Friday, August 14, 2020

Test = vaccine

"Cheap, frequent COVID tests could be ‘akin to vaccine,’ professor says" from the Harvard Gazette HT Miles Kimbal

Yes, I'm repeating myself, but maybe if we just try over and over again we'll get through. We could stop this disease now with tests. Vaccines are just a tool to stop disease transmission. Widespread, cheap frequent tests are just as effective a tool to stop disease transmission. So I'll keep quoting anyone who wants to say this! 

A Harvard epidemiologist and expert in disease testing is calling for a shift in strategy toward a cheap, daily, do-it-yourself test that he says can be as effective as a vaccine at interrupting coronavirus transmission — and is currently the only viable option for a quick return to an approximation of normal life.

“These are our hope,” said Michal Mina, assistant professor of epidemiology at Harvard T.H. Chan School of Public Health and Brigham and Women’s Hospital. “We don’t have anything tomorrow, other than shutting down the economy and keeping schools closed.”

....the paper-strip tests have already been developed and their shotgun approach to testing — cheap and widespread — provides a way back to the workplace, classroom, and other venues.

The tests, which can be produced for less than a dollar, can be performed by consumers each day or every other day. Though not as accurate as current diagnostic tests, they are nonetheless effective at detecting virus when a person is most infectious, Mina said. If everyone who tests positive stays home, he said, the widespread effect would be similar to that of a vaccine, breaking transmission chains across the country.

... What I would like to see happen is to start using testing [as] a true public health tool to break transmission chains in the same way that we know we can use masks to decrease transmission,” Mina said. “I want these tests to tell people they’re transmitting [the virus to others] at the time they’re transmitting, and [when] people can act on it because they’re getting immediate results. And I want them to take it every single day, or every other day.”

Several companies have developed such tests, Mina said.

Why aren't we doing this, voluntarily even? 

The Food and Drug Administration,..  has held up approval because the tests aren’t as accurate as nasal-swab, lab-based tests. While that would matter if they were intended as an individual diagnostic tool, Mina said that from a public health viewpoint, they are accurate enough to provide critical initial screening on a large scale. ....

“Everyone says, ‘Why aren’t you doing this already?’ My answer is, ‘It is illegal to do this right now,’” Mina said. 

In other words, the FDA says:  "Yes, you can use a thermometer to screen people out and send them home. Yes, you can use a questionnaire to screen people out and send them home.  No, you may not use a far more accurate $1 paper test for exactly the same purpose. And if you try, we'll ruin your company and send you to jail." 

Alex Tabarrok puts it nicely: We're testing for contagiousness, not for infection. 

President Trump seems to have discovered President Obama's phone and pen. A suggestion: Tomorrow morning, 9 AM, executive order: The sale and use of these paper tests shall be legal. We could be done with Covid 19 in a month or two. 


  1. Yes.

    But it's not just the bureaucrats who are risk averse, it's also the public that doesn't understand risk. Nobody has an incentive to explain that to the public, aside from a few kooky academics.

    If Mr. Trump legalized paper tests, it would be sabotaged and ignored.

    1. Many school districts are already being required to use testing-based metrics, usually at the country level, to decide whether to open schools. It wouldn't take much of a nudge to get these tests, which could targeted more narrowly to the school district community, to be used as the basis for these metrics. Sewer line tests, which are anonymous would be another less invasive way to track transmission.

  2. Every major public sector failure in this pandemic has been some variation on not being able to adjust levels of risk management to the balance of costs and benefits. Six months in, and we're still making the same mistake. So infuriating.

  3. Not sure how fast it would work, but approving these tests is a total no-brainer, and the FDA is without a doubt propelling the contagion by this idiotic regulation.

  4. Just because one researcher makes a claim in the press doesn’t make it so. I also research Covid-19 from a mathematical perspective and it is simply not true that test = vaccine. Yes testing is important and I agree that we need more of it with quick results. And I agree that bad regulation is a massive hindrance. We can interrupt transmission chains, but testing will not eliminate this disease. Not a chance. To eliminate a widespread communicable disease we need three ingredients: evidence of low incidence, high quality surveillance with rapid outbreak response, and high population immunity. Without a vaccine, the criteria of low incidence and high population immunity are mutually exclusive propositions.

