Pro Norman Fenton on probability and law
Worth a listen https://m.youtube.com/watch?v=RVQ6kReMLdw
An unusual pattern: Is Benjamin Geen a killer or England's unluckiest man?“Norman Fenton is a professor of risk management information at Queen Mary London University.
And Ben Geen, he says, might be the victim of what he terms an 'exaggerated coincidence'.
'What you've got is apparently a highly unusual sequence of events happening at this particular hospital.
'If you looked at that individually—what is the probability that this particular nurse would be present at 18 respiratory arrest events at the same hospital within such a short period of time—the probability is incredibly low.'
But Fenton is 'almost certain' that similar patterns would have occurred in a given two-month period in thousands of hospitals across the UK.
'Think of it this way,' Fenton says. 'In most hospitals, the nurses tend to be on duty for at least 50 to 70 per cent of the recorded time.
'For any given nurse to be present at all of those events is still—again—a very, very low probability.
'But what you've got is, let's say, 100 nurses at any given hospital where one of these sequence of events has been observed—again, there is quite a high probability that at least one of them would have been present at all, in this case 18, such events.'
In other words, statistically speaking, Geen won the lottery twice—but it was the wrong kind of lottery.
When Geen's team appealed his conviction on statistical grounds, the court didn't agree with Hutton or Fenton's analyses, arguing that there was no need to statistically validate the opinion evidence of experts.
It's worth noting: neither Hutton nor Fenton were arguing that Geen was innocent, merely making the case for greater statistical rigour.
The court claimed that 'academic statistical opinion, however distinguished, is divorced from the actual facts'.
After all, real people—patients, jurors, healthcare professionals in the high pressure environment of A and E—have to react to a lot more than raw data.
But in the case of Geen, what were they reacting to?
The spectre of Harold Shipman still very much haunts the UK healthcare system.
Shipman, nicknamed 'Dr Death', is the most prolific serial killer on record—arrested in 1998 and sentenced to life in prison by 2000, he was confirmed by an inquiry to have killed 218 of his patients, though it's suspected the true number could be as high as 250.
Shipman hanged himself in his cell on 13 January, 2004—just 23 days before the two patients that sparked the Horton General Hospital investigation suffered respiratory arrest.
'Since the Shipman inquiry, as soon as there's an accusation, you've got to think dirty,' says Dr Wendy Hesketh, who is writing a book about medico-crime.
'If you don't, you could find yourself having to answer as to why you didn't do a full investigation.'
Hesketh points out that in Geen's case very few incidences of respiratory deaths were considered suspicious when they occurred.
'Beforehand, they weren't flagged up to anyone, no one was suspicious, no one thought that there was any untoward behaviour or that anyone had been unlawfully killed.
'All of a sudden you've got this team who are actually potential witnesses themselves, and they are talking to each other and reinforcing their own suspicions.
Read more here:
https://www.abc.net.au/radionational/programs/healthreport/benjamin-geen-an-unusual-pattern/7287620Ben Geen: another possible case of miscarriage of justice and misunderstanding statistics?http://probabilityandlaw.blogspot.com/2014/11/the-ben-geen-case-another-miscarriage.htmlhttps://www.crcpress.com/Risk-Assessment-and-Decision-Analysis-with-Bayesian-Networks/Fenton-Neil/p/book/9781138035119http://probabilityandlaw.blogspot.com/