Oh dear lordy be, seriously?

Perhaps spending a fortune on tests the accuracy of which is unknown is a bad idea. Not because the supplier makes a profit but because the results can’t be trusted.

If half of cases are asymptomatic, appears to be the case, it clearly isn’t searching for needles in a haystack. Then again, if the failure rate (false positives) of the test is high enough, fake cases will dwarf real ones and we will stay in lockdown forever by a panicked government.

Pollock may well be that, Theo. But she’s not wrong this time, is she?

Chatting via Whatsapp to a friend who works for PHE. They had a talk by Patrick Vallance yesterday, where he talked about trying to get infectious people out of the community; Moonshot testing is aimed at finding those people who are infectious so they can remove themselves from public life for the period of infectiousness, rather than just finding people who have it.

njc – I’m not an epididdyman or nothing, but Covid is endemic and public health authorities are incompetent.

So this sounds like it wouldn’t really work. Test and trace is a fiasco, what would make this different?

Re: treatments, there are awesome treatments now available for Chinee Flu. Donald Trump (pbuh) is living proof. If you’re not incredibly fat or barely clinging to life as it is, you have very little to fear from Covid.

But lots to fear about our £2Tn national debt and the ongoing Dresdening of the economy.

Last week another huge trache of my colleagues were furloughed. Many of them won’t be coming back. These are good people, clever people, with those skills in “Cyber” the government keeps telling ballet dancers they should learn. In light of recent racial kerfuffles it’s sadly ironical many of them are “BAME” – and not the imported Indian kind either, smart black and Asian British blokes who worked hard to become engineers and techies. Tossed on the scrapheap because private sector demand has collapsed and the business is haemorrhaging cash.

Now some of them can’t afford to pay their mortgages because of a cough with a 0.028% mortality rate. Great work, Boris!

Testing the asymptomatic has never made any sense other than to politicians who use said ‘positive cases’ to support their illogical and draconian rules, to the ‘experts’ who need as many cases as possible to confirm their insane predictions and to the MSM who relish a constant source of cheap content.

I would test positive for smallpox, a disease with a 30% infection fatality rate, but have no symptoms, sickness, or ability to infect anyone with it. Positive case though, so isolate!!!

The lunatics etc.etc.

“But what’s the point of trying to find asymptomatic Covid cases?”

Purely scientifically… We still need to know the % of people that do get infected, but don’t develop symptoms because their body fights off the virus within the usual incubation period.
This will also contain the fraction of people where their mucal layer do their proper job, and trap the virus particles before they can do any harm at all.

Of course, that’s for the current definition of “CoVid cases”, where a positive test means you’re considered Deatly Ill With The Plague, regardless of whether or not the virus has actually managed to take root, you’ve developed actual symptoms, and whether or not you are actually contagious to others. (three different stages which all get conflated into one big mess by the Powers that Be and the Panick Pricks…)

As scientific data , the current CoVid test results are utter garbage, because there’s no follow-up, no baseline checking, no controls, and no common definitions of cases.
It’s a political excercise in obstinacy and idiocy.

I have a black belt in Karate and have done three marathons and an ironman.

Was he a thick.racist.prick

I wonder if she’s one of the Statists who point at Germany, South Korea and and Taiwan and complain that we should be like them, not noticing that they use decentralised and privatised systems both for health care and test and trace?

@njc

“Chatting via Whatsapp to a friend who works for PHE. They had a talk by Patrick Vallance yesterday, where he talked about trying to get infectious people out of the community; Moonshot testing is aimed at finding those people who are infectious so they can remove themselves from public life for the period of infectiousness, rather than just finding people who have it.”

Yep this is the rationale and it’s quite a different one from the current testing regime, and indeed the contact-tracing idea. Expect to see more widespread rapid testing at schools, unis, large workplaces and travel hubs in the near future, as well as perhaps some repeats of the Liverpool trial in other hotspot towns/cities if that goes well (logistics is the main challenge with it).

For the sharp-eyed, this is an idea which has been around since early in the epidemic – they even bought millions of rapid tests early on before evaluating them as useless. There have been trials going on – very quietly since that early failure – and PHE are now satisfied the current iteration of the test kits has sufficiently low false positive rate and false negative rate to be useful for what they’re tasking it with. So they’re aware of the potential criticism from Prof Pollock et al but they think this kit is the real deal. Though I suspect for population screening (as opposed to use in settings with unavoidable crowding and contact) they’ll reserve it for the worst-hit areas only.

