Testing a Random Sample

Finally. Somebody stating what is obvious to me. From The Hill:

John Ioannidis, a Stanford epidemiologist who is famous for debunking bad research, has been pushing for it. He told me that random sampling is needed and could be done with a couple of thousand tests. When I told him that I previously worked in the polling industry, he put it in terms that resonated with me. He said, “Random representative testing is like polling. We run thousands of opinion polls in this country. We should similarly get a representative sample of the population and get them tested. It is just so easy.”

A recent television interview with Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases and member of the White House Coronavirus Task Force, underscores the need. After estimating that 100,000 to 200,000 Americans could die of the coronavirus, he said that projections are a “moving target” and that models are “only as good and as accurate as your assumptions.” But how good are models if the data is insufficient?

Ioannidis warned of a potential evidence fiasco in a recent op-ed for Stat. He wrote, “The data collected so far on how many people are infected and how the epidemic is evolving are utterly unreliable. Given the limited testing to date, some deaths and probably the vast majority of infections due to SARS-CoV-2 are being missed.”

Eran Bendavid and Jay Bhattacharya, also professors at Stanford, echoed that concern in The Wall Street Journal, writing, “The true fatality rate is the portion of those infected who die, not the deaths from identified positive cases.” They speculate that due to how infectious the coronavirus appears to be, and because tens of thousands of people traveled from Wuhan to America in December, millions of Americans could have been infected.

Random sampling will tell us what percentage of the population has the coronavirus and its lethality. Only testing the very sick skews mortality rates and leaves us in the dark about how many Americans are unknowingly walking around asymptomatic or with mild symptoms. Looking at other countries’ data also has its challenges; age structures, climate differences, quality of health care systems and testing all vary.

New York State has administered nearly a quarter million COVID-19 tests, by far the largest number per 1M population of any state. Nearly two-thirds of its cases are in the counties surrounding New York City. Administering just a few thousand of those tests to a random sample of people in the Bronx, Queens, Manhattan, Kings County, etc. would have probably been more productive than the vast majority of those tests. Epidemiological testing is long overdue.

I have a question for physicians. In the absence of a treatment for COVID-19, how does testing a patient with the symptoms of COVID-19, change how you treat that patient? I doubt that it actually does. I think the same supportive care would be provided, as available, whether the test was positive or negative.

20 comments

Pernicious

What conclusions should we draw from this account of the story of testing for COVID-19 in the United States, published in the Houston Chronicole? I think it’s summed up pretty well in one quote, from an exasperated researcher:

“The most pernicious effect of the current regulatory environment is that it kneecaps our ability for preparedness should a true emergency emerge,” Greninger wrote to colleagues on Feb. 14.

or maybe the conclusion of the article:

On March 12, Fauci, who runs the National Institute of Allergy and Infectious Diseases, told lawmakers the problem was not simply the failure of the CDC test. The coronavirus testing debacle had exposed deep structural problems in the nation’s public health system, he said.

“Yeah, it is a failure, let’s admit it,” he said. “The idea of anybody getting it easily the way people in other countries are doing it, we’re not set up for that. Do I think we should be? Yes, but we’re not.”

Bureaucracts and political appointees may have succeeded in killing thousands of Americans.

14 comments

Test Sewage

I found this article in Scientific American fascinating:

More than a dozen research groups worldwide have started analysing wastewater for the new coronavirus as a way to estimate the total number of infections in a community, given that most people will not be tested. The method could also be used to detect the coronavirus if it returns to communities, say scientists. So far, researchers have found traces of the virus in the Netherlands, the United States and Sweden.

Analysing wastewater—used water that goes through the drainage system to a treatment facility—is one way that researchers can track infectious diseases that are excreted in urine or faeces, such as SARS-CoV-2.

One treatment plant can capture wastewater from more than one million people, says Gertjan Medema, a microbiologist at KWR Water Research Institute in Nieuwegein, the Netherlands. Monitoring effluent at this scale could provide better estimates for how widespread the coronavirus is than testing, because wastewater surveillance can account for those who have not been tested and have only mild or no symptoms, says Medema, who has detected SARS-CoV-2 genetic material—viral RNA—in several treatment plants in the Netherlands. “Health authorities are only seeing the tip of the iceberg.”

The TL;DR version of the article is that maybe we should put less emphasis on testing individuals for COVID-19 and more on testing their sewage. Something to think about.

