Coronavirus: Iran cover-up of deaths revealed by data leak
bbc.com
bbc.com
>The number of deaths from coronavirus in Iran is nearly triple what Iran's government claims, a BBC Persian service investigation has found.
This is false. There is a very important distinction between confirmed deaths due to coronavirus, and deaths of people who may have shown some coronavirus symptoms. One is clearly an undercount, one is clearly an overcount and they aren't measuring the same thing. So to publish on your state broadcaster a story deliberately mis-representing that fact is a bad move.
The data even shows that the confirmed coronavirus deaths track proportionately to suspected coronavirus deaths - actually showing that it's highly unlikely that the numbers are being manipulated.
If you go here: https://coronavirus.data.gov.uk/ the UK government gives the exact same measure that Iran uses as their official number.
That said, it is a bit awkward, given the UK's own undercounting.
Absolutely not true. From propagating the idea that life-long advocate of equality, Jeremy Corbyn, is antisemitic, to sending well-fed journos out on boats to chase migrants (never people, migrants) around asking them what they want, the BBC is a Conservative mouth-piece.
[1] https://www.thenational.scot/news/18620765.bbc-slammed-portr...
https://www.independent.co.uk/news/uk/home-news/jeremy-corby...
...which in my book means they must be doing something right. Their reporting certainly seems less rabidly partisan than pretty much anything in the US.
For the BBC they HAVE to keep the government happy to continue to exist, or they will be legislated away. Same as any other state owned broadcaster.
Although the threats from the current government against the BBC seem empty. Why would they destroy the most successful government controlled media operation in the world?
(I'm being slightly flippant - I am aware how it's possible to be wrong on both counts. However I do think calling the BBC a "Conservative mouth-piece" when many on the right of the Tory part think the BBC is left-wing and beyond repair is a little unfair)
The right-wingers get upset about social pieces, the left-wingers get upset about economic and political pieces. But it's not even possible or desirable to be utterly unbiased, so one of our broadcasters being "the voice of the Establishment" doesn't seem like that bad a thing to me. It's not like other broadcasters aren't avaiable.
However, that could be said of most (all?) large media organizations in the UK.
Does it refer to that as a "British government cover-up"?
It's like they always believe authoritarian leaders are much competent than democratic leaders, any mistakes in authoritarian nations must be cover-up.
Meanwhile, they also believe authoritarian nations are bound to collapse, because their leaders capabilities are eventually not a match to the complexity of managing the state.
Then, why isn't this the moment where the authoritarian leaders over reached their capabilities?
As far as I am aware there is no such thing happening in Iran.
So in the case of the UK the BBC is saying that UK govt response has been poor, and in the case of Iran that there is a "cover up". I would have thought that was bordering on a reasonable job of reporting the facts.
>The number of deaths from coronavirus in Iran is nearly triple what Iran's government claims, a BBC Persian service investigation has found. The government's own records appear to show almost 42,000 people died with Covid-19 symptoms up to 20 July, versus 14,405 reported by its health ministry.
Now those two things aren't the same - the number of deaths of people who could possibly have had coronavirus isn't the same thing as the number of deaths of people who were actually confirmed to have coronavirus. The first is a quite expansive definition which is what the BBC is using, and the second is the number the Iranian regime publishes. And that second definition is a perfectly reasonable thing to publish, it's what the UK publishes and reasonable people know that's likely to be an undercount.
Now, given that the BBC is already comparing apples to oranges, comparing the suspected cause of death with confirmed. Let's look at the claim you're quoting.
>The number of people known to be infected is also almost double official figures: 451,024 as opposed to 278,827.
So what I'd want to see, before I took the second quote seriously, is a defintion of what the numbers refer to.
Because if you read it in context, that sounds to me very much like it could be the BBC is doing exactly the same thing again, comparing the official figures which are the number of people actually tested confirmed cases, with the unofficial figures, which are simply people presenting with symptoms but not comfirmed.
Testing capacity is a worldwide issue, but apparently with Iran we're going to try and literally tie this to the Iranian revolution.