It depends what the test tests for. Knowing how many people have had the flu and recovered (anti-body test) would be useful in calculating how far away from herd immunity we are. Having a more accurate estimate of IFR ditto.
But if the test is unreliable then indeed it’s worse than useless.

@PF

“You don’t need to upscale generally unless you’re serious about eradicating”

No that’s not right. Read @njc for the PHE view on why they’re doing it (his information sources tally with mine).

Even the original “herd immunity strategy” (name disputed) required having some measures available to control the spread – they were hoping they could allow it to wash through the population but at a controlled rate, so it didn’t threaten to breach healthcare capacity. So even though there wasn’t a total lockdown initially they still brought in additional hygiene rules and soon closed sports events and concerts. If the curve continued growing to the point the capacity forecast warning lights were flashing, the idea was to phase more restrictions in. Attempting to control the spread isn’t synonymous with following an elimination strategy.

Broadly there are several ways to control spread. Reduce contact rates between people (stay at home orders, close schools etc), reduce the probability contact results in transmission (this is the concept behind the 1m or 2m rule, masks, handwash stations etc) and isolating potentially infectious people from the population (this was the point of contact tracing and will be the point of these rapid tests). The idea is that the better you can do the last point, the more liberal you can be opening up the economy on the first point, without risking trabsmission levels get out of control.

I’m just explaining the rationale. We’ll see how well it works in practice. It isn’t about elimination (the test isn’t sensitive enough for that and it probably wouldn’t be logistically possible to test regularly enough for that) and it is sod all to do with estimating IFRs.

@Ted S

What you’re saying is irrelevant for the UK. One thing the authorities here do deserve credit for is fairly quickly setting up a large-scale random survey to measure spread of the virus through the population.

https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/6november2020

I believe other countries are still struggling to figure out what their spread is, because they’re relying on the daily test results and this is being confounded by capacity (just how many tests are being offered and – importantly – whether they are being targeted in worst affected areas) and also by who is allowed access to a test (eg which symptoms you need to show, which varies from place to place and sometimes gives extra leniency eg to healthcare workers)..

As a result, countries like that have to make an educated guess what’s going on in terms of transmission by looking at measures like the positivity rate (which may be more sensible than looking at positive tests alone). You’re effectively fighting while blindfolded. In the UK we haven’t had to do that for months now, fortunately (except in respect to local outbreaks, the national survey is only finely grained enough to spot regional trends so we still rely on local test results to figure out what’s happening on a town by town basis).

It has never been part of PHE’s gameplan to test fewer people (they’ve been continually aiming to expand capacity and the most common complaint is they haven’t done so quickly enough), and it’s clear the government here is reticent about lockdowns (some would argue not reticent enough but they’re clearly not desperate for any excuse to launch one – they invested sufficient political capacity in avoiding one when their advisors told them this would cost lives, came under sustained pressure from opposition parties, and Lockdown 2 required a very embarrassing U-turn). This “moonshot” project of large-scale rapid testing was the government’s main hope for avoiding future lockdown so it has probably arrived about a month too late for them. Ironically it’s taken this long due to the accuracy trials it was being put through, which is exactly Pollock’s concern.

MBE

Fair enough, I take the point, but I simply don’t accept the rationale. Of locking down healthy people etc – I won’t go into detail, we’ve done it to death often enough on here already. And it’s never been about health capacity. If it (really) was, virtually none of the enforced stuff would have happened (once past one or two weeks into the 1st lockdown had been properly analysed, if we are still pretending we didn’t know at that stage), you are right, the voluntary advice was good (and worked throughout).

“searching for symptomless yet infectious people”: I’ve not seen any claim that the test reveals who is infectious.

Come to that I’ve seen no persuasive evidence that anyone knows the false positive rate.

@A

The number of false positives is hardly going to be affected by the prevalence in the population, because even in the worst towns/cities 90%+ of people won’t have the disease (might be different in smaller, more specific settings, e.g. food processing plant that’s had an outbreak). The thing that changes more markedly is the ratio of false positives to true positives. This can mess up the overall cost-benefit ratio of your intervention. But in terms of overwhelming the contact-tracing service, it’s the total number that matters. They’ve done enough trials of the new test that they reckon the false positive rate is manageable, and do intend to do contact-tracing on everyone who tests positives in the rapid testing in Liverpool. We’ll see how well it copes.

