Critical Issues: Spot the Error

This one could be a WitCH or a PoSWW, except we’re withholding some information. So, consider it an analysis puzzle.

The following was posted to an Australia-NZ statistics list. The email was from an American statistician, referring to and quoting from this article (Update: link fixed), on a Texas hospital adopting critical care guidelines. See if you can identify the problem.

UPDATE (30/7)

A weird hint: we are not playing fair. (And, boy are people gonna be pissed.)

As we wrote in reply to a comment “There is the post, and the links to which the post refers. You can consider what ever issues you see.” So maybe list quickly, without elaboration, anything you see as an issue.

UPDATE (31/7)

OK, time to end this, and so a final hint. As indicated in the comments, the “problem” is with the poster’s line “Statistics is a bitch”. We’re very, very pleased that no one has hit upon the “problem” with this line.

FINAL UPDATE (01/08)

Well, wasn’t that fun? Thanks to everyone for playing along.

As indicated above, the “problem” is with the line “Statistics is a bitch”. And what’s wrong with that line? Not a whole lot. It’s not a great line, since it can be read as treating the “statistics” as a from-nowhere reality, rather than the disastrous consequences of Republican screw-up. But no big deal.

So, why post on this as a problem? Because the stats email list to which the comment was posted thought it was a problem. A number of commenters took very serious issue with the poster’s use of the word “bitch”.

This began with an off-post email to the poster, indicating “the language used is not at all appropriate for [such] an email list” and a request: “I’d appreciate it if you could apologise for this choice of words.” (To whom?) The poster replied to the email list, with a long and unhelpful, but fundamentally reasonable, non-apology apology. In brief, the poster, who is in Texas, suggested that they had much, much bigger things to worry about. And then the bashing kicked in.

There were calls for the poster to “grow up”, to “stop using hurtful, offensive language”, suggestions that “the problem is the use of a term that is all too often directed at women in a derogatory way” (ignoring that this was not the case here), whining about “gendered name-calling”, and all manner of nitpicky and gratuitous complaint.

It was crazy and it was revolting, and all of it coming from proud and proper academics. Eventually there was some tepid defence of the poster, but way too little and way too late. No one stood up properly to these ridiculous, self-important language nazis.

Which is why we posted about it here. Of course this is the type of blog where those offended by strong language are unlikely to hang around. And, maybe some shyer types here agree with the poster’s critics. But, it was still very, very pleasing that no one who engaged here had a clue what we could have been on about.

Another Message to Fellow Melburnians

1) Stay home.

2) If you don’t stay home, wear a fucking mask.

3) Forget that, just stay home.

UPDATE (22/07)

Some gentle observations, after our first shopping trip in a month, and with Uncle Dan on the car radio:

4) If you’re sick and you don’t isolate before your test, you’re a fucking moron.

5) If you’re tested and you don’t isolate before your results, you’re a fucking moron. 

6) Just because it is not yet mandatory to wear a mask, it doesn’t mean you shouldn’t, you fucking morons. 

7) Bunnings is lying through their fucking teeth. 

Should Schools Now Be Opening?

Just for a change, we’re asking a non-rhetorical question. So, ignoring the fact that ScoMoFo is a thug, and ignoring the fact that “Dumbo” Dan Tehan is a thug, is it time for schools to reopen?

Our suspicion is that, at least in Victoria, the answer is “no”. We haven’t thought hard about it, however. So, while we (try to find time to) think some more, we’d be interested in what others have to say.

Very quickly, here are the arguments we see for opening schools in Victoria:

  • Federal health officials suggest schools are safe.
  • The NSW study (not yet peer-reviewed) suggests schools are safe.
  • Childminding.
  • Year 12 students are getting seriously dicked around.

Here are the arguments we see for keeping schools in Victoria closed:

  • ScoMoFo is an idiot.
  • Dumbo Dan makes ScoMoFo look smart.
  • Daniel Andrews and his Chief Medical Officer are not idiots.
  • No one has a real sense of what will happen when restrictions ease.
  • The kids (P-11) miss a term of school? Big deal.

Vanessa’s Appt Concerns

Today, the Australian government released COVIDSafe, the Government’s coronavirus tracking app, based on Singapore’s TraceTogether version. The release comes complete with the Government’s predictable reassuring and cajoling and guilt-tripping. Should Australians trust them and use the app? Really? For us, there is a very simple answer: when and only when Vanessa Teague gives the all clear.

Vanessa is an expert on cryptography and, as it happens, is an ex-student and a good friend. She is very smart and is as principled a person as we have ever met. Along with many of her colleagues, Vanessa has been critical of the Government’s needless (and entirely predictable) secrecy over COVIDSafe. She has written a series of blogposts about their underlying concerns, and in particular the Government’s failure to follow up on promises and release COVIDSafe’s source code. This is Vanessa’s current stance on using the app (as of 23/04):

“In its TraceTogether form, I would be happy to run it on the train but refuse to run it in my home or office. I need to see the details of Australia’s version before I decide.”  

