Thread by @DanKaszeta

OK. As promised, a thread about allegations of poisoning incidents in Iran. (Long thread.) Hold all questions until the end. (1 of xxx) #IranianSchoolgirls

A variety of reports have circulated about incidents involving schoolgirls in Iran becoming ill. Reports and descriptions of these incidents vary significantly.

These incidents have been going on for some time. The first such reports that I heard of emerged in Qom in November. bbc.co.uk/news/world-mid…

The various reporting that has come to me has a wide range of information. Some of the accounts are muddled and a few are contradictory. As there have been multiple incidents and I don’t read Farsi, this is an understandable variable.

Some of these have been attributed to “toxic gas” or suspicious odours. Other accounts have claimed that it was contamination of food or water as the possible source.

By means of historical background, let me point to three major things:

First, Iran was a victim of chemical warfare in the Iran-Iraq war in the 1980s so claims of chemical weapons or poisonings may resonate with the public more so than in some other countries.

Second, There have been dozens of similar incidents in a neighbouring country. Afghanistan has seen many such incidents, often but not entirely, at girl’s schools from 2009 onward.

These Afghan incidents were largely not resolved, although some theories continue to circulate. They were, in my opinion, under-investigated. Here are merely four examples of numerous Afghan incidents:

Third, there is a climate of odd claims in use of chemicals during the current wave of unrest in Iran. Military smoke grenades have been wrongly accused of being nerve agents, as one example of many.

Chemical and biological forensics can be extremely difficult and this is an area where I know a bit. More than a bit. I’ve written extensively on this. But this isn’t about me.

Getting to the bottom of these alleged incidents is complicated, and it is complicated for a number of reasons, which I will unpack in considerable depth.

Elapsed time makes investigation difficult. To explain simply.... if a substance is capable of causing harm in the human body, then that means the compound in question reacts with the environment.

Many substances are transitory in nature. Gases waft away. Vapors dissipate. Liquids evaporate. Also, many things react with the natural environment. Sunlight, oxygen, moisture, etc. all degrade things.

Elapsed time combined with this reactivity/degradation means that if you go looking for something, you may not find it. In a situation like this, actually finding the alleged causative substance is often the only useful evidence, but can be extremely difficult.

Actually getting a sample of a gas, vapor, or aerosol pretty much requires you to be there, with the right equipment, at the time of exposure. Which doesn’t happen very often.

Remember Birling Gap? Nobody ever figured that one out. Nor will they. Because nobody collected a sample at the right place at the right time. Nor was that ever likely to be possible. theguardian.com/uk-news/2017/a…

Another form of evidence is witness accounts.

Witness accounts can be unreliable. In situations like these, witness accounts can be a bit contagious. One person says “well, I smelled X” and other people latch on, often subconsciously, to the suggestion.

Some witness accounts turn out to be third party. In other words “Well, I didn’t see it but my friend did….”

Variability in witness accounts is a known and well-studied phenomenon in air crash investigations, criminal investigations, civil litigation, and criminal trials.

It would be highly unusual for this variability in witness accounts to somehow NOT exist in incidents such as the Iranian ones.

Many witness accounts have focussed on smells. Smell CAN be a clue in such instances, but it can also be misleading. Many truly awful poisons are odorless.

Strong smells can exist for reasons other than an attempted poisoning. Smell alone is not enough to make a diagnosis.

Smells and description of smells is highly subjective and varies greatly from person to person. There are also chemicals that some people cannot smell but other people can.

There are chemical hazards that have smells only at certain concentrations. Some hazards are smellable at low levels but become un-smellable at higher concentrations.

The various odors described in the Iranian incidents are difficult to tie to particular chemical hazards.

“Fruit” (unspecified), “oranges”, “rotting oranges”, “tangerines”, “rotting fish” have all been described to me. None of these descriptions have been particularly revelatory.

So, if we don’t have a physical sample and witness accounts are variable and not that useful, what about biomedical evidence?

Well, biomedical evidence starts with signs and symptoms.

“Signs” are things that a medic can observe objectively. Symptoms are things that are felt, and are thus inherently subjective.

With the Iranian events, the observable signs and reported symptoms are frustratingly vague.

Let me let you in on a little bit of a secret. Lots of chemical exposures result in fairly generic signs and symptoms.

Things like nausea, headache, irritated eyes and skin, vomiting, shortness of breath, fatigue, etc. are commonplace across many types of exposures to thousands of different substances.

But they are practically pointless in trying to identify an unknown substance(s) that may or may not be responsible in the absence of other information.

You also have to factor in concomitant problems. The situation in Iran is stressful.

Someone is going to yell at me and claim I’m patronising, but what I need to raise is valid and well established in the scientific literature.