Other countries (and going by this report that would seem to include Iran) are heavily underreporting to the tune of 60%.
And then obviously you have countries that are somewhat underreporting like the UK, where they have reported covid deaths of approximately 46 thousand, but excess deaths of 64 thousand giving an approximate underreporting in the order of 25+%.
As to the revolution, it's not likely the cause as merely another example of a tightly controlled society that is sensitive to criticism; a reflection of their fear of "losing control".
"Beware of him that would deny you access to information, for in his heart he dreams himself your master."
I don't know if the data were faked and I assume that everybody is applying "creative accountability" here. Not only in Iran. Is a political thing.
... but this is not so strange as it seems.
People can pass the test more than one time. As immunity is so short lived, each patient could test positive, then negative and then positive again, an "undefined" number of times. This should count as one patient, two events. I assume that each positive has a second try at least to assure that is not a false positive
Yes. Correct. Agreed.
If that's the measuring stick the USA has been overcounting. Healthcare has been authorized by the CDC to count suspected but unconfirmed (i.e., no test) as positives.
Add in the fact that healthcare facilities are compensated more for Covid-19 case, and most of those facilities are for-profit...what could go wrong?
You can't get away from the piles of corpses, and only a small number of these are explained by the increases in other deaths.
It's absolutely unscientific to say that ever death with C19 present was caused by C19. That's not how it works.
The fact that this ruse has been normalized doesn't make it right.
The truth is, at this point, the data is in such shambles that it can't be used to produce accurate information to drive intelligent decisions.
That's what we've been seeing, and continue to see. Believe what you want but the absence of comorbidies should be a red flag. Either you see the danger in that slight-of-hand, or you dont.
If you're claiming they died of cancer or heart attacks or cardiac arrests or stroke you still need to explain why there are so many more than normal.
That aside, you're assuming that any incremental death can be attributed to Covid and only to Covid. That's a false and unscientific assumption. Futhermore, in a country with the obesity rate of the USA's it's foolish to believe that Covid is the only thing leading to death. There's a reason it's called morbidly obese.
I have, which is why I talk about all cause excess mortality.
https://blog.ons.gov.uk/2020/03/31/counting-deaths-involving...
https://www.ons.gov.uk/peoplepopulationandcommunity/healthan...
https://www.ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386...
> You're assuming that any incremental death can be attributed to Covid and only to Covid.
No I'm not. I'm asking you what you've read that makes you believe what you're saying. Support the claims you're making.
Do you have a source for that? I have heard or read no such thing, and can't find any claims of such. If anything, the claims I see are of the CDC undercounting infections.
> Add in the fact that healthcare facilities are compensated more for Covid-19 case
Again, do you have a source for that? Many states are covering all costs not borne by insurance, but I have not seen anything to indicate any COVID premiums being paid.
> and most of those facilities are for-profit
No, most of those facilities are non-profit.
https://www.cdc.gov/nchs/data/nvss/vsrg/vsrg03-508.pdf
Now before you say "Well, it's not that many" the truth is, you can't. We don't know. The data collection is sloppy, inconsistent, etc. But if you can find a CDC report that addresses this and breaks it out (read: transparency) then I would like to see that. Please share.
For the second:
https://www.factcheck.org/2020/04/hospital-payments-and-the-...
Do you have a source for that?
Does Iran have similar numbers easily available? (Obviously, these don't have to be in English.)
In practice they're both very likely undercounts. In the developing world, many people die from illness without ever having been seen by a health care professional at all, or counted as anything.
It will be years before we have a good model for real infection rates and excess death statistics.
I think I stand by "years" until we have a good consensus model for how the pandemic spread and what the effects were.
That's really all you need; a historical time series of dead people. Frankly, the US, Europe; it's the only data for Covid I look at any more because the rest of it is politicized and site specific, and a changing process over time.
For example:
https://www.euromomo.eu/graphs-and-maps/
https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidvi...
