July 2026: Government declines to widen the Thirlwall terms of reference (16 July) · new 100-page insulin report to the CCRC challenging the trial evidence (9 July) · Thirlwall report still expected no earlier than September · inquests relisted to 2027 · Shoo Lee Panel: no medical evidence of deliberate harm.
Statistical explainer
A shift-rota chart visualises which staff were present when a defined set of events occurred. As a piece of statistical reasoning, a shift chart only carries weight when the events are defined independently of the staff being investigated and when the comparator base rate is known.
A shift chart was central to the prosecution presentation. The Royal Statistical Society and named professors of statistics (Gill, Fenton, Hutton, Spiegelhalter, Green, O'Quigley) have published critiques arguing the chart suffers from selection bias, base-rate neglect and Texas-sharpshooter logic.
A chart mapping 25 significant clinical events against the nursing rota was central to the Crown's presentation. It showed Letby on duty for every one of them, and that pattern was offered to the jury as demanding an explanation.
The Royal Statistical Society and named professors of statistics — Gill, Fenton, Hutton, Spiegelhalter, Green and O'Quigley — have published critiques arguing the chart suffers from selection bias, base-rate neglect and Texas-sharpshooter logic. Prof. John O'Quigley's peer-reviewed analysis (Medicine, Science and the Law, December 2025) concludes the chart should not have been shown to jurors; Prof. Richard Gill's June 2026 critique disputes O'Quigley's modelling while endorsing that conclusion.
The critique is of the chart's inferential weight, not of the underlying rota data, which is not disputed. What a properly constructed comparison would have shown was never put before the jury and cannot now be reconstructed from the trial record alone.
A visualisation of which staff were on duty when a defined set of events occurred. It carries statistical weight only if the events were defined independently of the staff being investigated and a comparator base rate is known.
Critics argue the events were selected because of their association with the suspect, and that no comparator base rate for nurse-shift presence at adverse events was presented.
No. The dispute is about what can be inferred from the chart, not about who worked which shift.
Not entirely. Gill's 2026 critique disputes O'Quigley's modelling while agreeing with his conclusion that the chart should not have gone before jurors.
That collapses occurring on shifts Letby did not work were not part of the chart. Including them changes the denominator the pattern is measured against.
Found an error or missing source? Send a correction with the page URL and supporting source.