And, if that’s what Vanessa suggests then that’s what we’ll do, right up until Vanessa and her colleagues suggest otherwise. We’ll regularly be checking on Vanessa’s blog and twitter account.

 

Postscript: We had planned on writing about Vanessa a month or so ago, when she made the news. That story is highly relevant, since it involves privacy concerns, government screw-up, an arrogant and inept minister, a limp lettuce watchdog, a thuggish department secretary being matey matey with a vice chancellor, and a spineless university. Yep, same old, same old. But, given the speed of the times, we’ll probably have to leave that story be.

 

UPDATE (27/4)

The Minister for Health has today made an undertaking to release the source code “within two weeks”. We’ll see. (The formal agency response on privacy (26/4) states that such release will be “subject to consultation with the Australian Signals Directorate’s Australian Cyber Security Centre”.)

Vanessa and her colleagues have a new blog post (27/4). The post has been written “on a best-effort basis using decompiled code from the app, without access to server-side code or technical documentation.” Their conclusion:

Like TraceTogether, there are still serious privacy problems if we consider the central authority to be an adversary. That authority, whether Amazon, the Australian government or whoever accesses the server, can

    • recognise all your encryptedIDs if they are heard on Bluetooth devices as you go,
    • recognise them on your phone if it acquires it, and
    • learn your contacts if you test positive.

UPDATE (01/05)

We’re not going to bother with the nasty guilt-tripping on the COVIDSafe app, including from numerous media nitwits who should know better. This from Bernard Keane suffices.

Vanessa now has a very good twitter thread on the seemingly contradictory safe/not-safe messages from IT folk.

UPDATE (11/5) Vanessa has a twitter thread (08/05) on ScoMoFo’s latest round of silly buggers.

UPDATE (13/5) This will come as a great surprise, but it turns out that Greg Hunt is a dishonest piece of shit.

UPDATE (15/5) Vanessa and her colleagues have a new blog post (14/5): The missing server code, and why it matters.

UPDATE (20/5) Vanessa and her colleague Chris Culnane have a new blog post (19/5), on flaws in and corrections to the UK covid app (and why this was possible). Vanessa also has an accompanying twitter thread.

Guest Post: Failing the Coronavirus Test

The following was written by my brother Dan, who has been watching the governing of the coronavirus pandemic very closely. The views are, of course, Dan’s.* 

In a real emergency situation, it tends to be less than useful to have everyone imagining that they are an expert whose opinion absolutely deserves to be heard. One of the least enjoyable aspects of the general response to this pandemic, apart from the virus itself, has been watching everyone on Twitter and on television starting from the unreflective assumption that whatever crap they happen to believe is automatically worthy of public pronouncement. In such a situation, and despite one’s better judgment, one almost begins to feel sorry for those in actual power who have to make actual decisions with actual consequences. Nevertheless, and even though I have absolutely no grounds for claiming any expertise in any area related to the current pandemic situation, I feel compelled to at least raise some questions about how governments and journalists are responding, in particular in Australia.

Let’s begin with an article published yesterday in The Guardian:

How many cases of coronavirus are there in Australia? Live Covid-19 numbers and statistics

What followed in that article was nothing more than a representation of the information that Australian state governments have deigned to provide about the numbers of confirmed cases. Not one word was written about the question of how many actual cases there may be. And yet: do not GPs and doctors have reporting requirements concerning serious infectious diseases? In other words, are they not obligated to report probable cases to the relevant medical authorities?

We have to distinguish three categories: confirmed cases for those who have had a Covid-19 test with a positive result, probable cases, where a doctor has assessed a patient as showing symptoms that mean they are likely to have it but have not had an actual test for the presence of the virus, and actual cases, that is, everyone in the community who does have or has had the virus. Currently we have no way of knowing the precise number of actual cases, but the best clue about how many actual cases there may be is to know the number of probable or suspected cases that have been detected. For journalists to treat data about confirmed cases as indicating anything meaningful about actual cases is obvious and needless distortion.

Why does this matter? It matters because currently GPs are seeing patients who have the symptoms of Covid-19, and they are telling those patients to go home and stay home. Clearly, this is a directive coming from above. How widespread is this practice? We cannot know, because we are not receiving information about probable cases. Whether it is a good idea is one thing (which we will come back to), but the main point is that there is a co-ordinated approach and one that systematically distorts the informational value of the data about confirmed cases. Creating pretty charts and graphs showing what is happening with confirmed cases only contributes to the obfuscation of the situation. But all those GPs must be following their reporting requirements, and government must be collecting that information, yet no Australian journalist seems to have thought of ever asking for it, let alone demanding it. Why?