Obviously Iranians "can count". What developing nations don't have is a way to get all the dead COUNTED. Some old guy dies in a little village with little to no government services: who's job is it to "count" that death? How do you get that entered? Do you just trust cousin Farhad when he tells you at your office in slightly-larger-village three days laters? How do you know that's not the same death as this other person called in? The names are spelled differently, but maybe that's just a mistake?
Be serious here. Running a government is a hard problem. If it were easy, we wouldn't even have "developing" nations at all.
[1] https://iranian-studies.stanford.edu/sites/g/files/sbiybj619...
This is one of the most ridiculous, preposterous and ... bigoted assertions I've heard all day: of course the Iranians can count the dead. They were probably capable of it 2500 years ago under Cyrus. They're certainly capable of it today[0]. You realize the Iranians make stuff like cars[1], aircraft[2], and have successfully hacked "advanced" US stealth drones[3], right? And you think they can't count dead people?
[0]https://irandataportal.syr.edu/census
[1] https://en.wikipedia.org/wiki/Automotive_industry_in_Iran
[2] https://en.wikipedia.org/wiki/Iran_Aviation_Industries_Organ...
[3] https://en.wikipedia.org/wiki/Iran%E2%80%93U.S._RQ-170_incid...
Sweden, for example, was pilloried because their policy had them run their marathon through all up-front. Now they get to essentially return to life as they once knew it, having never had to strictly lockdown like what you see in Australia right now.
So it’s not just death rate, because you have to adjust for whether you can reasonably say you’ve finished the race, or whether your running on a treadmill at Mile 1 just hoping the race gets “canceled” (effective vaccine in widespread deployment) before you run out of stream.
The UK establishment have influence on the stuff that really, really matters. Corbyn is a great example. The BBC accumulate political capital on smaller stuff that is unbiased, then spend that to keep people who want systemic change that impacts vested interests.
[1] https://www.reuters.com/article/us-health-coronavirus-iran/i...
Second list which has leaked now is people who had died with suspicious symptoms but not tested for COVID-19 for whatever reason.
They admitted way before this that they believe real death toll might be 3 times higher than official numbers, and real infected might be up to 5 times higher than official numbers by comparing average/expected death rate and clinics reports.
I think this is an issue in many countries right now. The numbers are real as much as they get to test their population and everyone knows that.
You're correct, but this is still relevant. It all depends on how the statistic is presented. Excess deaths shouldn't be presented as "deaths due to Covid-19 infection", but it could be presented as "deaths related to the Covid-19 outbreak", and people who died from cardiovascular disease because they didn't seek help due to the pandemic are still very much relevant since they are just that: related to the Covid-19 outbreak.
After all, if you save 100 people in ICUs after they contracted Covid-19 that's good, but if 1 000 people die of other causes because they were scared to go to hospital then that is still relevant to understanding how the pandemic affected deaths in general, and how the response could be improved in the future compared across countries.
If I saw this trend only in a handful of countries I'd say they are undercounting (FWIW I still think they are) but because we can see this in US, EU and Asia, it'd say virus is getting less deadly/treatment is getting significantly better.
That, and the age structure of the infected changed: the older people are doing what they can to avoid the infection, if they are able to do that.
Also, across the world, different measures are still in place, changing the dynamics of the spread, compared to the start of the pandemic when the spread was practically unconstrained. The dynamics of the spreading is also different in different settings.
There's no any scientific reason to believe that "the virus" changed in any way biologically. It's the world that does its best to adapt. Also, the schools and the universities are not opened for students at the moment in many countries, also slowing down the spreading.
Edit: answer to: "Viruses don’t mutate?"
The coronaviruses have additional mechanisms to correct the copying, slowing down the mutations, compared to e.g. flu viruses.
Edit2: answering: "The main reason for why Coronaviruses don't mutate much is because they don't have to": I wouldn't call that "the main reason", but it's a part of their success. Coronaviruses have longer genome than many other viruses, and having uncontrolled mutations in that longer genome would make them degrade too fast, so there is that molecular mechanism they have, correcting the copying errors. Additionally, they have other mechanisms to recombine their genetic material, something like "sex between (corona)viruses" where even more than two parents could be possible. But that's different from mutations and happens under different conditions. Knowing all that, and all that what est31 mentions, it is indeed true, the coronaviruses really simply don't need to "mutate fast."
