Talk:List of laboratory biosecurity incidents

1906

Shibbolethink why remove 1906 Plague case? It is textbook case of biosecurity mishap. --Francesco espo (talk) 00:23, 20 July 2021 (UTC)

It's a contaminated vaccine, which is an iatrogenic medical mistake made in a clinical laboratory. An analogous example would be the Cutter incident, which we do not cover here.--Shibbolethink ^{(♔ ♕)} 00:38, 20 July 2021 (UTC)

Might be worth it to note in the lede that the list does not contain such entries, with, perhaps, a link to an article that does. Bonewah (talk) 13:31, 20 July 2021 (UTC)

I support inclusion of the incident in this article, and agree with Bonewah that if it isn't, then the lede should be clear as to what types of incidents are not included, and link to an article that does include that type of thing. I disagree with the edit removing the incident. Also, whenever you do remove substantial amounts of material from an article, or any material that includes a citation (i.e. removing the citation from the article), please also state clearly in the edit summary that you're doing so; e.g. "removing Plague incident" rather than just "not an LAI", which could be put on the talk page as the reason for the edit but is not a description of the edit and doesn't help much at all if someone is searching the page history to find what was deleted. How did the vaccine get contaminated? Was it contaminated in a laboratory? Was it contaminated in a clinical laboratory or in some other type of laboratory? If some types of laboratories are excluded from this article the lede should say so. It seems to me that the contamination and/or transmission occurred in a laboratory so it should be included. ☺Coppertwig (talk) 19:32, 7 December 2021 (UTC)

What counts as a laboratory? Are all factories which produce vaccines also laboratories?

I think this article should be exclusively about "research laboratories" as that is what most people are thinking about when they hear the word "laboratory" in this context. That would exclude "manufacturing laboratory" and "clinical laboratory" which do not undertake original research.

There could easily be a separate article about "pharmaceutical contamination incidents" which would include these other laboratories. — Shibbolethink ^{(♔ ♕)} 00:18, 8 December 2021 (UTC)

Date format

Dates given as e.g. "2003-08" are disallowed by the MOS; also I found it quite confusing. I thought it meant from year 2003 to year 2008, thought is was incorrect, and was going to change it or propose changing it to just "2003". Is it supposed to mean August, 2003? If so, I think the only format allowed by the MOS is either "August 2003" -- or "Aug 2003" if brevity is required. I see that it says yyyy-mm-dd at the top of the table, but first of all the reader doesn't necessarily notice this, and secondly since the day is missing it's obviously not actually yyyy-mm-dd, so interpreting it as a range of years is still quite plausible even if the heading is noticed. ☺Coppertwig (talk) 18:48, 7 December 2021 (UTC)

bitten by a ferret

I disagree with this edit by Shibbolethink which deletes an incident where someone was bitten by a ferret. I agree with user 79.70.179.144 who said "regardless of whether a transmission was confirmed or not, its still an "incident"" in edit summary when reverting and restoring the incident to the page. The lead doesn't say that only incidents with confirmed transmission are included. If bites are numerous and excluded from the list, I'd like to see a source stating that bites are numerous and a statement in the lead that bites are numerous and are excluded from this list. With the lead as it currently is I support inclusion of the incident; and if bites are to be excluded perhaps the article title needs to be changed too. ☺Coppertwig (talk) 19:15, 7 December 2021 (UTC)

Here are the sources you requested: [1] [2] [3]

I don't think bites should be excluded. I think events where no transmission occurred should be excluded. An analogy: Hundreds of thousands of people are in almost car accidents every day, but no one cares about a list of those "incidents." They only care when a car actually hit another car. If no transmission occurred, why would that be a notable incident?