Someone who suddenly feels unwell for an unknown cause in a stressful environment may feel symptoms aggravated by anxiety, fear, and possibly panic reactions.

For example, a slight headache can become far worse if you think that someone has inflicted this upon you.

New symptoms can appear. If you are experience nausea and a serious headache for no apparent reason, it’s not unheard of to start to hyperventilate.

Some of the signs and symptoms reported in Iran – numbness or tingling in extremities, partial temporary paralysis, can be plausibly attributed to hyperventilation. It's called hypocapnia

nejm.org/doi/full/10.10…

When you hyperventilate, the CO2 levels in your blood change and your blood chemistry temporarily changes. Paraesthesia and tetany can result. I’ve seen it happen.

Do people in stressful conditions hyperventilate? Do schoolchildren hyperventilate? Yes they do.

So, for some of these signs and symptoms, we must rule out hyperventilation as a differential diagnosis

Now, am I pointing a finger and saying that this is all "mass hysteria"? Psychogenic illness in group settings is a recognised phenomenon and it's not my specialty.

Could it be a factor? Certainly.

Could it amplify an existing situation? i.e. A few people get poisoned and signs and symptoms proliferate as fear and anxiety works as a multiplier? Certainly can't rule this out.

But what about physical biomedical samples? Hair, urine, blood? Or swatches of clothing? Sometimes that sort of thing is revelatory. It was very much of use in the various Syrian Sarin incidents.

The thing is, unless you have a bit of an idea of what you are looking for, such testing can be useless.

Routine blood and urine work will not reveal many of the chemicals that could be, in theory, used as an irritant.

Some chemicals are nasty or at least unpleasantly irritating at very small concentrations. The amount of material in blood, hair, urine, etc may be too small to detect

Getting back to the original point I made about toxic chemicals being reactive - the way these substances ("xenobiotics" if you want to be elegant about it ) work is that they react with things to cause bad effects

Therefore, after a poisoning, you may need to also look for the reaction products, not the substance itself.

But the list of things that are plausibly nasty and irritating enough to make people sick runs into the hundreds of thousands of chemical compounds

It's asking a bit much of a regional hospital or clinic to try to ferret out anything but the most obvious culprits.

There are also procedural issues. For biomedical evidence to actually provide useful information, we need the same standards as for environmental evidence, i.e. chain of custody, tying the sample to an actual time and place, etc.

I could go on and on about the problems of biomedical evidence, but I'll leave you with the basic point: the people analysing the sample generally either cannot look for the right things or do not know how to look for the right things, because this is a niche specialty

Now, there have been some specific claims emerging out of Iran. Some of them are ridiculous.

Some people have alleged the use of the nerve agent Tabun. As best I can tell, this is for two reasons only.

Some people are likely looking at general military manuals about chemical weapons and seeing that various books give Tabun a fruity smell.

Second, Tabun has track record of having been used by the Iraqis in the Iran-Iraq war. Iranians died from Tabun. So, there's probably some historic memory going on there.

But we've seen no actual evidence of anything approaching a death or serious signs and symptoms of Tabun. If we had, I'd be doing a whole different thread.

And I know a bit more about the smell of Tabun. The smells reported in Iran don't tally. And even if they did, we'd need more than just a smell to go off of.

The latest news is that the regime is blaming nitrogen and that nitrogen has been detected in air samples. This is dumb beyond belief. Air is 78% nitrogen.

Some have blamed various tear gasses. Now, this is plausible in some ways, particularly as poorly made tear gas munitions actually give off a lot of junk other than just tear gas.

Here I must footnote, erm... my own self. Here's a cite:

Kaszeta, Dan. "Restrict use of riot-control chemicals." Nature 573.7772 (2019): 27-29.

Some culprits suspects in Afghanistan were pesticides. One cannot necessarily rule that out.

I expect further developments, but I don't expect necessarily to get further clarity. There's simply not the evidence base to work with.

We have to accept the distinct possibility that we will not know what happened or that, actually, multiple different things happened and we're muddling them up together.

Another thing: several well meaning people have asked me for advice on how locals can investigate these instances. I am reluctant to dispense such advice.

Well-intended amateur efforts to try to collect evidence are more likely to get someone hurt or in trouble than produce any useful information.

Now, please don't construe anything I said as being pro-regime or anti-democracy or whatever. Nor am I trying to make light of human suffering. No doubt someone is going to have a go at me over this thread.

Save your oxygen. Whatever insult you hurl, I've probably heard it already.

TLDR; sometimes people expect an easy answer in scenarios where none is likely.

Adding this to the thread for clarity

https://twitter.com/DanKaszeta/status/1631987886886203392