    Sars-Cov-2 will not disappear from humanity at this point. It is too widespread, it is in all corners of the world. Maintaining and protecting a large susceptible population in the presence of a worldwide highly contagious pathogen is unstable. As soon as people let up on their “home pregnancy tests” it will explode again. It will re-enter the US sooner or later from an outside source. Unless you imagine Americans testing themselves every day for years and years, this strategy doesn’t come close to a vaccine. That is not to say high levels of testing are not useful.

    Please understand the logic here. Anytime you have a susceptible population in the presence of a globally dispersed highly contagious pathogen, the disease will find a way in. If we had this technology in the very early days of the Sars-Cov-2 in China, then maybe we could have eliminated it that way, but this strategy simply will not work as a vaccine substitute for the US population.

    Finally, if test = vaccine, why stop at Covid-19? Why not the flu? Or certain STDs? If we have vaccine substitutes, thinking just about Covid-19 seems a bit myopic. But alas...

    1. The point is not whether or not test = vaccine. That's irrelevant.

      "Condoms" are not equal to "vaccine" as far as AIDS is concern and that it’s not a good reason for the FDA to put condom approval on hold. They are not fully reliable either, and sure can create a "false sense of security".

      And still, it would have been extremely dumb to withhold condom approval on these bases.

      Difficult not to agree with Steamboat Lion: the "central planners" failure to understand risk-rewards and price-benefit tradeoffs for this virus is just infuriating.

      Give people the information required to do their own risk analysis and the tools to mitigate (not "to eliminate", you don´t need that) those risks (i.e. do you know the prevalence of the virus in your county?, why 6 months into this problem we don´t have a daily estimation of this all relevant variable on a county basis).

  5. It's strange how you put the onus for this decision on President Trump (who is neither a doctor nor an expert in infectious diseases), and not on the scores of medical experts, like Dr. Anthony Fauci, Director of the National Institute of Allergy and Infections Diseases.

    If legalizing the sale and use of these paper tests is such a smart, obvious decision, why isn't Dr. Fauci calling for it? He certainly has no difficulty making public statements about other policy decisions, like masks, goggles, and suppression of the economy. Authorizing and encouraging widespread at-home testing is a medical decision, and such decisions should be made by medical experts like Dr. Fauci and his colleagues at the CDC (which can grant emergency use authorization for at-home testing any time it wants to).

    Considering that Dr. Fauci is paid more than the President ($417,608 for 2019, compared to the President's $400,000 - Fauci also keeps the money while Trump donates it back to the Government), it seems reasonable to demand that he do his job, and not that the President do it for him.

    1. +1.

      If Dr Mina and Dr Cochrane are right, and there is a silver bullet solution blocked by bad federal regulation, then the medical experts given the authority to lead the virus response at the CDC such as Dr Fauci should be loudly advocating this.

      If the CDC experts aren't advocating this, I can think of three possibilities:

      1. This isn't nearly as great as it sounds.
      2. They need a few more days to evaluate this.
      3. They are terrible experts.

      Either way, the credit or blame for the medical decisions belong first with the CDC.

  6. "We could stop this disease now with tests. Vaccines are just a tool to stop disease transmission. Widespread, cheap frequent tests are just as effective a tool to stop disease transmission."

    Simply not in evidence. Want to understand how hard it is to get rid of this virus? Look at New Zealand floating in the South Pacific in splendid isolation athousand miles from anywhere. They thought they had wiped out the disease. But, then:

    This disease will be tamed by a vaccine just as every other viral disease has been tamed by vaccination. And, the vaccine will be available in record time

    1. "New cases on the remote island country raise questions about whether it’s possible for any country to keep out the virus." That video piece is just as bad as all the other media you decry. It just conforms to your worldview so it's OK I guess.

      I would have thought New Zealand illustrates the benefits of testing even for a country that has almost stamped out the virus. Authorities would find it far easier to track and monitor the virus if they knew who had it. NZ could have had a bunch of latent cases for weeks without knowing.

      Current COVID-19 vaccine trials suggest 50-70% efficacy. No better than a flu shot. If this one mutates every year like flu, it's never going away and vaccines will at best mitigate the damage. An alternative to daily tests for everyone would be frequent testing of sewer lines coupled with high intensity testing anywhere the sewer line tests flash red. Either would be a huge addition to the policy arsenal and could be used for more than COVID-19.