Another analogy: If I bump your car in a line of cars before a stop light, but no scratches are left behind, very few if any drivers would report that incident to their insurance. It just wouldn't be interesting or worthwhile to do so. Why would anyone consider an incident where someone was bitten by a lab animal, but where no disease transmission occurred (a relatively common, not constant or ever-present, but relatively common occurrence) a notable enough incident to merit inclusion? — Shibbolethink ^{(♔ ♕)} 00:18, 8 December 2021 (UTC)

For the same reason mishaps involving nuclear weapons are notable even if they do not involve anyone dying or any further harm ([Bent Spear]), because the potential harm is far greater than any car accident could ever be. Moreover, the fact that these incidents are appearing in reliable sources is strong evidence that they are notable enough for inclusion. Bonewah (talk) 14:36, 8 December 2021 (UTC)

There is a huge difference between a mouse infected with yersinia pestis (aka Black plague) biting someone in a lab, and a nuclear event. You can literally go to Arizona and grab any prarie dog, and chances are that prarie dog is infected with yersinia pestis [4]. You can go out to any cow pasture, and poke most dead cows, or literally just dig up a few handfuls of soil, and there will be bacillus anthracis spores (AKA anthrax) present in a reasonable quantity. [5]

A majority of the experiments we're talking about involve the natural versions of these pathogens. (way north of 50%, though difficult to quantify. There are statistics on NIH grant proposals and how many need to get recombinant DNA approval) [6]. These things exist already in nature. The same is not true of nuclear fission on the scales found in nuclear weapons or nuclear power plants. That does not happen on earth except deep underground inside uranium mines, and even then rarely, such that we are only detecting it much later, by its results [7]. These pathogens, on the other hand, are infecting people in the wild all the time, and nothing happens on any appreciable societal level. ~ 7 people get infected with the black plague in the US each year. [8] Around 1-2k worldwide. Between 2 and 20,000 anthrax cases are happening worldwide every year, probably around 5 in the US [9].

/That/ is why just getting bit by such an animal is not that noteworthy. people also get bit by random wild animals, and it is not noteworthy. I agree it is noteworthy when transmission actually occurs, or when a huge lapse in security actually occurs, because of the laboratory environment and how special it is. But a random naive animal bite is so common, as demonstrated above, that comparing this to any sort of nuclear accident or near miss is a misunderstanding of the probabilities here. — Shibbolethink ^{(♔ ♕)} 02:12, 9 December 2021 (UTC)

I should further note that the lede includes what, in my opinion, would include a ferret bite or any other non-exposure near miss: "This list of laboratory biosecurity incidents includes accidental laboratory-acquired infections and laboratory releases of lethal pathogens, containment failures in or during transport of lethal pathogens, and incidents of exposure of lethal pathogens to laboratory personnel, improper disposal of contaminated waste, and/or the escape of laboratory animals." If we are going include only those incidents which resulted in an actual infection (which im opposed to), then most of the opening sentence would be wrong. My strong preference is to update the lede with a note that 'incident; in this context includes infection, potential release or notable mishandling (like the 2014 discovery of smallpox in White Oaks, for example). Bonewah (talk) 14:48, 8 December 2021 (UTC)

So just to clarify, Bonewah, do you believe that any and all times that an autoclave throws an error should be included on this list? Because that is a “potential exposure” by CDC standards. Even if every chemical and mechanical indicator shows that the contents of the autoclave were sterilized. Even if the temperature of the autoclave was several dozen degrees above the killing temperature of any included pathogen. Still technically an “exposure” if you have to open the door before the cycle is technically completed. Would that count in your mind?

What about if an animal that isn’t infected with any pathogen, but is held in the same facility as an infected animal, urinates on an investigator?

Personally I believe the only criteria should be WP:DUE. And none of these events would count as due because they likely would only be reported in primary sources. And If mentioned in RS, it would only be briefly and not in any plethora or predominance of sources. — Shibbolethink ^{(♔ ♕)} 18:00, 8 December 2021 (UTC)

I think if an incident is mentioned in reliable secondary source, then it should be included unless a very good case is made for exclusion. So in the this case, the ferret bite would be included as it was reported in a presumably reliable secondary source, the Times UK. I would note that i would consider scientific papers such as this to be secondary sources as well as they are not reporting the incident itself, but rather performing an analysis of same. Bonewah (talk) 18:55, 8 December 2021 (UTC)

Any secondary source whatsoever? Even a cursory random mention that is not substantially the subject of the article? That would go against WP:DUE. — Shibbolethink ^{(♔ ♕)} 02:16, 9 December 2021 (UTC)

