Recognising and managing a concussion is now well codified, and the companion page on concussion covers assessment, the graduated return and persisting symptoms. Two questions sit outside that pathway and are far less settled: what actually prevents concussion, and what repeated head impact does to the brain over decades. Both are areas where the confident claim is usually ahead of the evidence, and where a clinician is likely to be asked for an opinion by an anxious parent, a governing body or a journalist. This page covers what prevention strategies are supported, the distinction between concussion and sub-concussive head acceleration load, what the long-term studies do and do not show, and how to hold that conversation honestly.
The strategies with the best evidence are the ones that reduce exposure to head impact rather than those that try to protect the head once impact occurs.
Rule and exposure modification generally has the strongest and most consistent evidence. Disallowing body checking in youth ice hockey substantially reduces concussion rates. In rugby union, lowering the legal tackle height has been introduced in community and age grade rugby, and video analysis has found a reduction in head to head and head to shoulder contact following the change. That is a contact mechanism outcome rather than a concussion incidence outcome, so it is supportive rather than proof, and implementation depends on consistent refereeing decisions that are themselves difficult. In football, heading restrictions in youth training were introduced before 2024, and deliberate heading in matches is being phased out across under-seven to under-eleven football in England over the seasons from 2024-25 to 2026-27.
Neuromuscular and technique interventions sit in the middle, with emerging and context-specific evidence, including a reduction in concussion with neuromuscular training in adolescent rugby. Coaching safer tackle technique is plausible and widely recommended. Neck strengthening is the intervention most often promoted, on the reasoning that a stronger head and neck segment reduces head acceleration for a given impulse. The evidence for it is genuinely mixed: a systematic review found only three eligible studies, and reviewers have concluded that current evidence is insufficient to establish that cervical strengthening reduces concussion incidence, even where it successfully increases neck strength. It is reasonable to include and dishonest to promise.
Equipment evidence is mixed and, importantly, sport, product, age and sex specific rather than uniformly negative. Helmets clearly reduce skull fracture and superficial head injury. Their effect on concussion differs by sport and product: soft headgear has not shown consistent protection in rugby, while some low-certainty studies in adolescent female field sports suggest possible benefit, so a single universal answer is not supportable. Mouthguards prevent dental and orofacial injury, and the Amsterdam consensus identified concussion prevention evidence for mouthguard use in ice hockey specifically, while the picture across other sports remains mixed. Emerging small-scale work suggests a custom mouthguard may reduce peak linear head acceleration during heading, but a crossover study in eighteen amateur players is a signal rather than a basis for advice. Concussion is a multiaxial biomechanical event, so equipment limitations cannot be reduced to an inability to reduce rotation alone. Education and recognition sit alongside all of this, improving reporting and management rather than preventing injury.
A concussion is a clinically apparent event. A head impact without diagnosed concussion produces no clinical signs, and such impacts are considerably more numerous than diagnosed concussions in collision sport, though the ratio is sport and exposure specific. The older term sub-concussive is still widely used but should not be taken to imply the impact is biologically harmless. Head acceleration load is the cumulative exposure to both, and the shift over the past decade has been from counting diagnosed concussions to attempting to quantify total exposure.
The technology driving this is the instrumented mouthguard, which measures linear and rotational acceleration of the head during play and has been deployed in professional rugby to identify impacts exceeding a threshold and trigger off-field assessment. The biomechanical rationale is that injury risk relates to head acceleration, particularly rotational acceleration, rather than to the force of the impact as such.
Two cautions belong with this. First, a measured acceleration threshold is not a diagnosis: it identifies an event warranting assessment, and clinical judgement still determines whether concussion occurred. Second, head acceleration measures cannot diagnose concussion, define tissue damage, identify a universally safe threshold or quantify an individual's long-term risk on their own. The relationship between cumulative load and long-term outcome is plausible and incompletely established, so the honest position is that reducing unnecessary head impact exposure is sensible on precautionary grounds rather than because a safe threshold has been defined. No such threshold exists.
The most influential UK evidence comes from a retrospective cohort of former Scottish professional footballers, which found higher mortality from neurodegenerative disease and greater use of dementia medication than in matched controls.
The finding that carries most weight is the position gradient. Risk estimates were highest among defenders, who head the ball frequently, with no statistically clear elevation among goalkeepers, who have the same career in the same environment but rarely head it. That within-sport gradient supports an exposure-related interpretation, though it does not eliminate confounding or establish that heading alone caused the difference. Comparable concerns have been raised in rugby and in Australian rules football, using different study designs and with their own uncertainties.
Chronic traumatic encephalopathy (CTE) is a neuropathological diagnosis made after death, characterised by a distinctive pattern of tau deposition. Several points are routinely misstated. It cannot currently be diagnosed in life. Much of the case series evidence comes from brain banks to which families donated because symptoms were present, which is a powerful selection bias and means prevalence cannot be inferred from those series. The relationship between the pathology and the clinical syndrome is not fully established, and not everyone with the pathology had symptoms. Research criteria exist for a clinical entity termed traumatic encephalopathy syndrome, but meeting them does not diagnose chronic traumatic encephalopathy in a living person.
The defensible summary is that an association between repeated head impact exposure and later neurodegenerative disease is now supported by several independent lines of evidence, that the position gradient strengthens the causal case considerably, and that the evidence remains largely observational with dose, threshold and individual susceptibility all unresolved.
Parents, athletes and administrators ask a question the evidence cannot fully answer, and the temptation is to resolve it in one direction or the other. Neither reassurance that there is no evidence of harm nor an assertion that contact sport causes dementia is honest.
A defensible position has several parts. The health benefits of participation in sport are large and well established, and should not be omitted from a conversation about risk. The association between repeated head impact and later neurodegenerative disease is real enough to justify action, and action is already being taken through rule change. Exposure can be reduced without abandoning the sport, principally by reducing unnecessary repetitive contact in training while retaining the exposure genuinely required for safe skill development and competition preparation. Recognising each concussion and removing the athlete promptly, with no same-day return, remains among the clearest controllable elements, though it is not the sole determinant of long-term outcome.
For an individual athlete with a history of multiple concussions, the assessment is individualised rather than a fixed count. Relevant factors are the neurological examination, the number of events, the force required to produce each one and whether that is decreasing, the symptom trajectory and whether recovery is lengthening, neuropsychological and vestibular findings where indicated, mental health, the sport and position, cumulative exposure, the athlete's own preferences, and the genuine uncertainty about future risk. There is no evidence-based number of concussions at which retirement becomes mandatory, and any clinician quoting one is expressing a policy rather than a finding.
Sign up to get full access to 10 topics of your choice, including all sections, clinical pearls, and exam tips.
Sign up free10 free topics included with your account. Full access from £24.17/month.
Sections included with full access