      We don't have to make a discrete choice. We can have tests AND vaccines AND masks AND quarantines AND social distancing and put them into place in varying degrees as circumstances warrant.

  7. Don't even think about random, voluntary testing either nationally or by state...

  8. Not quite what you had in mind, not as fast nor as cheap, but close:

    "Saliva-based coronavirus test funded by NBA, NBPA gets emergency authorization from FDA" August 15, 2020 11:56 AM ET By Zach Lowe |
    ESPN Senior Writer |

    The U.S. Food and Drug Administration issued an emergency authorization on Saturday allowing public use of a saliva-based test for the coronavirus developed at Yale University and funded by the NBA and the National Basketball Players Association.

    The test, known as SalivaDirect, is designed for widespread public screening. The cost per sample could be as low as about $4, though the cost to consumers will likely be higher than that -- perhaps around $15 or $20 in some cases, according to expert sources.

    Yale administered the saliva test to a group that included NBA players and staff in the lead-up to the league's return to play and compared results to the nasal swab tests the same group took. The results almost universally matched, according to published research that has not yet been peer-reviewed.

    The leading coronavirus saliva test, developed at a Rutgers University lab and given the same permission by the FDA in mid-April, costs individual consumers up to $150 -- though that can be reduced to $60 or $70 in some circumstances, said Andrew Brooks, an associate professor at Rutgers and chief operating officer of RUCDR Infinite Biologics, the lab behind the test. The Rutgers test can be taken at home and returns results in 24 to 48 hours.

    Several NBA teams used the Rutgers test in June, and Brooks said several sports teams are still using it. Those teams fly saliva samples to one of several labs -- including the Rutgers lab in New Jersey -- approved for administering the test, which adds time and cost.

  9. The Yale test funded by the league and players' union is simple enough to be used by labs everywhere provided they go through required accreditation processes, said Nathan Grubaugh, an assistant professor of epidemiology at Yale and one of two senior authors, along with Anne Wyllie, an associate research scientist in epidemiology, behind the saliva studies. Consumers dribble saliva into a narrow tube. Depending on the proximity of the lab, consumers could get results back within a few hours -- and definitely within 24 hours, Grubaugh said.

    The Yale test removes one cumbersome and expensive step -- the extraction of RNA from samples -- that is a core part of nasal swab tests and the Rutgers test. Scientists warned early in the pandemic about supply chain bottlenecks and shortages in equipment required to extract RNA.

    Extraction makes for a clearer and more certain result, according to both Brooks and Grubaugh.

    "(The Yale test) loses a little bit of sensitivity, but what we gain is speed and that it should be up to 10 times cheaper," Grubaugh said. The Yale test replaces the extraction step with the introduction of a reagent -- chemicals mixed with the saliva sample -- and a short heating process that releases the virus genome. The team found successful results using reagents that are commonly available, meaning labs everywhere could implement the Yale protocol, Grubaugh said.

    "My goal is not to test athletes," Grubaugh said. "That's not my target population. My target population is everybody. There were concerns about partnering with the NBA when all these other people need testing. But the simple answer ended up being the NBA was going to do all this testing anyway, so why not partner with them and try to create something for everyone?"

    The NBA, Yale and the players' association do not intend to take royalties from any use of the testing method, Grubaugh and others said. The NBA and union contributed more than $500,000 combined to fund the Yale work, sources told ESPN.

    * * *

  10. The genesis of the Yale-NBA partnership happened in early April, when Grubaugh and the Yale team published preliminary research indicating saliva tests conducted on coronavirus patients and health care workers were as accurate as nasal swab tests. ...

    When Yale released its initial findings in April, officials in the NBA league office and sports scientists across the league were calling labs and scouring literature for possible clues on how they might develop fast, cheap and easily accessible testing for players. ...

    One team official -- Robby Sikka, vice president of basketball performance and technology for the Minnesota Timberwolves -- came across the Yale paper and emailed Grubaugh.

    "We had a lot of strange requests, but this one was at the top," Grubaugh said. "I saw Timberwolves in the subject line and said, 'What the heck?'"

    The two connected. The research quickly reached the desk of NBA senior vice president David Weiss, the league's point person for coronavirus response ...