Government should want to give out that data, because it will make the situation seem more serious than the perpetual reporting of confirmed cases can ever do, and this seriousness seems to be what they are struggling to convey. The governments of both China and New Zealand decided at a certain point that it was better to use probable cases as the official data, rather than confirmed cases, but other countries including Australia do not. Perhaps the reason the Australian government is reluctant to do so is because it will have to admit publicly that it is telling GPs to send home probable cases (as long as they are mild), and not to make any further therapeutic intervention. Perhaps they believe that such an admission will cause consternation, disruption, or panic. Who knows? But the passivity of journalists about this means that Australians are unable to form any clear picture of how widely the epidemic has spread throughout the country.

It matters because there is one fundamental question animating every decision in relation to this pandemic: the calculation of the health benefit of highly restrictive strategies compared to the economic cost of those same strategies. What’s the connection? The connection is that the real issue is not “how tight should the lockdown be?” or “when can we begin to relax the restrictions?”. The real question is “what possibilities does a lockdown open up?” In other words, the lockdown is not an end in itself, but a means to another end: a way of setting the conditions for doing those things that actually need to be done. This is exactly what Tedros Adhanom Ghebreyesus, Director-General of WHO, said yesterday:

“We understand that countries are trying to assess when and how they will be able to ease these measures. The answer depends on what countries do while these population-wide measures are in place.”

Dr. Ghebreyesus’s point is that a lockdown itself is not the “solution”. It is only the basis on which any possible solution could be enacted. After that, the WHO Director-General itemises what must be done during the lockdown. First, expand and train the health care workforce. OK, no problem: presumably, hopefully, governments and authorities understand this point and are trying to do something about it (and should have been trying since Wuhan).

But then we come to the second point:

“implement a system to find every suspected case at the community level”.

 This, it seems to me, is fundamental and crucial – and it is being completely avoided by Western countries. Currently we hear a lot about “how much testing is going on” and “how wide is the testing”. By “the testing” is meant Covid-19 tests, those that lead to the “confirmed cases” being confirmed. Everyone who wants to seem sensible and rational says “we need lots and lots of testing”. Is this true?

This, it seems to me, is not what WHO is saying. They are saying “find every suspected case”. Sure, Covid-19 tests are one way of finding those cases “at the community level”. If you have enough tests. If you have the resources to implement them. If you have enough time to do those tests. But it may be that this is a highly inefficient way of “finding every suspected case at the community level”.

What would be more efficient? Well, firstly, 90% of cases start with fever, and temperature testing is much simpler and quicker than testing for the virus itself. Anyone can do it. Anyone can be employed to do it. Once someone has been found to have an elevated temperature, then the question becomes what procedure to follow after that. It was precisely this point that Donald McNeil of the New York Times explained with great lucidity two weeks ago in this interview, describing in some detail what they actually did in China that actually worked:

 

From McNeil, we can see that “testing” in China actually meant a sequence of steps, so that the Covid-19 test would only be necessary after having passed through the earlier steps. Notice that this is exactly what Dr. Ghebreyesus said in point 2.

This properly expanded notion of “testing” ties into Dr. Ghebreyesus’s third point, on the need to “ramp up production capacity and availability of testing”, as well as the fourth point, on the need to “identify, adapt and equip facilities you will use to treat patients”, and his fifth point, on the need to “develop a clear plan and process to quarantine contacts”. Of course, governments will say that they are doing precisely the things Dr. Ghebreyesus demands. But are they doing those things? Is this the reason that the Australian government prefers not to mention that GPs are quietly telling patients with mild symptoms to go home and stay home? Because doing so amounts to crossing their fingers that those with the virus will do as they are instructed, that they will know how not to infect their family members, that they will not take a trip to the supermarket, or to Dan Murphy’s, etcetera etcetera etcetera. And maybe crossing their fingers will work. It is also possible, however, that instead of finger-crossing, it just might actually be better to pull patients suspected of having Covid-19 out of wherever they are, send them to one place for further investigation, and then, if confirmed to be infected, send them to another place for treatment.

Of course, no one is keen to follow the Chinese example of welding people into their apartment buildings. But that’s not the real story of China, and we’ve known that for some time now. Pointing out what China might have done wrong is not very important, if they are not things we are likely to do ourselves. But pointing out what they did right might be very important, if anyone cares enough to think about it. The real story of China’s response is: having a systematic approach to finding every suspected case, having a multi-step and efficient way of finding out if they do have Covid-19, removing those people from general circulation before they have a chance to infect others, offering treatment in dedicated facilities, and then sending out teams to trace their contacts.

The questions are: What is the real reason that Western countries are not willing to do what WHO has told them needs to be done? Why are they not willing to do what China has shown them can work? Why are they not willing to hire masses of people (plenty of whom have recently had space freed up in their calendar) to stand in a mask and gloves at train stations and public building entrances to give each person a temperature test? Why have they not been willing to insist that suspected cases be isolated until they are confirmed or otherwise?