4 comments

After COVID-19

Henry Kissinger weighs in on the COVID-19 outbreak in an op-ed in the Wall Street Journal:

The coronavirus has struck with unprecedented scale and ferocity. Its spread is exponential: U.S. cases are doubling every fifth day. At this writing, there is no cure. Medical supplies are insufficient to cope with the widening waves of cases. Intensive-care units are on the verge, and beyond, of being overwhelmed. Testing is inadequate to the task of identifying the extent of infection, much less reversing its spread. A successful vaccine could be 12 to 18 months away.

The U.S. administration has done a solid job in avoiding immediate catastrophe. The ultimate test will be whether the virus’s spread can be arrested and then reversed in a manner and at a scale that maintains public confidence in Americans’ ability to govern themselves. The crisis effort, however vast and necessary, must not crowd out the urgent task of launching a parallel enterprise for the transition to the post-coronavirus order.

Ever the consummate courtier. Much of the op-ed focuses on what should happen after the crisis is over:

Drawing lessons from the development of the Marshall Plan and the Manhattan Project, the U.S. is obliged to undertake a major effort in three domains. First, shore up global resilience to infectious disease. Triumphs of medical science like the polio vaccine and the eradication of smallpox, or the emerging statistical-technical marvel of medical diagnosis through artificial intelligence, have lulled us into a dangerous complacency. We need to develop new techniques and technologies for infection control and commensurate vaccines across large populations. Cities, states and regions must consistently prepare to protect their people from pandemics through stockpiling, cooperative planning and exploration at the frontiers of science.

Second, strive to heal the wounds to the world economy. Global leaders have learned important lessons from the 2008 financial crisis. The current economic crisis is more complex: The contraction unleashed by the coronavirus is, in its speed and global scale, unlike anything ever known in history. And necessary public-health measures such as social distancing and closing schools and businesses are contributing to the economic pain. Programs should also seek to ameliorate the effects of impending chaos on the world’s most vulnerable populations.

Third, safeguard the principles of the liberal world order. The founding legend of modern government is a walled city protected by powerful rulers, sometimes despotic, other times benevolent, yet always strong enough to protect the people from an external enemy. Enlightenment thinkers reframed this concept, arguing that the purpose of the legitimate state is to provide for the fundamental needs of the people: security, order, economic well-being, and justice. Individuals cannot secure these things on their own. The pandemic has prompted an anachronism, a revival of the walled city in an age when prosperity depends on global trade and movement of people.

Sadly, I think it’s superficial. Is is possible that people have little confidence in our institutions because of their repeated failures? And that failure along a single plane—the benefit of those in charge. What I think we have witnessed is a panic among political leaders and what we should have learned is that we cannot afford to allow supply chains for vital goods to run through large, powerful authoritarian countries. It isn’t even that difficult. All that would be required in the case of health care, for example, is for the only goods eligible for compensation through Medicare or Medicaid would need to be made in the United States from components with completely American supply chains. Strict liability for fraudulent claims. Yes, it would increase costs and probably cut margins. You’d probably be surprised at how quickly production can be repatriated.

Enlightenment values do not require us to sacrifice our populations and our security on the altar of globalism.

13 comments

Test Or Bust

In op-ed at the Wall Street Journal Paul Romer and Rajiv Shah highlight the importance of testing:

For now, social distancing is the best America can do to contain the Covid-19 pandemic. But if the U.S. truly mobilizes, it can soon deploy better weapons—advanced tests—that will allow the country to shift gradually to a protocol less disruptive and more effective than a lockdown.

Instead of ricocheting between an unsustainable shutdown and a dangerous, uncertain return to normalcy, the U.S. could mount a sustainable strategy with better tests and maintain a stable course for as long as it takes to develop a vaccine or cure. The country will once more be able to plan for the future, get back to work safely and avoid an economic depression. This will require massive investment to ramp up production and coordinate the construction of test centers. But the alternatives are even more costly.

Two types of testing will be essential. The first test, which relies on a technology known as the polymerase chain reaction, or PCR, can detect the virus even before a person has symptoms. It is the best way to identify who is infected. The second test looks not for the virus but for the antibodies that the immune system produces to fight it. This test isn’t so effective during the early stages of an infection, but since antibodies remain even after the virus is gone, it reveals who has been infected in the past.

Together, these two tests will give policy makers the data to make smarter decisions about who needs to be isolated and where resources need to be deployed. Instead of firing blindly, this data will let the country target its efforts.

We’ve got to do more, faster. Local, state, and federal governments need to clear the path and use the powers at their disposal to accomplish more testing. The first priority will necessarily be health care workers. Then onwards to the whole population.