    "I was hesitant," Grubaugh said. "We do research. We are not developers of diagnostics. But this was an opportunity. They were willing to fund it. This is a crazy time for everyone anyway. I studied mosquitoes before this."

    With players returning to team markets in April and May, the league put out a call for volunteers to take saliva tests -- for the purpose of comparing results to the nasal swab tests the same group would also take. The results showed close to universal agreement between tests, according to Yale's research.

    * * *

  11. See, it pays to follow the sports news.

  12. "Though not as accurate as current diagnostic tests. . ."

    Wow. Talk about burying the lede.

    Accuracy is the whole ballgame, particularly with low-prevalence SARS-CoV-2.

    It seems strange for an economist to not instinctively realize that bad testing is worse than no testing.

  13. Sure, there is a trade-off between accuracy and "easiness of use + speed of results".But, the relevant variable is the transmission rate, and it can be lower with a "less accurate-easier to use-faster results" test.

    In any case, this is a very "static" analysis. Once these tests are out there and different providers are competing to develop the best ones, accuracy and easiness to use will improve fast.

    Performance improves by doing, not by writing academic papers on "Accuracy vs widespread use: an econometric analysis for Covid-19 testing"

  14. "'If everyone who tests positive stays home,' he said, 'the widespread effect would be similar to that of a vaccine, breaking transmission chains across the country.'" -- quoting Michael J. Mina

    In that single sentence, quoted in the blog post, lies the difficulty. "If... ." If all persons venturing outside the family home were to wear masks..., etc. ...the economy could return to "normalcy". Or, some semblence of "normalcy". A nice thought, but it turns on the doctrine of universal conformance to some ideal state of behaviour, a doctrine that is not universally held to in the U.S. voluntarily. "Life, liberty, and the pursuit of happiness", is the basis of the social construct in the U.S.

    Micheal Mina is pinning his hopes on three assumptions: (a) universal availability of an inexpensive test ("paper strip"), (b) self-administration of that test daily or every other day, and (c) self-isolation in the event of a "positive" test result. What is the likelihood of all three assumptions being true? Firstly, we calculate the probability of all three events occuring for a representative individual P("a" AND "b" AND "c") = P("a")P("b")P("c"), where P(x) is probability that "x" occurs. If one out of the three probabilities is zero, the resulting probability is zero; if one of the three probabilities is non-negative, but a low fraction, then the resulting probability is a low fraction even if the other two probabilities are unity or a high fraction.

    The likelihood of universally testing 350 million individuals on a daily or every other day basis, has to be seen as unlikely. Even if the likelihood of a positive test in the case of an infectious individual is high, say 80%, the proposed 'solution' can fail if individuals testing 'positive' do not (or cannot) self-isolate for any reason (economic, or home situation, etc.).

    ... continued below...

  15. ...continued from above...

    The paper "Test sensitivity is secondary to frequency and turnaround time for COVID-19 surveillance", pre-print, D.B. Larremore, et al., medRxiv (June 27, 2020), uses computer simulations to demonstrate the 'efficacy' of a second-best test procedure. The paper does not delve into the behavioural or socio-political aspects, but relies on two scenario constructs, (i) a 20,000 closed 'complete-mix' population with 1/N infectious agent leakage into that population, and (ii) a closed 8-million population "agent-based model" with an initial seeding of 100 infectious agents, to demonstrate the efficacy of the testing protocol with test frequencies ranging from daily to every third day to weekly to bi-weekly, at two levels of detection--10^3 and 10^5 cp/ml. Scenario (i) is a standard SIR model with a behavioural element--i.e., self-isolation on occurrence of a 'positive test result' which reduces the number of infectious individuals active in the population. Scenario (ii) relies on a model constructed based on events earlier this year in the City of New York. The scenario (ii) model is therefore a retrospective 'what-if' scenario reliant on perfect foresight--i.e., what might have been if we had the test and applied it daily or every third day, etc., to a population of 8 million souls. As computer models go, it is a 'blockbuster' in terms of modelling skill and computing power.

