Expert Witness Journal Issue 66 April 2026 - Flipbook - Page 18
Testing the Evidence, Not the Expert: What
De Francisci v Hampshire Hospitals NHS
Foundation Trust Teaches About Causation
by Dr. Carolina Stamboulid, PhD.; Founder and Scienti昀椀c Evidence Analyst,
at Epistēmē Scienti昀椀c Consulting
De Francisci v Hampshire Hospitals NHS Foundation
Trust is a clinical negligence case that warrants
close attention from expert witnesses and those
who instruct them. Breach of duty was admitted,
the relevant clinical guidance was accepted, and the
underlying biological mechanism was not in dispute.
Yet causation failed.
Answering that question required the court
to examine what the medical literature could
legitimately demonstrate, rather than what
might appear biologically plausible or intuitively
persuasive.
What makes this judgment particularly instructive
is not its outcome, but the court’s reasoning. The
case did not turn on a preference for one expert
over another. Instead, the court examined the
type, quality, and methodological robustness of the
scienti昀椀c literature relied upon by each expert and
determined which body of evidence was capable of
answering the causation question on the balance of
probabilities.
Both experts were experienced obstetricians
drawing on reputable research. Although they
reached di昀昀erent clinical conclusions, the divergence
in opinion re昀氀ected the di昀昀erent bodies and types
of scienti昀椀c evidence upon which those conclusions
were based.
Two Experts, Two Evidence Bases
The claimant’s expert placed primary reliance on the
meta-analysis by Roberge et al. (2017), supported by a
small randomised trial by Ebrashy et al. (2005). This
evidence base relied predominantly on aggregate
data (AD), with conclusions drawn from study-level
summaries and subgroup analyses. It emphasised
large protective e昀昀ects of early aspirin, with severe
pre-eclampsia outcomes used as a proxy for HELLP
syndrome.
For expert witnesses, the judgment shows a clear
judicial willingness to look beyond headline
conclusions and interrogate study design, statistical
robustness, and applicability to the individual
claimant. For solicitors, it highlights the value of
testing the evidential foundations of a case at an
early stage, before signi昀椀cant costs and positions
harden.
The defendant’s expert relied instead on the
Cochrane review by Duley et al.(2019), together with
the individual participant data (IPD) meta-analysis
by Meher et al. (2019), which pooled raw patient-level
data across trials involving approximately 40,000
women. This evidence base provided the most robust
assessment of overall e昀昀ect size and timing, while
the Cochrane review examined HELLP outcomes
directly.
The Case in Brief
The claimant, a woman with polycystic kidney
disease, developed HELLP syndrome during
pregnancy. The Trust accepted that aspirin should
have been advised at 12 weeks in accordance with
guidance from the National Institute for Health and
Care Excellence for women at higher risk of preeclampsia. Instead, aspirin was commenced at 23
weeks.
The court’s task was not to decide which expert
was more persuasive. It was to determine which
body of literature was methodologically capable of
supporting a conclusion on causation on the balance
of probabilities.
The central question for the court was therefore
narrow but demanding: if aspirin had been started at
the correct time, would this claimant, on the balance
of probabilities, have avoided HELLP syndrome?
EXPERT WITNESS JOURNAL
16
APRIL 2026
