany have been doing more testing, so the number of untested cases (and therefore cases not showing in these numbers) in the UK may be considerably higher than in Germany.

To see the impact of the lockdown in daily case numbers, we need to wait until 12 – 14 days after the lockdown.

If we look at the right-hand third of the graph – post day 14 – we can see that all countries have a slight downward trend, or an erratic curve but with no clear upward or downward trend.

For the UK, we have just 5 days data in that period, and so drawing any conclusions – as our scientists keep telling us – is premature.

Back to our original questions:

Did it work? The simple answer to this is that it is just too early to say.

Did we do it at the right time? Hindsight is a wonderful thing, but if we look at the basic data above available on the day the UK lockdown was enacted (23rd March), we were at a lower end of the range of new daily cases per million compared to other countries (which suggests we acted earlier based on case numbers).

Looking at deaths per million it is less clear. A look at the total number of deaths per million shows that the UK had reached 4.2 before lockdown. This compares favourably to Italy (who had reached over 7, although to be fair were also below 4 just two days before their lockdown started), but less well with Germany, France and Spain (who locked-down at 0.8, 1.9 and 2.5 per million respectively). Or to look at it another way, if we compare our lock-down date with the date at which the UK data matched the deaths-per-million data when other countries imposed theirs, we were:

• 4 days ahead of Italy;
• 2 days behind Spain;
• 3 days behind France;
• 6 days behind Germany; and
• 9 days behind China.

This gives us one of many ways to compare the timing of our lockdown. However, it is not the only way to measure it, and it still doesn’t tell us who made the right judgement call; who responded fast enough to allow their health system to cope; fast enough to minimize Covid-19 deaths; and late enough to minimize the other fatal consequences of lockdown.

So let’s all keep sane: based on what we can see today, the UK decision to lock-down was broadly in line with what other countries have been doing. The government have not, it would seem, gambled with our health any more than any other government.