Sounds like we are saying the same thing, that an incident is notable if it is covered in a secondary source so long as that coverage isnt trivial. I dont think an incident has to appear in multiple sources necessarily, just more than a passing mention. Bonewah (talk) 14:06, 9 December 2021 (UTC)

A Review of Laboratory-Acquired Infections

The above discussion brings up a good point with regard to one of the secondary sources we cite: A Review of Laboratory-Acquired Infections in the Asia-Pacific: Understanding Risk and the Need for Improved Biosafety for Veterinary and Zoonotic Diseases. In this source, we find a list of confirmed laboratory acquired infections (LAI's) from the Asia-Pacific region, most of which do not yet appear on our list. Do we transclude that list here? They all involve actual infection, not merely potential exposure as discussed above. Bonewah (talk) 14:13, 9 December 2021 (UTC)

Personally, I would consider these ‘’ passing mentions’’ because none of them receive a dedicated history or investigation. It’s just a review of a bunch of random incidents. We will soon overwhelm this article with random incidents because while the incidence of LAIs has been decreasing over time, it is still likely in the thousands overall.

I think we should have sentences in this article describing the overall frequency of LAIs over time among different geographic areas (e.g they are more common in Southeast Asia, likely as a result of lab infrastructure challenges).

But if we begin including an incident every time it’s mentioned in even a ‘’single’’ article, a list review like that, then we will quickly turn this list into a database more suitable for NCBI than Wikipedia. Wikipedia is not an indiscriminate list of things that have happened. It is only for ‘’notable’’ WP:DUE things. Correct me if I’m wrong (and I very well may be), but I thought DUE required more than just a single RS. It requires proportionality in the overall ‘’landscape’’ of sources. If this is one of many thousands of incidents… it should be afforded 1/1,000ths of space, aka not included. — Shibbolethink ^{(♔ ♕)} 17:42, 9 December 2021 (UTC)

Im inclined to agree, although the proportionality might be fairly low due to the small size of the landscape of reliable sources. I went ahead and updated the lede to note that this list is not all-inclusive and added some external links to papers on the subject and the LAI online database for the interested reader. Bonewah (talk) 18:46, 9 December 2021 (UTC)

I think it'd be reasonable to limit our list to items that have their own Wikipedia articles or are at least mentioned in other Wikipedia articles. That is, we should be able to fill in the "main article" column for each row in the list. –Novem Linguae (talk) 06:15, 10 December 2021 (UTC)

On second thought, a lot of the current entries don't have a main article, and I don't think we want to delete them. So hard to say what a good inclusion criteria would be. –Novem Linguae (talk) 06:18, 10 December 2021 (UTC)

use leak dates, not symptom dates

The prion leak row is listed as 2019. Yet that is the date the researcher died, not the date of the leak, which is listed as 10 years earlier. Recommend using 2009 as the date for this incident. 2600:1700:8A14:1A70:7DB3:4086:EF27:B9C5 (talk) 22:12, 3 March 2023 (UTC)

Part of the issue is that we don't really know that's when the exposure was, since Prion diseases are so insidious. — Shibbolethink ^{(♔ ♕)} 22:18, 3 March 2023 (UTC)

2016 Chikungunya

https://theintercept.com/2022/11/01/biosafety-lab-accident-chikungunya-virus/ 72.222.92.103 (talk) 23:21, 9 December 2023 (UTC)

Possible Incident

https://www.timeshighereducation.com/news/police-investigate-death-spanish-researcher

https://english.elpais.com/science-tech/2023-10-20/an-investigation-has-been-opened-into-the-death-of-a-scientist-who-was-studying-a-transmissible-and-deadly-disease-in-spain.html?ssm=TW_CC 72.222.92.103 (talk) 23:31, 9 December 2023 (UTC)

High level lab breach

In Canada top level scientist who worked on viruses such as Covid-19, Ebola, and other have been fired for shipping viruses to the wuhan lab of virology among other top level Chinese institutions in early 2019. They have also allowed access to Canadian level 4 laboratories to top Chinese officials and have been working with the Chinese government for years https://www.theglobeandmail.com/files/editorial/politics/nw-na-labs/winnipeg-scientists-doc.pdf 2001:1970:4AE5:A300:5447:C8DA:5C49:E5A8 (talk) 03:26, 11 March 2024 (UTC)