Ben Geen was caught red-handed – a serial killer, or so it seemed. In the third of his series The Justice Trap, Jon Robins shows how a misreading of statistics led to a life sentence
Timothy Stubbs was admitted to Horton Green General Hospital, Banbury in Oxfordshire on the afternoon of Thursday February 5th, 2004. The 42-year-old arrived complaining of abdominal pains, within hours he was fighting for breath and put on a life support machine.
It was the second time that day that a patient appeared to suffer an unexpected respiratory arrest. Over the previous two months some 18 patients had mysteriously stopped breathing. Timothy Stubbs was one of two patients to die.
According to the prosecuting barrister, the syringe was the ‘smoking gun’.
Staff were instantly suspicious. A couple of doctors had a theory. When an ambulance arrived at A&E, more likely than not staff nurse Ben Geen was there to greet it. As the only male nurse in the department, the 25-year-old stood out. A lieutenant in the Territorial Army, he was considered by colleagues ‘a bit gung-ho’.
The day after Thomas Stubbs was admitted, Ben Geen rang in ill. Senior staff called in doctors from nearby John Radcliffe Hospital in Oxford and together they went through case notes. By the end of the weekend, the team identified 18 patients as having suffered an unexplained respiratory arrest, hypoglycaemia or both. On each occasion Geen had been on duty.
The following Monday, Ben Geen was arrested. As he arrived for a night shift, he was called into a meeting. He had a syringe in his jacket pocket and discharged its contents into its lining.
‘Angel of Death’
On 10 May 2006, Ben Geen received a prison sentence of a
‘Like an arsonist who sparks a blaze and is then transfixed as the firefighters arrive, Geen would make his patients collapse and then join the adrenalin-charged frenzy as doctors and nurses battled to save them,’ one newspaper said.
At his three-month trial, it was alleged that incidents of respiratory arrest in A&E were extremely rare and, unheard of, without a cause. The Crown’s lawyers argued that ‘an unusual pattern’ had emerged. According to the prosecuting barrister, the syringe was the ‘smoking gun’. His jacket tested positive for a muscle relaxant and an anaesthetic. The syringe was said to be old and worn from repeated use.
The medics misread an innocent statistical blip as evidence of a serial killer.
There is a phenomenon known as ‘prosecutor’s fallacy’ where people – lawyers, jurors and the judiciary – confuse the probability of finding the evidence on an innocent person with the probability that a person on whom the evidence is found is innocent.
Ben Geen is currently in HMP Long Lartin, Worcestershire. His supporters, including his barrister Mark McDonald, argue that no crime was ever committed. The medics misread an innocent statistical blip as evidence of a serial killer.
Six years earlier, a jury found Britain’s most prolific serial killer, Harold Shipman, guilty of 15 murders. McDonald believes that senior hospital staff convinced themselves in the wake of Shipman hysteria that there was ‘a maniac on the loose’ (as the prosecution put it).
Guilty Until Proven Innocent
McDonald has represented Ben Geen on a pro bono basis for ten years. I asked him about the ‘smoking gun’. Before MacDonald became a lawyer he used to work in A&E. ‘I used to walk backwards and forwards with a load of rubbish in my pockets, including syringes.’ What about the discharging of the syringe? ‘It could be that he just knew he shouldn’t have been carrying it,’ he says.
In a report prepared for Ben Geen’s 2009 appeal, a medical statistician Professor Jane Hutton wrote that the evidence of an unusual pattern was of ‘no value in supporting a conclusion that there was an unusual pattern, nor a conclusion that any unusual pattern was not a chance event’.
The biased evidence-gathering at the Horton Hospital was conducted for one reason: making a murderer.Ben Geen’s Father
The Court of Appeal refused to hear her evidence. ‘We are satisfied the jury was entitled to, and perfectly capable of, drawing proper inferences from the evidence that this pattern of events was as a result of deliberate action rather than mere chance,’ said Lady Justice Hallett.
An application to the miscarriage of justice watchdog the Criminal Cases Review Commission was backed by five leading statisticians including Sir David Spiegelhalter, of Cambridge University and Professor Norman Fenton, of Queen Mary University of London.
Speaking at the launch of Guilty Until Proven Innocent last year, Prof Fenton argued that the cluster of events was not at all unusual. ‘Given the number of hospitals in the UK, in any four-year period it is almost certain that there will be several instances of exactly this kind of “abnormally high” sequence of respiratory events,’ he said. ‘It is actually quite likely that, purely by chance, in at least one such sequence there will be the same nurse present at each event.’
I interviewed Prof Hutton. She had been ‘surprised and disappointed’ not to be called to give evidence in the Court of Appeal. ‘I felt it was rather arrogant and the court simply didn’t take the point,’ she told me.
Making a Murderer
The case has now been with the CCRC for almost six years. I spoke to Ben Geen’s father this week. ‘We have serious concerns about the way that the prosecution’s medical experts relied on anecdotal and opinion evidence without any statistical basis to secure Ben’s wrongful conviction,’ he said. ‘The biased evidence-gathering at the Horton Hospital was conducted for one reason: making a murderer.’ He complains that the NHS Trust’s reluctance to meet with the watchdog has created unacceptable delays.
The eminent statisticians do not pass comment on Ben Geen’s likely guilt or innocence (although another statistician Richard Gill, of the University of Leiden in the Netherlands does and sees an uncanny parallel with the Lucia de Berk case).
What they experts agree on is that there was no evidential basis for the prosecution claim of an ‘unusual pattern’. There is widespread concern about the courts failure to get to grips with probability and how wrong assumptions drive prosecutions.
‘If you look at the way these cases are prosecuted, the statistics by no means dominate,’ Dr Fenton told me. ‘It’s all about the medical evidence. It doesn’t look like it has anything to do with a problem to do with statistics. But it is the statistics that initially drive the case.’