It seems to me that the answers to these questions are not just about a different assessment of what’s the best strategy, but instead have significant cultural and psychological aspects. That’s a discussion for another time. However that may be, the results of not asking these questions are clear in one Western country after another, as they take one step after another in what seems a completely reactive way, as the completely predictable progress of the pandemic unfolds.

The benefits of a systematic and carefully prepared approach are obvious: you know when to lock people down, in what way to do it, and what you hope to achieve. For example, consider public transport, which is obviously a place where infection can spread (see the photos and videos from the London tube). In the West, the question is always framed: should you stop it to prevent infection or keep it going because it fulfils an essential service? But this is the wrong question. No doubt it is dangerous, and no doubt it is essential, so the point is to create the conditions under which it is justifiable for it to continue. That may well mean significantly reducing the number of passengers. Why? Precisely so that it will be possible to give every single person who enters a train station a temperature test. Will that pick up every infected person? No, it won’t. But it will very significantly reduce the numbers exposed, while providing an opportunity to “find suspected cases at the community level”, and still allow some necessary movement.

The same thing applies to schools. Instead of “should we keep them open or shut them down?”, the question should be, “under what conditions would it make sense to keep them open?” Those conditions should include testing…not Covid-19 testing: temperature testing. Every single day. Why not? Why is that so impossible? Why is it never proposed and why do journalists never ask?

It is not that every country has to take exactly the same measures. Each country must evaluate its specific situation and make the most rational decisions possible in line with the points made by the Director-General of WHO yesterday. Some things that China did may be simply impossible for Western countries to do. So then, why not take a look at how Singapore was able to contain the virus so successfully? Was it by having a complete lockdown, close schools, and so on? No, schools remained open in Singapore and no total lockdown was necessary, yet it has been vastly more successful than Western countries. As Dale Fisher explained on 18 March, Singapore was already prepared by the end of January, had a vast number of hospitals (for a tiny island) specifically dedicated to handling patients, kept confirmed cases (even mild ones) in hospital and separated from friends and family, made sure that school children showing any illness whatsoever did not attend, and communicated clearly and explicitly with the public:

Why Singapore’s coronavirus response worked and what we can all learn

This kind of effective preparation simply did not occur in Australia, just as it did not take advantage of its specific situation as, like Singapore and Taiwan, an island nation.

And then there is the question of mandating the wearing of masks in public situations, and governments supplying them to those who need them… (Edit 27/03: links added)

It is only by asking questions such as these that it will be possible to adopt strategies that will allow a gradual transition from highly restrictive “lockdowns” to situations where there is greater freedom of movement and activity. This is what Singapore anticipated from the beginning, and it is clearly what needs to be thought about here and everywhere, if the medical and economic “balance” is going to be approached in anything like a rational way, and if there is in fact a way of preventing a global economic catastrophe. This shift to a less restricted approach is what China is beginning to do right now, but the problem is that, in a globalized world, each country is connected to all the others, and so such strategies must include very strong border control until such a time as everyone starts to behave like grown-ups. Currently, most of the Western countries are still behaving like scared or defiant children, to one degree or another.

In the wash-up of this pandemic, it will be found that governments, media and populations each had their true strengths and weaknesses exposed. Those strengths and weaknesses may not be the ones we thought they were beforehand. Based on current evidence, Western governments, media and populations seem, in general, very far from being top of the class.

 

*) Dan and I agree on my most things. Except, Dan believes that I have too much faith in the wisdom and goodness of human beings. 

Scott Morrison is Fucking Up

From the opening remarks of the March 25 WHO press briefing:

Asking people to stay at home and shutting down population movement is buying time and reducing the pressure on health systems. But on their own, these measures will not extinguish epidemics.

The point of these actions is to enable the more precise and targeted measures that are needed to stop transmission and save lives. We call on all countries who have introduced so-called “lockdown” measures to use this time to attack the virus. You have created a second window of opportunity. The question is, how will you use it?

There are six key actions that we recommend.

First, expand, train and deploy your health care and public health workforce;

Second, implement a system to find every suspected case at community level;

Third, ramp up the production, capacity and availability of testing;

Fourth, identify, adapt and equip facilities you will use to treat and isolate patients;

Fifth, develop a clear plan and process to quarantine contacts;

And sixth, refocus the whole of government on suppressing and controlling COVID-19.

These measures are the best way to suppress and stop transmission, so that when restrictions are lifted, the virus doesn’t resurge.

Dr. Tedros Adhanom Ghebreyesus, Director-General of WHO

UPDATE (26/03/20)

Donald McNeil Jr talks about what worked in China (and further discussion here)