    What we know from real-time behavioural studies and anecdotal reports of behaviour observed since the various state governments have lifted 'lock-down' measures is that people, young people in particular but older people as well, are less likely to follow voluntary recommendations or rules in the presence of the virus even when they know what the consequences are. This has resulted in resumption of the 'first-wave' of infections across a broad swathe of the states. We must ask ourselves just how the protocol that Michael Mina is proposing will be implemented and whether it will be as effective as the computer simulations reported in the paper he co-authored with D.B. Larremore, et al. Colour me sceptical

  16. Any test for infection shows that the infected person has been shedding virions and infecting others for a while before the test detected it. Tests are a foolish diversion. But Mina shows his bias with "We don't have anything tomorrow, other than shutting down the economy and keeping schools closed." There is a huge continuum of choices. Shutting down the economy will surely kill more people than this virus - which is on track to match the 1968/1969 Hong Kong flu epidemic. Remember? Maybe not, because - even though it killed 150,000 of 200 million Americans - we didn't undertake foolish lockdowns or other horrors. Masks generate a lot of fear and strife but have no measurable effect on the course of this virus.

    In fact, there's lots of evidence that a fourth of the population already has significant immunity, and if another 20% is infected and recovers, the virus burns out. Even on cruise ships and in nursing homes, no community has exceeded 47% prevalence. Immunity in those most likely to be infected - supermarket, food service, and transportation workers - creates effective herd immunity.

    The correct answer is to protect nursing homes and hospitals, and for at-risk people (the elderly, obese, people with heart disease) to self-isolate. The rest of us should just go about our business.

  17. Viruses live on, some beyond vaccines, because vaccines are not effective 100% or the time. Get used to living with viruses. The hope is for effective, and cheap cures. The FDA did serious damage to curing, if not reducing the strength, of viral infections, scotching the use of chloroquine. Another problem is the fear public health caused by citing the potential of millions of deaths, making entire nations cocoon. By their mistakes, the WHO, the CDC, the FDA, and the science/opinion of Dr. Fauci, resulted in an unnecessary shutdown of economies, and social/political stability.

  18. It's amazing how as more information comes in, policy becomes more entrenched in protecting the current practice

  19. Frame the question differently. Recognize the fact that elimination of the virus is not possible. How can you reduce the death rate from viruses like Covid 19? Answer: Make available medications that attack/cure the illness widely available, e.g. over-the-counter. The damage/deaths caused by the CDC edict against cloroquine is/was a killer. Why? Misplaced grant ofpower to government? Collusion of interests with vaccine/medication industry funding CDC? Cloroquine cost pennies per pill. Approved ant-viruses medications cost thousands of dollar. For sure, it was a mistake, and tragic.

  20. Will testing = behavior modification? Maybe. There are still people out there acting irresponsibly, not wearing masks (I see this everyday) and engaging in high risk behavior. This only makes things worse and increases the chances for super-spreader events. Employees are getting assaulted for enforcing mask wearing before people can be served as consumers - can we act with civility? Enforcement is critical right now and until we get serious about enforcement, no amount of testing will matter. Impose fines and make it hurt real bad. People may hate radical changes that affect their choices available, but it's truly necessary right now. Maybe all of this is Darwinism in action, but people need to understand they're putting others at risk by not acting responsibly. This virus mess has exposed how small pockets of irresponsibility can destroy lives and the economies that serve them.

  21. Hope you’ve seen:

  22. This debate turns on two things. First, are the mass of people capable of carrying out directions well enough to achieve vaccination levels. Given the higher risk reward ratios yet low individual consequences involved, as shown in comments, there are a good deal of grounds to be sceptical of success.

    Second, do we have any proof of concept that demonstrates that this theoretical solution can be practiced? For example, take a higher IQ polity, a nation like Singapore (several millions), or possibly Israel, implement and show that this can be done.

    As much as I love Johns reasoning that a technical and simple fiction can succeed, I share the carpers doubts. So, persuade me!

    John, please take your argument to decision makers to Singapore!

  23. Just picture asking people to spit on a strip and wait 10 minutes for the line to show up. How many will actually keep track of time? How many will check in 5 minutes when the signal is too faint? How many will in 20 minutes and get a false reading?
    How many will decide they just don't want to bother?

  24. We don’t need perfection. We just need to drop R below 1.00. Let me buy paper tests and I will use them wisely. So will many other people. Paper tests will push R down, saving thousands of lives.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.