Most people outside the law assume a claim succeeds when the evidence proves something happened. It does not need to. A civil claim succeeds when it is more likely than not to have happened, which is a lower and quite different bar, and understanding exactly how low is essential to reading, writing or challenging a medico-legal opinion.
Prof. Sandip Hindocha, a Consultant Plastic Surgeon and the UK’s National Lead Forensic Medical Examiner, prepares expert witness reports for claimant and defendant solicitors and as a single joint expert, and every opinion in those reports is pitched against this one standard.
In short: balance of probabilities means more likely than not, over 50 percent, not certainty and not a strong suspicion. It is the standard that applies to every disputed fact in a civil claim, breach of duty, causation, and the claimant’s own account of events, and it does not shift for a serious allegation. The court has to end up finding that something did or did not happen, there is no finding that it might have. An expert’s job is to say plainly which side of that line the evidence falls on, and to be honest when it genuinely does not fall clearly on either side.
What Balance of Probabilities Actually Means
The classic formulation comes from Lord Denning in Miller v Minister of Pensions [1947] 2 All ER 372, who described the standard as requiring the tribunal to be satisfied the thing is more probable than not. If the evidence tips the scale, however slightly, the party carrying the burden wins. If the scale sits level, that party loses, since the burden itself carries no weight.
That is the whole test. There is no requirement to eliminate other explanations, no requirement for a specific percentage above 50, and no separate, higher standard hiding inside serious cases. A finding of 51 percent likely and a finding of 95 percent likely both simply mean the fact is proved.
How the Civil Standard Differs From the Criminal One
Criminal cases require the jury to be sure, often expressed as beyond reasonable doubt, a standard close to certainty. Civil claims, including clinical negligence and personal injury, never reach for that bar. A clinician can be acquitted of a criminal charge over the same facts a civil court later finds against them on the balance of probabilities, and both outcomes can be correct, because the two courts are answering different questions to different standards.
| Standard | What it requires | Where it applies |
|---|---|---|
| Balance of probabilities | More likely than not, over 50 percent | Civil claims, including clinical negligence and personal injury |
| Beyond reasonable doubt | The fact-finder must be sure, close to certainty | Criminal prosecutions |
Why the Standard Does Not Rise for a Serious Allegation
It is a common and reasonable-sounding misconception that a serious allegation, a birth injury caused by negligence, or a suggestion that a clinician acted dishonestly, needs to clear a higher bar than an ordinary one. It does not. The House of Lords settled this in Re B (Children), where Lord Hoffmann held that the seriousness of an allegation makes no difference to the standard of proof to be applied, though it may affect how much evidence is needed in practice to satisfy that standard, since people do not usually act in a way that is seriously discreditable.
What changes with seriousness is the evidence a sensible fact-finder expects to see, not the arithmetic of the test itself. A vague, unsupported allegation of a catastrophic failure is unlikely to satisfy anyone that it is more probable than not, but that is a comment on the quality of the evidence, not a hidden second standard.
The Binary Nature of the Standard
The same judgment contains the passage that matters most for how an expert should write. Lord Hoffmann described the law as operating a binary system in which the only values are 0 and 1, the fact either happened or it did not. There is no room for a court to record that something probably happened, or might have happened, as a standalone finding sitting between the two.
Once a fact clears 50 percent, the court treats it as having happened, full stop, and reasons forward from there as a certainty for every purpose that follows. Once it falls short, the court treats it as not having happened, even if the true likelihood was 49 percent. This is why the standard is sometimes described as all or nothing, a fact that only just crosses the line is treated identically to one that plainly does.
Where the Civil Standard Applies in a Medicolegal Case
The same standard runs through every disputed factual question in a claim, not just the headline issue of whether the defendant was negligent.
Breach of Duty
Whether the care given fell below what a responsible body of similarly qualified practitioners would have provided is a factual question decided on the balance of probabilities, informed by expert opinion but ultimately a matter for the court.
Causation
Whether the breach actually caused the harm complained of is judged to the same standard, and it is usually the harder question of the two. Causation in complex cases, including where more than one factor may have contributed, is a substantial topic in its own right and covered separately, but the underlying standard of proof is the same 50 percent test discussed here.
Quantum and Prognosis
What the claimant’s future holds, whether symptoms will resolve, whether further surgery will be needed, is also assessed on the balance of probabilities where the question is binary, though where the future is genuinely uncertain the court can instead value a percentage chance, which is a different and narrower exception discussed below.
The Claimant’s Own Account
Whether the claimant is telling the truth about symptoms, pain, or the circumstances of an accident is weighed on the same standard, tested against contemporaneous records, since a report built on an account that does not survive scrutiny cannot itself survive scrutiny.
