Serial Killers and Statistical Blunders - Why Lucy Letby might be wrongly imprisoned: John O'Quigley

John O’Quigley is professor of statistics in the Department of Statistical Science, University College London. He has worked in the Laboratory of Probability, Statistics and Modelling, University of Paris – Sorbonne and in the Department of Mathematics, University of California San Diego as Director of the Laboratory for Mathematics and Statistics. Also at University of Washington, Seattle, the University of Virginia Medical School and Dept of Mathematics at Lancaster University, U.K. He has multiple publications in the field, more than I could count. QUESTIONS CONSIDERED 1) Can we start with a summary of Letby's case. What was she convicted of and what was the evidence presented that convicted her? You mention similar cases Sally Clark – 2 babies died of SIDS / 3 years in prison then overturned. Roy Meadows / she died aged 42 Lucia de Berk – found guilty of 7 murders and 3 attempted murders in 2003 and after 6 years in prison was fully exonerated in 2010 2) What is the work you have done on this, and why did you have an interest? Police said statistics didn’t play any part in the conviction - How do we decide what is an unusually high number? How ca it be measured? Can you say something about the shift chart? Letby and de Berk were shown to be the only ones who were present at all the events 3) 3 What was your methodology and your findings? What are some of the common statistical blunders made in court cases? What is the lottery blunder? What constitutes a spike in the death rate? How do we know it didn’t arise by chance? 4) Were you disappointed it was rejected for publication? I believe your paper was rejected and the reason given was it might upset the victims? 5) Are there any implication for those trying to publish papers on controversial issues? What is the state of science today? Is there is still a focus on finding truth or is that taking second place 6) What should happen in the Letby case now? How can we avoid poor thinking in criminal cases in future. PAPERS QUOTED For the two papers that were mentioned, here are the links. Logical and statistical errors in the investigation and prosecution of suspected serial killer nurses: https://osf.io/nk9da Use of roster charts in the investigation and prosecution of nurses suspected of inflicting deliberate harm on patients: https://osf.io/hqzrw