Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

thejusticegap.com

Post reply on HN
Print view View original

41–50 of 93 posts

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#41

Post by avianlyric »

rwmj wrote: post #11

While the statisticians are right to be cautious about clusters of deaths => serial killer, we also have to remember that a 9 month trial presented lots of evidence to a jury, they deliberated for over a week, and returned seven guilty verdicts. Unless you were also sitting at that trial or have read every line of court transcript, it would be very unwise to start claiming there's a miscarriage of justice.

The other half of the equation is bias in the investigation. Often a poor understanding of statistics is used to start investigations, unfortunately those investigations are started with the premise that a particular person “is probably guilty”. Turns out that when medical investigations are performed looking for wrongdoing, they’re much more likely to find wrongdoing that if the investigation was performed simply to understand if correct policy and procedure are followed. As a consequence it’s very normal for deaths previously ruled “normal” to suddenly become “suspicious”.

When people have studied these types of investigations, and has the evidence re-evaluated by unbiased investigators, the results are usually that nothing untoward occurred, or that deaths are better explained by systemic failures rather than the actions of specific individuals.

There’s a good reason why we don’t let police go on fishing expeditions. Medical investigations, frequently initiated by untrained administrators, and performed by individuals within a hospital with no formal training in performing unbiased investigations, nor given enough to perform proper investigations, simply don’t hold up to the standards we would expect from a criminal investigation.

None of this is to say that the Letby verdict is unsafe, but it wouldn’t be the first time a medical investigation with lots of evidence resulted in an awful miscarriage of justice.

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#42

Post by fiftyacorn »

One thing the Letby trial left me wondering was how do doctors and nurses working in areas with higher levels of mortality protect themselves from this sort of accusation? And whether this would put doctors off doing riskier procedures?

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#44

Post by littlestymaar »

quietbritishjim wrote: post #13
rwmj wrote: post #11

While the statisticians are right to be cautious about clusters of deaths => serial killer, we also have to remember that a 9 month trial presented lots of evidence to a jury, they deliberated for over a week, and returned seven guilty verdicts. Unless you were also sitting at that trial or have read every line of court transcript, it would be very unwise to start claiming there's a miscarriage of justice.

Those arguments included that the suspicious deaths all lined up perfectly with Lucy Letby's shifts, that she was often caught interacting with those babies when she had been instructed not to, and that when she was taken off the ward the deaths instantly stopped. That is a level of causality checking far stronger than most Bayesian analyses.

> that she was often caught interacting with those babies when she had been instructed not to

This is only a piece of evidence if:

- it's not a common behavior among nurses in her unit.

- she did that specifically to the victims.

Otherwise it could be that this rule isn't applied in practice, or that she is just the kind of people who don't really listen to instructions if they think they should not, which albeit a questionable attitude in an hospital isn't the same thing as “she killed them”.

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#45

Post by littlestymaar »

automatic6131 wrote: post #23
quietbritishjim wrote: post #13

Earlier quoted context omitted.

Those arguments included that the suspicious deaths all lined up perfectly with Lucy Letby's shifts, that she was often caught interacting with those babies when she had been instructed not to, and that when she was taken off the ward the deaths instantly stopped. That is a level of causality checking far stronger than most Bayesian analyses.

And she took home documents from several of those patients, which she'd have no ordinary reason to have fished out of the trash.

Like with the unauthorized interaction with the babies[1], this is only relevant to the case if (1) it's not a common behavior among nurses in her unit and (2) she did that specifically to the victims, and not with many of her patients.

[1] https://news.ycombinator.com/item?id=37810382

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#46

Post by jdietrich »

fiftyacorn wrote: post #42

One thing the Letby trial left me wondering was how do doctors and nurses working in areas with higher levels of mortality protect themselves from this sort of accusation? And whether this would put doctors off doing riskier procedures?

The suspicious deaths in this case were systematically ignored by hospital management. Doctors who raised concerns about Letby were disciplined for doing so. The hospital is now under police investigation on suspicion of corporate manslaughter, because it is highly likely that deaths could have been prevented if they had investigated rather than silenced those concerns.

https://www.bbc.co.uk/news/uk-66120934

https://www.bbc.co.uk/news/uk-england-merseyside-67006930

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#47

Post by mrangle »

rwmj wrote: post #11

While the statisticians are right to be cautious about clusters of deaths => serial killer, we also have to remember that a 9 month trial presented lots of evidence to a jury, they deliberated for over a week, and returned seven guilty verdicts. Unless you were also sitting at that trial or have read every line of court transcript, it would be very unwise to start claiming there's a miscarriage of justice.

No one forgets. 'We have to remember the jury" isn't a persuasive argument. Juries are no match for statistical misrepresentation. Even beyond stats, juries get it wrong constantly. Courts don't have staff math experts, including the judge. Were expert witness mathemeticians trial participants at any point? It may be premature to claim that there was a miscarriage of justice, but it may be similarly unwise to imply that there likely wasn't.

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#48

Post by kitd »

crote wrote: post #34
kitd wrote: post #16

Earlier quoted context omitted.

On the afternoon of the conviction, there was an expert criminal psychiatrist on BBC radio who had been following the trial closely. Apparently those notes, though they made all the headlines, played little part in her conviction, and are typical for someone suffering stress &| depression. The thing that did for her apparently were messages she had sent some of the parents, the times of which contradicted her testimo…

That's a very good reason to warn the public about it, is it not? People generally do not read court transcripts, they read media headlines. It's basically a guarantee that some people are already convinced their statistically unlucky coworkers are also serial killers. It can easily lead to witch hunts.

You're right, but I think the article is about the RSS warning the official inquiry rather than the public. One would expect them to have a higher threshold of quality of evidence than the typical tabloids or Twitter feeds.

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#49

Post by munch117 »

rgrieselhuber wrote: post #39
rwmj wrote: post #11

While the statisticians are right to be cautious about clusters of deaths => serial killer, we also have to remember that a 9 month trial presented lots of evidence to a jury, they deliberated for over a week, and returned seven guilty verdicts. Unless you were also sitting at that trial or have read every line of court transcript, it would be very unwise to start claiming there's a miscarriage of justice.

Also “experts warn” is argumentum ad verecundiam.

What's the latin name for "argument by using fancy latin names for things that could just as easily have been said in English"?

Re: Experts warn Letby inquiry of misreading stats to explain 'suspicious' deaths

#50

Post by lolinder »

rgrieselhuber wrote: post #39
rwmj wrote: post #11

While the statisticians are right to be cautious about clusters of deaths => serial killer, we also have to remember that a 9 month trial presented lots of evidence to a jury, they deliberated for over a week, and returned seven guilty verdicts. Unless you were also sitting at that trial or have read every line of court transcript, it would be very unwise to start claiming there's a miscarriage of justice.

Also “experts warn” is argumentum ad verecundiam.

No, the headline is just documenting what the experts are saying, it can't contain a fallacy because it doesn't contain an argument. It's the experts themselves who are making the argument, and they're not simply saying "trust us, we're statisticians".
Post reply on HN