Facebook post 15 April
Today’s post about COVID-19 will be a shorter one. The geek within me is loving the challenge of research, data interpretation and then trying to distill and articulate all of that on social media. If it wasn’t for the fact that it is a ’situation’ we are all living right now, it would actually be fun!!
But, I have studies (I am following an on-line lecture program in nutrition for GPs and clinicians which is becoming remarkably relevant to answering the question “what do we do next regarding coronavirus?”) which are getting neglected. Last week I had got into a good routine progressing my Icelandic language skills and that has lapsed over the last couple of days. Plus my online teaching and ‘content’ creation, so allow me just to throw some thoughts at you today and perhaps tomorrow get back ‘on it’.
One thing I was looking at yesterday was the economic consequences of lockdown. Discussion of whether to end lockdown is often dismissed as a ‘money over lives’ argument. Lockdown is saving lives (it is alleged) and so the pressure to get people back to work is, according to the argument, putting money over lives. Let me challenge that and show you it is actually an argument more like “lives now versus lives later’.
An analysis I have seen by Bloomberg posits that each percentage point increase in unemployment leads to 40,000 deaths over the following six years (an analysis based on U.S. data). The mechanisms are many but stress, change of lifestyle, potential poverty, all contribute to this figure. Now you scale that by worldwide estimates of total unemployment a prolonged lockdown will create, and that is hundreds of thousands of deaths in the next six years. Again remember this is from U.S. analysis.
In the U.K., winter excess deaths were nearly 50,000 in Winter 2017-18 That is, in that winter 50,000 more people died that you would have expected in an average winter. Fifty thousand. That is not in everyone’s consciousness. Now, those deaths were not ‘spiked’ into a a few short weeks which is what a viral pandemic will do (data profiles from Italy and China show that very clearly in the current pandemic) but that is a big number. Total deaths from coronavirus at the moment, and let me extrapolate for potential under-recorded COVID-19 care home deaths (the deaths are not under-reported but the fact that they might be COVID related, is) then we are probably at around 15,000. But the rate of new cases is levelling-off. Allowing for the lag in the death data compared to new case data, models will put total COVID deaths at 30,000 to 40,000. Still less than the total excess of 2017-18.
So, a few tens of thousands of deaths of COVID-19 (which may in the full year data be similar to a bad flu year) versus a few hundreds of thousands of deaths over the next six years because of economic down turn.
Another question: Has lockdown made this number smaller?
Probably not.
In the Italy data, the rate of new cases as a proportion of existing cases was already beginning to decline BEFORE lockdown was effected. That may seem odd, as we can all remember the ever increasing daily death figures from Italy during March, but (and this is something I will explain in a later post using the analogy of water filling a bath) when you have an exponential phenomenon like viral spread – one person infects two, those two infect four, those four infect eight etc – then you have to look not at absolute increases, but increases relative to the current case number to spot trend.
I suspect when the data for the UK becomes available we will find that here in the UK, new case rate as a proportion of existing cases was already on its downward trajectory at the point of lockdown. It is unlikely that our profile is any different from Italy’s.
We also need, in order to support the ‘flatten the curve’ meme that became as viral as SARS CoV2, an analysis of the extent to which the demand on healthcare resources versus the availability of those resources is, or would have been, a contributor to excess deaths. The hole in the data I analysed yesterday is partly, maybe wholly, explained by deaths in Care Homes at deaths at home, which are probably under-reported as COVID cases. But this raises the question, are healthcare resources really a limiting factor if nearly half of COVID deaths occur out of the hospital system? Did the curve need flattening?
Do not get me wrong, any death is a tragedy whether it is from this viral infection or the economic consequences that follow from lockdown, we just need to be realistic over long term consequences.
So, to quote Winson Churchill: “No matter how good the strategy, you should occasionally look at the results”. Maybe now is one of those times.
ADDENDUM
Again following some great comments I have more to add. This was in response to the “damned if you do, damned if you don’t comment from Fiona:
I would agree about the “damned if they do damned if they don’t.” At the time of the decision to lockdown, it probably was the best route.
But I would say, and this is the context of the Churchill quote, is that just because it is the approach that has been adopted, it is important to keep tracking its effectiveness as more data becomes available. And more data is becoming available.
I think a rationale of lockdown was to reduce death (“Stay at home, save lives”) and this worked on the model that there is a horizontal line on the deaths over time graph, that represents the limit of our healthcare system. Lockdown, was aimed at lengthening the curve so that it became less ‘peaky’ and more of the curve stayed below this posited horizontal line.
Using data we now have and data I will share in tomorrow’s post, the severity of COVID-19 symptoms IS becoming more predictable. It is not the lightening strike or snake in the grass that could get any of us unexpectedly but its severity of symptoms is very correlated to some key health markers.
The other thing about lockdown is that by delaying exposure it allows more time for treatment research. What I will share at some point is the fraught history of SARS vaccine research since 2002 and how unlikely it is a vaccine will be found. Data from this week shows the mutation rates of SARS-CoV2, and so notwithstanding the tech issues with making a vaccine it is unlikely to be a robust solution.
Again this is all data since lockdown so the discussion point is not whether we should have done it, but is there a valid case to continue if all we have is herd immunity and improving the metabolic health of the population as our management strategies.
One thing I think is urgent to address is the panic and how scared the population are becoming. This is a strong contributor to the lives now versus lives later analysis. And a scared population, when lockdown is ended, will be less robust immunologically and that is a whole extra consideration…