The Misguided Discussion of “Herd Immunity”
Needless to say, Boris Johnson having become an unintended participant in his Government’s earlier “herd immunity” strategy might seem to constitute sufficient evidence that this wasn’t a great idea. Admittedly, the UK switched course rapidly, but the damage of the initial barmy idea was done. As summarized in Fortune:
His government was pursuing a strategy that rested, at least in part, on the idea of shielding the most vulnerable members of the British public from infection while allowing a large percentage of others to catch the virus. The hope was that most of these people would experience relatively mild symptoms, recover, and wind up immune, stopping the virus’s further transmission.
But this kind of herd immunity, experts said, could require upwards of 60% of the population becoming infected. Herd immunity as a deliberate policy, epidemiologists said, is usually achieved through a vaccination program. It was untested as a tool for responding to a pandemic.
Epidemiologists and medical experts immediately assailed the plan as a dangerous gamble. And, when epidemiologists at Imperial College London, who had been advising the government on the likely spread of the virus, updated their models to take into account information on the number of hospital patients requiring intensive care in Italy, it became apparent that the minimal, voluntary restrictions Johnson had suggested were unlikely to save the NHS from being overwhelmed.
And the cost will be high. From the Guardian:
World-leading disease data analysts have projected that the UK will become the country worst hit by the coronavirus pandemic in Europe, accounting for more than 40% of total deaths across the continent.
The Institute for Health Metrics and Evaluation (IHME) in Seattle predicts 66,000 UK deaths from Covid-19 by August, with a peak of nearly 3,000 a day, based on a steep climb in daily deaths early in the outbreak.
The analysts also claim discussions over “herd immunity” led to a delay in the UK introducing physical distancing measures, which were brought in from 23 March in England when the coronavirus daily death toll was 54. Portugal, by comparison, had just one confirmed death when distancing measures were imposed.
Readers also felt compelled to weigh in on the barminess of this approach. From Dr. Larry B via e-mail:
The notion of herd immunity in the present context seems to have been advocated by Dominic Cummings, which after serious criticism he appears to have abandoned. Well, it shouldn’t have been advocated, in the given context, in the first place, as it was inapplicable.
If you wish to observe a herd’s immunity to a given pathogen, you allow it to be infected and then observe the course of the disease. Those left are either immune to the disease or robust enough to fight off the infection. This immunity of course applies only to the given pathogen and can’t be easily generalized. This is not usually a strategy adopted for human beings.
The term is usually used in the context of the introduction of a vaccine where you are creating, to the extent you can, herd immunity. You then observe how well the vaccine works. If the vaccine works, you have created herd immunity for a certain period of time, a period associated with the pathogen and the immune system’s response to it. It will be different for each pathogen. For standard flu, it appears to be about a year. For covid-19, no one seems to know.
To have a scientific government spokesman like Vallance support this dangerously absurd conception was a travesty. Cummings’ suggestions, and the basis of possible bullying of scientific advisors, stem, perhaps for the most part, from readings he has summarized in his blog, which shows ill-digested and sometimes incoherent summaries of things he has read. He is widely read, but it is unclear whether he understands what he has read. From the evidence of his blog and the things he has said in the period he has been Johnson’s advisor, it does not seem as if he understand much, if anything, that he has read.
As for herd immunity in the absence of a vaccine, what effectively you are doing is allowing a pathogen to cull the herd. This sort of thing has been advocated by eugenicists. I am not certain but it may be that Cummings is a believer in eugenics, a discredited theory of genetic purity of a population. There is a long and well developed literature discrediting eugenics theorizing. That Johnson and some of those around him listen to this kind of bullshit is disturbing, to say the least. Some are still working to this proposition, preposterous as that may seem.
Cummings’ understanding and apparent infatuation with complex systems and complexity, applied, say, to epidemics, does not seem to translate into coherent applications, simply because he doesn’t seem to understand them. The rest of the cabinet appear to be similarly ignorant and incompetent. And with Johnson out of the way, they also appear rudderless.