Edit3: "There different strains" is not true. They are different isolates, where the completely minor differences exist, but for all it is known, until somebody proves otherwise, and nobody has, there is still just a single "strain" of SARS-CoV-2.
Edit4: Thanks to Gibbon1, yes, comparing with influenza is tricky, but maybe it's good to give readers the idea once again that this is surely not flu and that the viruses aren't the same and don't behave the same.
Viruses don’t mutate?
The main reason for why Coronaviruses don't mutate much is because they don't have to. While influenza constantly comes up with mutations so that it can come back seasonally, it seems that Coronaviruses take a different approach by evading adaptive immunity instead. E.g. feline Coronaviruses can infect cats over and over again, without large increases in ability for cats to fight it. That's also the case a bit for SARS-Cov-2, there are reports of humans getting infected a second time, but they are rather rare reports, we'd have far more of them if humans had no good adaptive immunity.
There different strains, but apparently the virulence is pretty similar [0].
[0] https://coronavirusexplained.ukri.org/en/article/cad0013/
Single stranded viruses are a lot more stable.
More testing does not explain falling CFR in South Korea for one thing.
A month or so ago, there was a research paper (admittedly an unreviewed preprint) that suggested there are different strains circulating in the US - the part I remember is that the New York City and Chicago strains did indeed act slightly but measurably differently. I think there was a third as well.
As Dr. Fauci would say, it doesn't matter, even if it were a peer reviewed it can still be bad. (1)
There are a lot of bad studies, especially on the "preprint servers" that in more quiet times would not appear at all, or which nobody would take a bit serious. Now there is a lot of wishful thinking or bias involved even in their perception.
The scientific process doesn't protect anybody from some studies simply being bad, the process is there that eventually the bad ones are going to not be reconfirmed, and the really good ones are those that have many confirmations and have even the power to make new predictions, that remain true.
There are known issues with these studies claiming different "strains" too easily.
1) https://www.rev.com/blog/transcripts/dr-fauci-hydroxychloroq...
Note that the trend is worldwide. Even in countries with very high testing in the beginning. South Korea for one!
I also think it's unlikely to have the virus pathology synchronized across the world with current travel restrictions.
Last time I checked the virus actually seem to feature some "dual gene pool" mechanics, which should make it more robust and adaptable to "fading out".
https://www.reuters.com/article/us-health-coronavirus-italy-...
57% in Bergamo province with antibodies.
Italian lockdowns didn't stop COVID - human immune systems did.
This also means for every 1 detected case, there were 46 actual COVID instances in Bergamo.
Should actually read
57% of a "random" sample of 9965 people who were tested in Bergamo province have antibodies.
That might be true for Bergamo, but certainly not for the rest of Italy, especially outside of Lombardy which had much lower infection numbers and death numbers.
Having as many dead as in Bergamo everywhere (0.58% of the whole population according to a short google search) would be unacceptable.
If all of the world reached Bergamo stage, there would be no way to test 57% of the world population.
That doesn't mean there are 46 * 18 million cases world wide -- the deaths would be catastrophic world-wide, just like they were in Bergamo.
However, that is the stuff of the future and the ivory tower. In the course and flow of this public health crisis it is necessary for reasonable speculation and extrapolation to be used to drive our responses.
There is no evidence that I have heard that the most susceptible have been culled from the population yet, and there is no logical reason to believe that this has happened. In fact the contrary seems to be true when the pattern of the outbreak is observed in the USA. Potentially there remain a huge well of victims left in the population.
To claim that this abet contingent knowledge is illegitimate in the face of the human catastrophe that could result from rejecting it strikes me as an rejection of all of the principles of the enlightenment. Our rationality is our sword and shield for fighting the shadows on the wall of the cave. I for one will not be putting these tools down. It is wrong to cower in the dark when our brains can light a light and let us see a path to safety.