Waiting six months to mobilize such an effort is far too long. We’ve got days or, at the most, weeks, not months and years.

15 comments

Why Hasn’t Amazon Already Done It?

Close but no cigar. The editors of the Wall Street Journal have an idea:

Their respective comments highlight how difficult it is for leaders to make decisions when they still lack answers to basic questions. One response might be for the federal government to develop a website where the states update their numbers, outline the policies they have put in place, and note the results.

That would let people and public officials find out what’s happening in, say, Louisiana; where other states are on the infection curve; what is being done to flatten it—and give people a better sense of what their unmet needs are likely to be. Precision is inherently difficult, but a federal website could be a shared clearinghouse for shaping policies informing the public.

Many early projections about Covid-19 deaths were based on worst-case scenarios if governments did nothing. But state governments have taken many different steps. It will take more testing and tracing to know for sure, but the more information we have the more we can close the gap on dramatically different estimates and learn what works and what doesn’t.

Right idea. Wrong strategy. If undertaken by the federal government, the crisis would be long over before the site came up. The federal government simply does not have the human resources to accomplish it. That’s not what it does.

What it does is let contracts for private corporations, drawn from a list of qualified vendors, to do things. The contract process alone will take months.

Amazon, Google, or Facebook have the know-how, resources, and the ability to have such a site up in days. Amazon, in particular, is making millions in additional sales. C’mon, Jeff, do your bit. Set up a clearing house for state and local governments to pool their information.

6 comments

Reminder

The difference between a state governor and the President of the United States is that for a president blaming the federal government is not a satisfactory response to problems. At least not without a workable plan for changing the situation.

0 comments

Accidents Do Happen

David Ignatius is jumping on the “it came from a bioweapons lab” bandwagon. From his most recent Washington Post column:

To be clear: U.S. intelligence officials think there’s no evidence whatsoever that the coronavirus was created in a laboratory as a potential bioweapon. Solid scientific research demonstrates that the virus wasn’t engineered by humans and that it originated in bats.

But how did the outbreak occur? Solving this medical mystery is important to prevent future pandemics. What’s increasingly clear is that the initial “origin story” — that the virus was spread by people who ate contaminated animals at the Huanan Seafood Market in Wuhan — is shaky.

Scientists have identified the culprit as a bat coronavirus, through genetic sequencing; bats weren’t sold at the seafood market, although that market or others could have sold animals that had contact with bats. The Lancet noted in a January study that the first covid-19 case in Wuhan had no connection to the seafood market.

There’s a competing theory — of an accidental lab release of bat coronavirus — that scientists have been puzzling about for weeks. Less than 300 yards from the seafood market is the Wuhan branch of the Chinese Center for Disease Control and Prevention. Researchers from that facility and the nearby Wuhan Institute of Virology have posted articles about collecting bat coronaviruses from around China, for study to prevent future illness. Did one of those samples leak, or was hazardous waste deposited in a place where it could spread?

Richard Ebright, a Rutgers microbiologist and biosafety expert, told me in an email that “the first human infection could have occurred as a natural accident,” with the virus passing from bat to human, possibly through another animal. But Ebright cautioned that it “also could have occurred as a laboratory accident, with, for example, an accidental infection of a laboratory worker.” He noted that bat coronaviruses were studied in Wuhan at Biosafety Level 2, “which provides only minimal protection,” compared with the top BSL-4.

Ebright described a December video from the Wuhan CDC that shows staffers “collecting bat coronaviruses with inadequate [personal protective equipment] and unsafe operational practices.” Separately, I reviewed two Chinese articles, from 2017 and 2019, describing the heroics of Wuhan CDC researcher Tian Junhua, who while capturing bats in a cave “forgot to take protective measures” so that “bat urine dripped from the top of his head like raindrops.”

And then there’s the Chinese study that was curiously withdrawn. In February, a site called ResearchGate published a brief article by Botao Xiao and Lei Xiao from Guangzhou’s South China University of Technology. “In addition to origins of natural recombination and intermediate host, the killer coronavirus probably originated from a laboratory in Wuhan. Safety level may need to be reinforced in high risk biohazardous laboratories,” the article concluded. Botao Xiao told the Wall Street Journal in February that he had withdrawn the paper because it “was not supported by direct proofs.”

Accidents happen, human or laboratory. Solving the mystery of how covid-19 began isn’t a blame game, but a chance for China and the United States to cooperate in a crisis, and prevent a future one.