Another concept Cummings doesn’t understand, which he has used, is Schumpeter’s notion of creative destruction. I will leave that, though he contends that this concept is central to his approach to his ‘job’.
Cummings has also contracted coronavirus. Per the Daily Mail, he is still missing in action:
Dominic Cummings is yet to return to Downing Street after developing coronavirus symptoms and entering self-isolation more than a week ago.
Boris Johnson’s top aide has not been seen in public since before Monday March 30 when he put himself into isolation after getting symptoms over the weekend.
Number 10 has been insistent that Mr Cummings is working remotely but his continued absence is likely to spark further scrutiny of the health of key government players, especially after Mr Johnson was hospitalised with the disease.
There are growing concerns about the state of the Downing Street operation after numerous staff were laid low by the killer bug.
he PM’s other top adviser, Sir Eddie Lister, 70, has not been seen in public since the start of lockdown with his age putting him in an at-risk group. A number of other aides have also been off with symptoms.
Readers in comments yesterday also pointed out, as Dr. Larry did in passing, that it isn’t even clear how much immunity contracting coronavirus might confer. From Phacops:
You cannot make that presumption of immunity until there is clinical testing for immunoglobulin (G) serum titre and followup for reinfection. We are not anywhere capable of that in our for-profit medical system as is clearly evident by the fact that we are in our current situation by putting profit ahead of resilience of care for all our communities.
Even with the 4 most common coronaviruses that cause respiratory infections (the common cold) immunity is frequently weak and not long lasting. Magical thinking will not promote immunity. Rather, what this pandemic is demonstrating is that class/social inequalities in the lack of adequate provision of healthcare to all Americans creates the underlying health issues that make class and race factors in COVID comorbidity and risk of death. Want to recover from this and prepare for other zoonotic diseases in the future? Then, we need universal healthcare that is free at point of use to encourage use and prevention.
Note that findings are decidedly mixed. One study of SARS suggested that survivors had antibodies, and hence immunity, for three years, while those who had MERS appeared to have immunity for only a year.
And then there’s the question of mild or asymptomatic infections. There is evidence that at least some of those individuals may not have developed immunity. From Time:
A study on recovered COVID-19 patients in the southern Chinese city of Shenzhen found that 38 out of 262, or almost 15% of the patients, tested positive after they were discharged. They were confirmed via PCR (polymerase chain reaction) tests, currently the gold standard for coronavirus testing. The study has yet to be peer reviewed, but offers some early insight into the potential for re-infection. The 38 patients were mostly young (below the age of 14) and displayed mild symptoms during their period of infection. The patients generally were not symptomatic at the time of their second positive test.
In Wuhan, China, where the pandemic began, researchers looked at a case study of four medical workers who had three consecutive positive PCR tests after having seemingly recovered. Similar to the study in Shenzhen, the patients were asymptomatic and their family members were not infected.
Outside of China, at least two such cases have also been reported in Japan (including one Diamond Princess cruise passenger) and one case was reported in South Korea. All three of them reportedly showed symptoms of infection after an initial recovery, and then re-tested as positive.
In other words, these cases do not appear to be a combo of false and real positives.
Finally consider this observation from ZacP about the limits of antibody testing, now touted as the way to be sure of that someone contracted coronavirus:
It is important to note that an antibody test can mean different things depending on the specific infection. For example, a person who tests positive for the HIV antibody still also is a carrier of the virus and can infect other people. It can also take months, as many as six, for antibodies to become detectable via the standard labwork.
Combine that with false +/- results, length of time to ramp up production of testing kits, and we still need evidence to be able to interpret what a +antibody result would even mean in the case of COVID19…….serum testing still seems to be a long, long way out from actually being able to guide decision making.
A new article by Fierce Biotech confirmed these concerns:
With over two dozen different molecular diagnostics now authorized by the FDA for COVID-19, the field is beginning to see the first of a new group of tests aimed at screening for people’s immune responses to the disease and cataloguing past infections instead of active ones.
While overall testing capacity of any kind remains far short of meeting demand, antibody blood tests would provide additional data on the spread of the novel coronavirus, and results showing immunity could be used to give people an all-clear to leave quarantine and return to work.