A few countries are close enough to be considered accurate. Some are off by over 100%.
It will be interesting, when it's all over and all numbers are in, to look at the statistics of this.
Anybody claiming to know specifically what to expect from something that has never happened before should be treated with a healthy dose of skepticism. Especially if their only way of taking measurements is so imprecise and inconsistently applied.
The deaths may not be directly attributable to the infectious agent and disease mechanism of SARS-COV-2 or COVID-19, that is, the virus or disease. But they are, in the absence of another plausible causal mechanism, part of the overall impact on public health.
Gross mortality anomoly is the truest measure of impact.
(Deeper discussions of causality and epistemology omitted.)
I’m not sure this is accurate. From what I’ve read, many people simply aren’t tested. They are old, got pneumonia, and died. The cause of the pneumonia may have been any number of things, but some large proportion of them are expected to be flu during flu season.
More starkly, visits to the ER for strokes and heart attacks plummeted during lockdown. But there weren’t actually fewer strokes and heart attacks. Many of those people just died at home.
In short, no one can say how many of excess deaths were COVID vs lockdown deaths, but it certainly isn’t the case that the “lockdown”‘category is actually a net negative.
It is not just Iran.
For example, I believe that Belgium has a very inclusive count (proven or suspected) and other countries count only if proven.
This also means that some countries undercount when they have insufficient testing capacity.
Also, with co-morbidities there is a choice whether to assign the cause of death to these factors or to COVID-19. If someone was dying from lung cancer and also contracted Covid-19 at the end what was the cause of death?
So the rule of thumb is: don’t try to compare numbers between countries.
And also note that in some cases you cannot event compare data locally: for example, in the UK, England and Scotland have different definitions.
But to put that in a realistic perspective, it is a little bit like if you corporate Enterprise Architecture function published new guidelines for how to ensure data consistency across business units. Publishing a document does not make it so overnight.
> Tehran, the capital, has the highest number of deaths with 8,120 people who died with COVID-19 or symptoms similar to it.
Not sure how Iran had formalized the similar symptoms and if they count death from another deceases that were triggered by COVID-19.
Imo the cover-up reasons seems weak.
Why would you want to make that distinction and not count those people with pre-existing diseases who died as a consequence of contracting COVID-19?
For example weak cardiovascular health in overweight people already puts them at higher risk of dying after contracting COVID-19. I don't see why anyone would not count those cases unless they want to paint a rosier picture.
The stats about the total number of death seems quite noisy.
Of course this will also include deaths not directly caused by infections with the virus. But it gives an overall picture of the impact of the virus and its downstream effects on a country and allows comparisons between countries.
Might be the better, or more illuminating way, but it's a standard for whom?
Most western governments report covid deaths it in a totally different way - not by taking into any account excess deaths. So it's not like Iran is doing anything special in this regard...
Yes, but you can only do that in aggregate over a larger period, you can't use it as a basis for daily statistics, not in the least because not all deaths are recorded the same day they occur. The daily death tallies are based on confirmed cases only, in pretty much every country.
That means underreporting of cases is the norm, not the exception [0]. I must say those graphs are a bit hard to interpret, but if I understand correctly: a gray bulge above the dark gray dotted line implies excess deaths not officially attributed to Covid-19. From those figures, it seems only Denmark and Germany are accurately reporting Covid deaths, and Belgium, France and the US may be overreporting the death tally (although it can also mean that not all deaths in the past weeks have been recorded yet in the official figures).
[0] https://www.theguardian.com/world/ng-interactive/2020/may/29...
Of course, using incomplete data may not tell us anything either. It's also not an actionable graph, because there's too much uncertainty.
All available data are seen through the rear-view mirror and tell us where we were not where we are. Those who work in such scenarios are aware of this, and other limitations: test coverage and limitations, sampling, selection criteria, reporting lags, comorbidity biases, etc.
Most of the rapidly-available data are subject to greater bias. Much of the more accurate data are subject to greater lags. But with sufficient history we can use the latter to adjust the former (pending final calibration measurements) to improve the clarity of current assessments.