How Experts Should Express a Probabilistic Opinion
Practice Direction 35 requires an expert’s report to state the degree of certainty behind each opinion. That single requirement is where most of the practical difficulty in writing a good report actually sits.
Medicine rarely deals in certainty, and an expert who is asked whether a delay caused a worse outcome is not being asked whether it is scientifically proven beyond doubt. They are being asked whether it is more likely than not. Saying an outcome was merely possible, when the honest opinion is that it was probable, undersells a case the evidence actually supports. Saying an outcome was probable when the honest opinion is that it was merely possible does the opposite, and exposes both the expert and the case to a difficult cross-examination.
A report that genuinely cannot resolve a question one way or the other should say so in those terms, rather than reaching for a form of words that sounds more confident than the underlying evidence justifies.
The All or Nothing Effect in Practice
The binary nature of the standard produces results that can feel harsh. In Gregg v Scott [2005] UKHL 2, a delay in diagnosing cancer was found to have reduced the claimant’s chance of a cure from around 42 percent to 25 percent. Because his chance of a good outcome was already below 50 percent before the negligence, he could not show that the delay had made the difference between recovery and non-recovery on the balance of probabilities, and the claim for that head of loss failed.
The claimant had genuinely lost something real, a meaningful reduction in his chance of survival, but English law does not generally compensate a lost chance below 50 percent as a distinct type of loss in personal injury and clinical negligence claims. The case is a stark illustration of why the 50 percent line matters so much in practice, not just in theory.
Where the Standard Bends for Material Contribution
The strict binary approach has one well established softening. Where more than one cause has contributed to an indivisible injury, and it cannot be shown on the ordinary balance of probabilities which single cause was decisive, a defendant can still be liable if their breach made a material contribution, meaning more than negligible, to the harm, following the long-standing principle in Bonnington Castings Ltd v Wardlaw [1956] AC 613 and applied in the clinical setting since.
This does not lower the standard of proof itself, the material contribution still has to be proved to the same balance of probabilities standard, it changes what has to be proved. It is a narrow exception rather than a general escape route, and the full detail of how causation is argued in a multi-cause case is a substantial topic beyond what fits here.
Common Mistakes With the Standard
A few misunderstandings recur often enough in medical negligence claims to be worth naming directly. Treating a serious allegation as needing stronger proof than the standard actually requires is one, covered above. Assuming an expert opinion must be certain before it is useful is another, when what the court actually needs is an honest probability.
Confusing possible with probable is the most consequential, since the two words describe entirely different positions under the standard, and using them loosely in a report or in correspondence with instructing solicitors can misrepresent the strength of a case in either direction.
What This Means for a Medicolegal Expert Report
A report that engages properly with the standard states, for each disputed question, whether the expert’s opinion is that the answer is more likely than not, and does not hide behind hedged language where a clear view is actually held. Where the evidence is genuinely balanced, the report says that too, since an honest admission of uncertainty is more useful to the court, and more defensible under cross-examination, than false confidence in either direction.
How that opinion is structured within the report, and what else Practice Direction 35 requires alongside it, is covered in more detail in how to write a high-quality expert witness report, and it sits within the wider framework of an expert’s duty to the court described in how expert witnesses are regulated in the UK.
Frequently Asked Questions
What percentage is balance of probabilities?
There is no fixed percentage in the rules, but it is generally understood as anything over 50 percent. A fact found to be 51 percent likely and one found to be 95 percent likely are both treated as proved.
Does a serious allegation need a higher standard of proof?
No. The House of Lords confirmed in Re B (Children) that the standard does not change with the seriousness of the allegation, though more serious or inherently unlikely allegations often need stronger evidence in practice to reach that same 50 percent threshold.
What happens if the evidence is exactly balanced?
The party who carries the burden of proof loses. The burden itself has no evidential weight, so an evenly balanced case is not decided in favour of whoever is arguing it, it goes against them.
Can a claimant recover damages for a lost chance below 50 percent?
Generally not, in personal injury and clinical negligence claims. Gregg v Scott confirmed that a reduction in the chance of a better medical outcome is not itself compensated as a distinct loss where the original chance was already below 50 percent.
Is balance of probabilities the same standard for every issue in a claim?
Yes. Breach of duty, causation, quantum and the credibility of the claimant’s account are all decided on the same balance of probabilities standard, there is no sliding scale between different issues within a single civil claim.
Prof. Hindocha prepares CPR Part 35 compliant expert reports that state the degree of certainty behind every opinion, for solicitors and insurers across England and Wales. Details of his expert witness services and how to instruct him are on the instruction page.
This article is general information about the standard of proof used in civil and medicolegal cases in England and Wales, and is not a substitute for legal advice on a specific claim.