As someone else once said in another context, at this point what difference does it make?

What we do know, through genome studies, is that COVID-19 originated in China. We know that more than a million people have been infected. We know that the best hope for preventing its spread would have required China to be forthcoming about the virus much, much earlier. We know that its cost in terms of lost income and revenue will be in the tens of trillions and the cost in terms of lost lives immeasurable.

We also know, with a confidence founded on experience, that we can’t trust the Chinese government’s official statistics. As long as the Chinese Communist Party is in power, cooperation is meaningless.

And I’ve already made my view clear. Whether the virus is a weapon, an accidental release, or naturally occurring, as long as the CCP holds the reins, trade, travel, and, indeed, any interaction whatever with China is simply not worth the risk.

11 comments

The Best Case Scenario

South Korea has been touted as a best case scenario for the response and outcome of the COVID-19 pandemic so I’ve been following COVID-19 events there pretty closely at Worldometer. Two of the graphs there concern me. The first is this:

If that can be believed, it suggests that almost half of the people with COVID-19 may remain chronically ill.

That’s augmented by this graph:

New cases seem to have stabilized but there’s been a sharp decline in new recoveries that is very concerning. If the new cases exceed new recoveries in coming weeks, the South Koreans may have succeeded in making their health care system collapse in slow motion.

IMO this supports what I have been saying for some time. We are not just in desperate need of more testing we really need an effective treatment.

That’s not very encouraging as best case scenarios go and we’re unlikely to equal even that.

15 comments

Are There Too Many Flights?

The editors of the Wall Street Journal are worried that U. S. airlines will be nationalized as a result of the policy response to COVID-19:

America’s beleaguered passenger airlines are allocated roughly $50 billion in the coronavirus relief bill that passed last week. Loans represent half of the amount. The rest is grants to backfill wages and benefits, offered on the condition that the airlines won’t cut compensation or furlough anyone before Sept. 30.

The idea is simply to freeze the staff list for six months, at which point the pandemic might have receded and air travel recovered. In exchange, Congress has authorized the Treasury Secretary, at his sole discretion, to “receive warrants, options, preferred stock, debt securities, notes, or other financial instruments” that constitute “appropriate compensation to the Federal Government.” The law also empowers him to take “a warrant or equity interest” in airlines receiving loans.

The desire to get something for the taxpayer’s buck is understandable, but there’s a real risk here of a long-term nationalization. Look at the latest valuations of the biggest airlines: roughly $15.3 billion for Delta, $6.4 billion for United, and $4.6 billion for American. Leave out Southwest, which has fewer debts than its peers. Add about $8 billion combined for Alaska, JetBlue, Allegiant, Spirit and Hawaiian. Compare that with the federal government’s offer of $50 billion in loans and grants.

The point is that if the government takes equity, its stake might not be a sliver, especially if the employee subsidy isn’t discounted. At the end of 2019, American Airlines was paying about $1 billion a month in salaries, wages and benefits, according to its latest financial statement. With the number of flying passengers now down 90%, how long would the Treasury have to float an airline’s payroll before taxpayers became its majority owner?

Several observations. First, although I think that the payroll subsidies are probably good policy at least in the near term, the loans, which appear to be the editors’ main concern, shouldn’t exist at all. The airlines able to weather this storm, the leading candidate being Southwest, should survive and the others should be allowed to fail. Managing risk is one of a manager’s primary jobs. If managers are incapable of doing that, we shouldn’t subsidize their feckless behavior.

Second, I strongly suspect that there will be a lot less flying for the foreseeable future and the decrease in passenger miles may not be temporary. IMO the best policy WRT the airlines is gradually decreasing subsidies including wage subsidies after the immediate crisis which I would define as the next several months, say, starting in June. Airline employees will be increasingly motivated to look for other jobs as the subsidies decrease.

Third, I hope there’s some consideration of executive compensation in the legislation. $1 million of that billion a month American shells out in payrolls goes to the CEO. When I wasn’t an AA stockholder, that was none of my business but if I have been dragooned into becoming one to keep the airline afloat it is. $12 million a year plus options sounds like a lot for the CEO of a $6 billion company. Microsoft is a $1 trillion company and Nadella earns less than four times that. Something does not compute.

Finally, IMO we’re still doing far too much flying. Even before COVID-19 flying was a pretty miserable experience, especially compared with what it was like, say, 40 years ago. I don’t understand why we have been subsidizing it so heavily.

0 comments