Article published: 16 November 2026
Key points
- Retrospective diagnosis must be supported by evidence relevant to each diagnostic criterion; Detailed transaction records do not necessarily establish subjective motivations or experiences; The court preferred the psychiatrist who explained the limits of the available evidence; A qualified or inconclusive opinion may be more helpful than an unsupported firm conclusion; Expert disagreement on reasoning does not itself imply bias or misconduct
How much can an expert really conclude when the evidence does not tell the whole story? In Ashton & Ors v TSE Malta LP [2026] EWHC 2935 (KB), two psychiatrists considered substantially the same records of an individual’s gambling activity but reached sharply different conclusions about whether the diagnostic criteria for gambling disorder had been met. The High Court preferred the more cautious opinion, largely because the available records could not establish the thoughts, motivations and experiences central to several of those criteria.
The background
The proceedings followed the death by suicide of Luke Ashton in April 2021. His family brought a negligence claim against TSE Malta LP, the operator of the Betfair exchange, arguing that the company should have recognised the signs of gambling disorder and taken steps to intervene. The case involved substantial evidence about Mr Ashton’s betting history, including the frequency of bets, deposits, losses, changing deposit limits and use of the platform. It was a deeply serious and distressing dispute, but the expert-evidence question has implications well beyond gambling-related litigation.
Mr Justice Swift considered the scope of any common-law duty of care and whether a psychiatric injury of the relevant kind had been sufficiently foreseeable. A necessary part of the claimants’ case was that Mr Ashton had suffered from gambling disorder as defined in DSM-5. The evidence about that diagnosis became central to the outcome.
Two experts, very different conclusions
The claimants’ psychiatric expert, Dr John Barker, considered that eight of the nine DSM-5 diagnostic criteria were satisfied, indicating severe gambling disorder. The defendant’s expert, Dr Humphrey Needham-Bennett, considered that only two criteria were established, a further criterion might be satisfied, and the available evidence was insufficient to assess most of the remainder. Both were attempting a retrospective assessment: Mr Ashton had not been diagnosed with gambling disorder during his lifetime, and it was not possible to obtain the kind of direct clinical history that would ordinarily inform an assessment.
The disagreement was not principally about whether the account records existed or what they recorded. It concerned what those objectively recorded behaviours could reliably show about the person’s internal experience. Frequent betting, financial losses and changes in account settings may be important evidence, but a clinical diagnostic criterion can require more than proof that particular transactions took place.
What the records could not tell the court
Some diagnostic criteria concern motivations, emotional states and the reasons for gambling. The transaction records could describe what Mr Ashton did on the platform, but they did not necessarily establish why he did it, how he felt when doing so, or the subjective symptoms required to satisfy particular criteria. Nor did the absence of a direct clinical assessment become less important simply because the volume of digital data was substantial.
Dr Barker drew inferences from the betting evidence in favour of satisfying numerous diagnostic criteria. Mr Justice Swift considered that this reasoning did not overcome the absence of reliable information about Mr Ashton’s reasons, motivations and emotional experience. At paragraph 76 the judge described the claimants’ expert’s reasoning as unconvincing on that question. By contrast, at paragraph 77 he regarded Dr Needham-Bennett’s analysis as careful, considered and thorough and accepted it.
The court concluded that the evidence did not establish gambling disorder under the relevant diagnostic criteria. That meant there was no adequate evidential basis for finding that the defendant knew, or ought to have known, of the relevant disorder. The claim was dismissed. That conclusion was a determination of the evidence and legal issues before the court; it should not be treated as a general statement that gambling behaviour cannot cause serious harm or that such harm can always be assessed from account records alone.
The difference between inference and assumption
Experts are often asked to reach opinions from incomplete information. In medicine this may involve reviewing records after the relevant events; in other disciplines the expert may have to reconstruct a transaction, an accident, a building’s condition or a sequence of decisions from the documents that remain available. Drawing reasonable inferences from that information is part of the professional task. The difficulty is knowing when the evidence genuinely supports an inference and when it leaves a gap that cannot safely be filled.
There is no simple rule that the more information an expert has, the firmer the conclusion ought to be. Thousands of entries may describe one aspect of an event in considerable detail while leaving another, essential aspect largely unknown. Here, the records were useful evidence of gambling activity, but they were not a substitute for evidence about psychological motivations. The judge’s treatment of the competing psychiatric opinions illustrates why an apparently comprehensive dataset must still be tested against each element of the question the expert has been asked to answer.
That also means it may be entirely appropriate for a report to say that a conclusion cannot confidently be reached. A properly explained limitation is not a failure of expertise. It may be the most valuable assistance the expert can provide, particularly when the alternative is a conclusion that appears more certain than the evidence permits.
Putting the opinion in its proper scope
Experts should distinguish observations supported directly by the available material from interpretations that depend on assumptions. Where a diagnostic or technical framework has several distinct requirements, the report should address each of them and explain what evidence supports or fails to support it. If a crucial matter is not recorded, saying so may be more helpful than adopting an assumption merely to complete the analysis.
This does not mean experts must refuse to express an opinion whenever information is incomplete. It means explaining the limits of the available evidence, considering plausible alternative explanations and avoiding the suggestion that detailed objective records necessarily establish matters which those records could not have captured. In Ashton, the court preferred the expert who more clearly maintained that distinction. The disagreement was about the strength of the reasoning and the evidence, not an allegation that either psychiatrist lacked independence or integrity.
What experts can take from the judgment
The case is a reminder of the difference between knowing a great deal about a person’s recorded activities and knowing enough to answer a specialist question about their underlying condition or state of mind. It is not a reason to avoid making carefully reasoned inferences, but it is a reason to show the route from evidence to opinion and to be candid about what remains unknown.
Perhaps the hardest part of being an expert is sometimes recognising not what you know, but what the available evidence simply does not allow you to know.
Articles reflect the law and practice at the time of original publication, except where a dated update is shown. Subsequent developments may affect their relevance or accuracy. They are provided for general information and do not constitute legal or professional advice. Obtain appropriate advice on your particular circumstances. Full disclaimer.
