Famed Epidemiologist (and former Southampton footballer) Matt Le Tissier produced some surprising and deeply researched analysis that he, rather than publishing in a scientific paper for peer review, posted to Twitter on May 2nd (week 18 of 2020, which although it sounds like an incidental fact, will become pretty relevant):
Matt went on to explain that this was posted as a form of education, as he wanted to fight back against the “narrative of scaring the sheeple” and was “just stating [the facts]”:
Although it would seem foolish to go against such robust analysis from an esteemed intellect, it might be worth considering, for a few minutes at least, what other facts Matt has excluded from his analysis, and what a more accurate and considered picture might show.
Is Matt’s Fact Correct?
Matt may have stumbled on to a fact that confirms his analysis that a “narrative of scaring the sheeple” is seriously flawed. If we look at ONS stats for deaths [1] up until Week 15 for the previous 5 years, we see that 2018 was indeed a year that outstripped the supposedly deadly COVID-19 year of 2020:
It’s odd that Matt chose Week 15 given that on May 2nd, when he tweeted, we had data that included Week 16 deaths from the ONS.
2018 – Excess Winter Deaths hit 40 year high
2018 was no ordinary year, as Matt implies. In fact, 2017/2018 recorded the highest number of excess winter deaths for more than 40 years [2], “the number of excess winter deaths observed in 2017/18 was higher than all years since the 1975/76 winter period”.
This was down to several factors. The 2017/2018 winter saw an especially ineffective flu jab [2], and the pernicious and ever increasing effect of unprecedented NHS underinvestment [3] and the chaotic Lansley reforms of 2012 [4] no doubt played their part. This underinvestiment correlates with a rather unfortunate fact that since the Conservatives came into power in 2010, deaths had been showing year on year increases, which reversed the promising trend of the 2000s [5] where deaths reduced, in large part due to unprecedented investment (note investment) in the NHS under successive Labour Governments.
The Ineffective Flu Jab
Nick Stripe, head of health analysis and life events at the ONS, said at the time of this analysis in November 2018 that [2] “The number of excess winter deaths in England and Wales in 2017-2018 was the highest recorded since the winter of 1975-1976. It is likely that last winter’s increase was due to the predominant strain of flu, the effectiveness of the influenza vaccine and below-average winter temperatures”.
The flu jab had been particularly ineffective that year, with statistics showing that [2]
“Last year’s flu jab protected just one in 10 pensioners. The worst protection was among over 65s – the age group most vulnerable to flu – with effectiveness of 10.1 per cent and none at all against some key strains, an evaluation by Public Health England shows. Figures among younger adults were little better at 12.2 per cent, with rates of 26.9 per cent among children, the provisional end-of-season estimates show. The failings contributed to the worst flu season for seven years, with 15,000 deaths from the virus, around twice the average figure, and the worst NHS performance on record”
NHS Investment and Reforms
Also worthy of comment is the general upwards trend in deaths since the Conservative Government came into power in 2010, with the hard to ignore record underinvestment and chaotic Lansley reforms no doubt significantly contributing to this general malaise.
A More Considered Analysis – Is COVID-19 really all that bad?
So with this context in mind, what would a more considered analysis look like? Is COVID-19 so extraordinary after all?
It is obviously spurious of Matt to choose a selective reporting period and compare that to 2020 without the necessary context. The 2017/2018 winter was the worst winter on record for excess deaths for decades, and the period Matt chose had a significant number of weeks included where COVID-19 had barely gotten started. A classic manipulation of Statistics to fit a narrative, rather than using Statistics to inform or shape one.
If we look at excess death data, in particular comparative weekly excess deaths, we see just how extraordinary COVID-19 really is, even when comparing to Matt’s favoured year of 2018. It goes to show (ignoring for a second the arguments that it was too late) how necessary a lockdown was (how much worse it could have been with no intervention [6]):
Perhaps Matt’s epidemiological qualifications should be questioned, not the unprecedented nature of the COVID-19 outbreak and the UK’s precarious position.
References
[1] ONS, “Deaths registered weekly in England and Wales,” 03 05 2020. [Online]. Available: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/datasets/weeklyprovisionalfiguresondeathsregisteredinenglandandwales.
[2] L. Donnelly, “Ineffective flu jab blamed as extra winter deaths hit 40-year high,” 30 November 2018. [Online]. Available: https://www.telegraph.co.uk/news/2018/11/30/winter-deaths-hit-highest-level-40-years-experts-blame-ineffective/.
[3] F. F. Team, “Spending on NHS in England,” 09 July 2019. [Online]. Available: https://fullfact.org/health/spending-english-nhs/.
[4] L. Goodall, “Lansley reforms of 2012 Trusts,” 21 April 2020. [Online]. Available: https://twitter.com/lewis_goodall/status/1252701845367263232.
[5] D. Clark, “Death rate per 1,000 resident population in the United Kingdom from 2000 to 2018,” 26 November 2019. [Online]. Available: https://www.statista.com/statistics/281478/death-rate-united-kingdom-uk/.
[6] C. Cookson, “UK’s original coronavirus plan risked ‘hundreds of thousands’ dead,” 16 March 2020. [Online]. Available: https://www.ft.com/content/249daf9a-67c3-11ea-800d-da70cff6e4d3.
Infinite Jest Thoughts 