The thrust of what you’re saying is correct, which is why it seems the use of tests for mass population screening is going to be limited to localities with high rates. If we ever did a Slovak-style national screening, extensive contact-tracing would clearly be impossible. But expect them to pop in settings deemed high risk for transmission (e.g. universities, food factories) across the country.

@PF

Yeah you’re right that hospital capacity didn’t drive everything, I should rephrase that. The big calls over both lockdowns seem very driven by fear over hospital capacity. In between, the UK mostly seems stuck in a kind of holding strategy (or no strategy at all?). Sturgeon wanted to go all-out for the elimination option, but Westminster pushed back against that. But they didn’t revert to their earlier plan of controlled spread either – they have been optimistic about a vaccine for some time (from what I’ve gathered from my contacts “on the inside” – certainly more than they’ve generally admitted publicly, lest they be accused of wishful thinking), so perhaps they didn’t think the deaths from controlled spread would be worth it, or politically acceptable.

I might be being harsh to say a holding strategy seems like no strategy at all. It does buy time for development of vaccination and treatments, getting test-and-trace and smartphone app systems together (here, on the contrary, optimism about their potential effectiveness has nosedived, so nobody serious seems to be suggesting these systems can “take the strain” in the R-budget of substantially relaxing other restrictions, which had been the initial hope), and more research into properties of the coronavirus we didn’t have great understanding of (how is it physically spreading from person to person, how long might immunity last etc). But the lack of a visible exit strategy was a common complaint from various quarters for some time – opposition and backbench Tory MPs, businesses, the devolved governments (Sturgeon’s idea of elimination was absurdly high cost and would have serious long-term complications for ever opening the UK up again, but at least it would have been a clearly defined goal).

The localised “tier” system could easily have been implemented much earlier, in fact would have been the logical replacement system to bring in after Lockdown 1: it didn’t require much work technically/legally, came in too late, and ended up being abandoned before its effects could show. The ONS survey figures that came out this Friday did indicate a decline in the growth rate but too late to bolster the case of sticking with tiers and avoiding Lockdown 2. Moreover, I’m not sure the goals of the tier system were ever clearly set out – if, as publicly stated, it was to avoid future lockdowns altogether, rather than merely postpone the time the next one was called, they should have been used more proactively to control the spread of infection, rather than simply letting figures drift up until the tightest measures were brought in. Using the tiers as a local emergency response rather than a brake on growth meant that capacity forecasts were allowed to drift into the danger zone (particularly since the effects of tiers could not yet be evaluated, so any slowing effect couldn’t be incorporated into the models!), at which point national measures became inevitable. Even if the tiers had been brought in a few weeks earlier, I wonder if it could have made a difference – at least letting its effectiveness be judged in the worst areas, which if successful would have made future healthcare capacity forecasts less alarming and bolstered the case for holding off on Lockdown 2.

Re the wind-up of Lockdown 1. Patients in hospital peaked in early April (see https://coronavirus.data.gov.uk/details/healthcare) but the curve only flattened out just before the peak, so with the noise in the data it wasn’t entirely clear it really had been the peak until a bit later. The ONS survey data only starts from May so until then there was a lot of uncertainty re how much transmission was going on; daily testing data had many limitations especially while its capacity was being expanded. Overall, infections, hospitalisations, and deaths seemed to decline much more slowly here than many other European countries, which led to persistent caution about lifting restrictions or relying on voluntary measures. The fear among advisors was if we were only just keeping R below one with all the things we had in place, there didn’t seem much scope to loosen up without triggering another upsurge. Particularly because schools and (probably) universities were expected to restart in September and this alone was expected to consume a big chunk of the R-budget. Indeed the second peak seems to have been driven initially by a rise in infections among young people. This doesn’t necessarily justify the policy. Just stating my understanding of what the thinking at the top was.

As you say, purely about health capacity

“Re the wind-up of Lockdown 1. Patients in hospital peaked in early April (see https://coronavirus.data.gov.uk/details/healthcare) but the curve only flattened out just before the peak”

That’s fair, I did notice that, it was clear even when looking at that data back latter half of April.

“so with the noise in the data it wasn’t entirely clear it really had been the peak until a bit later”

Except that since then we know there was also solid data available from 111 call trends, which was an earlier proxy. I’m struggling to believe that any half competent analysts on the payroll didn’t in reality have a good handle. Actually quite simply, I don’t believe it. And if they really were clueless and thought come early April (my one to two week excuse for them above, absolute tops) that the whole thing was still flying through the roof – then, they weren’t competent, and sack the lot of them and start again.

“Just stating my understanding.”

Fair enough, I get it.