Covid: when the figures become the problem

By Richard North - January 21, 2022

I see that the government is proposing to end the daily publication of figures on Channel crossings by illegal immigrants.

The move, we are told, comes after the UK Statistics Authority criticised the Home Office for the quality of data on Channel crossings in November because of a lack of transparency and accessibility. The Home Office response is to increase transparency and accessibility by publishing the figures every three months, in line with the publication of other official statistics.

Whatever the official justification, one cannot help but observe that this is one of the things that governments do when statistics become inconvenient. They simply stop publishing them. Sometimes, they make a vague attempt to offer a plausible excuse; mostly, they don’t bother.

That’s what they did in the early 1990s, in the wake of Edwina Currie’s 1988 salmonella and eggs scare. When she blamed eggs for the observed rise in food poisoning, the government imposed a wide range of controls on the egg industry. This should have cut the incidence but when, week-on-week, the figures obstinately continued to rise, they simply stopped publishing them.

Now, the Telegraph reports that there is pressure to do the same with the daily Covid figures, this time with slightly more justification.

It seems that “experts”, trailing after just about everybody else, have realised that the statistics are increasingly unreliable, and are calling for them to be scrapped “before people become addicted to them”. This is especially the case as data suggest that 70 percent of “Covid” patients in hospital are being primarily treated for other problems.

In support of this argument, the paper cites the latest statistics from NHS England which show that in the East Midlands, just 533 (29 percent) of the 1,817 people included in coronavirus hospital data were being treated primarily for the virus. For England as a whole, nearly half (47.9 percent) of Covid patients were admitted to hospital for other conditions, but also tested positive.

The situation is said to be exacerbated by an apparent divergence between the number of daily reported deaths and the registered Covid deaths recorded by the Office for National Statistics (ONS). It seems that so many people are now being diagnosed with omicron that a large proportion of natural deaths are ending up in the figures.

However, it has been the case for some time that ONS have recorded data separately on people who died “with” Covid, suffering from pre-existing conditions, and those who died “of” Covid with no other complications.

For instance, data for the year 2020 show that, of the 73,743 deaths (excluding line order anomalies) attributed to Covid, only 9,432 were recorded as due to Covid-19 with no pre-existing conditions (12.8 percent). The top two contributory causes were dementia and Alzheimer’s and diabetes. Heart disease, in various forms, features prominently.

It is worth noting, incidentally, that ONS reports that the average age of those whose deaths “involved” Covid-19 was 82.3 and for those “due to” Covid-19 was 82.5.

From these snapshots, it would be reasonable to infer that the daily published figures give a poor and highly distorted profile of the epidemic in the UK and, for that matter, its severity. The average age of death from Covid compares favourably with average life expectancy of 81.2 years.

Despite this, the Telegraph wheels out professor Balloux (not a parody), who does computational systems biology at University College London. In his view, “Until recently, monitoring deaths within 28 days of testing positive was a good proxy – it mirrored the real world”.

One is not sure whether the “real world” of Balloux is the same that the rest of us inhabit, but the revered professor now believes things have changed. “At the moment”, he says, “the numbers look horrible and worse than they should”.

The next “expert” wheeled on is professor Ball (singular), who does molecular virology at the University of Nottingham. He tells us that “It’s always been the case that the data was unable to discriminate between people in hospital, or people who died, because of Covid-19 rather than with Covid-19”.

That, of course, only applied to the daily figures, but Ball (singular), being the expert, is allowed to make a statement of the bleedin’ obvious, declaring that “It will be important to understand the future impacts of Covid-19 and the only way you can do that is to record numbers, just the same way we do with other important infections. But the assumptions and potential errors inherent in those figures do need to be acknowledged more”.

Now we are told that experts are also growing concerned that Britain is becoming “addicted” to the figures, on which basis they believe they (the figures, not the experts) should be phased out in the coming months.

For this we get more Balloux, who adds: “Part of the transition out of the pandemic is stopping people feeling so obsessed by case numbers and hospitalisations, because it’s not entirely healthy. We’re all a bit addicted to it”.

Amazingly, we are then informed that SAGE is “concerned” that the data is (sic) becoming difficult to interpret in real-time, because of the high community prevalence of omicron and changes in behaviour and testing.

This is a bit rich coming from SAGE which has had difficulty interpreting Covid data in any time frame, but at least it admits to “uncertainty” about current trends in the number of new infections, “particularly as a result of changes to testing policy and behaviours”.

Another statement of the bleedin’ obvious then follows, as SAGE intone: “An increasing proportion of these reported admissions are positive tests amongst people admitted primarily for reasons other than Covid-19, reflecting the very high community prevalence”.

Other scientists (so called) say it was important to keep publishing daily Covid figures to help keep track of the epidemic. However, they said more should be done to highlight the problems with the data.

Nigel Marriott, an independent statistician, seems to be one of these. He says: “The dashboard is an excellent tool so I think we should still use it, but people do need to understand what the data says and what it doesn’t say”.

Bizarrely, we then get Paul Hunter, professor in medicine at the University of East Anglia, telling us: “It does look like we are starting to see people dying with Covid but not because of Covid in the data, though the majority 75 percent of deaths are still because of Covid”.

Yet, ever since Covid has been around, people have been dying “with” but not “because of” the virus. Hunter, though, is referring to the figures for the week ending 7 January, when the government reported 1,282 deaths. By contrast, the ONS registered only 992, of which just 712 had Covid as the primary cause of death.

The “primary cause”, however, is a reporting artefact, reflecting the layout of the death certificate and completion protocols, and it is only in the quarterly figures that we begin to see the fine detail which gives a more accurate profile of the disease.

What has only been referred to tangentially, though, is the explosion of testing which accompanied the emergence of omicron, which has undoubtedly exaggerated (by comparison with previous figures) the scale of the epidemic – and spooked SAGE and “experts” such as prof “lockdown” Ferguson.

I am reminded, therefore, of the Dutch response to the salmonella epidemic of the late 80s and early 90s, when it introduced a charge for testing to identify the epidemic types, thereby ensuring that reports were minimised. At the time, it became the only major poultry producing country to show a downturn in human salmonellosis.

In all probability, therefore, those who want to see the daily Covid figures disappear have a point. We will not return to normal until the media are unable to obsess over the totals, sending out their messages of despair and despondency with every spike in the increasingly meaningless graphs.

Whether the government should stop the daily reports of illegal migrant flows, though, is another question. Here, we will not return to normal until the flows stop – the figures simply record the scale of the problem. With Covid, the figures have become the problem.