Their accuracy would, of course, have to be paramount—any false positives could send unprotected people back into harm’s way. But researchers at the University of Oxford tasked with evaluating these serological tests say they’re still weeks away from solid validation and that no versions to date have performed well.
We see many false negatives (tests where no antibody is detected despite the fact we know it is there) and we also see false positives,” wrote Sir John Bell, the Regius Professor of Medicine at the University of Oxford and a government adviser on life sciences, in a university blog post.
“None of the tests we have validated would meet the criteria for a good test” described by the U.K.’s Medicines and Healthcare products Regulatory Agency, Bell added. “This is not a good result for test suppliers or for us.”
Bell stated a month is the bare minimum time before a test would be sufficiently improved and validated to be authorized for mass rollout.
And yet another bottleneck: a lack of staff to run tests on who is infected now. How many more would be needed to check for having had the coronavirus? From CIDRAP earlier this month:
Health experts from the Mayo Clinic to King County, Washington caution, that although testing is critical, it might not be possible at the levels needed to facilitate proposed plans. Even now, testing in most states is reserved for healthcare workers, or patients so sick they require hospitalization…
To confront testing material shortages, Pritt said Mayo has at least three different testing platforms to use, so if one runs out of a necessary reagent, another test can be used. Still, even with back-up plans, she said labs have to get creative….
For Jeff Duchin, MD, the limits of testing are only worsened by the limits of a stripped down public health workforce. Duchin is the health officer for Public Health Seattle–King County….
“A lot of plans proposed imagine unlimited testing capacity and instantaneous results, which would necessitate a public health army that doesn’t exist,” said Duchin. He said that in absence of a total Wuhan, China-style lockdown, containment of the virus would require a robust, boots-on-the-ground contact tracing effort that would require public health employees that no state or county currently has.
“Is the bang worth the buck in respect of the tremendous investment of time and energy needed to perform those tasks?” Duchin said. “If people are willing to fund and staff [them], it might be feasible if transmission dynamics of the disease make it a rational option. But in absence of a total lockdown, it’s too much work.”
Duchin said he sees serologic testing, which would detect COVID-19 antibodies in the blood, as potentially informing public health strategy. “With serology we would know what proportion of the population remains vulnerable, and which remain susceptible.”…
At the Minnesota Department of Health (MDH), health workers have dealt with a shortage of testing for weeks. Stephanie Yendell, DVM, MPH, senior epidemiology supervisor at the MDH, said the agency has started using other avenues to gain information about how the disease is spreading within state communities, and for that she’s using syndromic surveillance.
“We can’t be relying on testing moving forward to be the only indicator,” Wendell said. “We needed to have eyes on other indicators for illness, in order to put down on paper how all these pieces fit together.”
Yendell said most people with COVID-19 won’t seek medical care, because they experience only mild to moderate symptoms. For those people, self-reporting to existing surveillance websites is a way for public health departments to track community spread of the virus. Other patients will likely report only to outpatient clinics, where COVID tests are unlikely to be available.
“The power and drawback of syndromic surveillance is you can capture people who didn’t get a COVID test, or maybe, if they did see a doctor, their healthcare provider was suspicious of COVID but the visit was not coded as such,” said Yendell.
Now, she said MDH scientists are working on establishing which symptoms they want to track via electronic medical health records and user-interface websites.
Notice the considerable gap between the mainstream coverage of the idea that testing, particularly for immunity, would relatively soon become widely available, versus experts anticipating that it would be able to be done only at most to ascertain levels of infection and recovery in communities, and not to identify that individuals can work or travel safely. Look at how much the front-line public health officials discount the idea that testing will be the main source of information. Bear in mind that South Korea, held as the gold standard for testing, had actually performed as of mid March only 338,000 tests on a population of 51 million.
That does not mean that better tests won’t give very valuable insight. But there’s been way too much hope invested in science as a magic bullet. As Lambert pointed out early on, we’re still having to rely on 19th century methods more than 21st century ones.