Trampoline parks generate strong opinions and weak data, usually in that order. Depending on who's talking, they're either an injury epidemic or no more dangerous than a football pitch. Both camps tend to argue without numbers, so this post collects what UK data actually exists, says plainly where it's thin, and draws the operational conclusions the figures genuinely support. If you run a park, this is your evidence base. If you're deciding whether to add trampolines to a venue, read it before the salesperson calls back.
What the UK numbers say
The most-cited national figure comes from 2017, when the BBC reported that ambulances were called to 1,181 incidents at trampoline parks across England in a single year, more than three a day, during the sector's fastest growth period (BBC News, 2018). The same reporting cycle carried a hospital comparison that remains the sharpest data point in the UK debate: of patients treated for trampoline injuries, 44% of those hurt at parks had fractures, against 36% for garden trampolines. Park injuries weren't just frequent. They were, on average, worse.
For scale across all trampolining (parks and gardens together), a study by RoSPA with the Royal College of Emergency Medicine and Oxford University Hospitals estimated around 13,000 trampolining injuries treated in English emergency departments each year, costing the NHS roughly £1.5 million annually. And RoSPA's long-standing headline finding about mechanism: around 60% of trampoline injuries happen when more than one person is on the trampoline (RoSPA trampoline safety). Hold onto that one; it matters more to operators than any other number here.
The most granular UK evidence is a published hospital study, Trampoline Park Injuries and Their Burden on Local Orthopaedic and Emergency Services, which tracked every patient arriving at one district general hospital from its local trampoline park over sixteen months. The numbers deserve a table:
| Finding | Figure |
|---|---|
| Patients from one park, Jul 2014 to Nov 2015 | 71 |
| Mean age (range) | 20 years (7 to 48) |
| Soft-tissue sprains | 41% |
| Fractures | 35%, mostly lower limb |
| Needed surgery | 20% |
| Admitted to hospital | 25%, mean stay 2 days |
| Severe cases | 2 open tibial fractures, transferred to major trauma centres |
| Care costs from one park | Over £80,000 |
Read that middle column twice. A fifth of the casualties from one park needed an operation. And the mean age of 20, with a range running to 48, is the quiet surprise: this is not purely a children's injury problem. It's parents' tibias too, which any park operator who has watched the adults-only hour could have told the researchers.
Where the data is honestly weak
Now the part most articles skip. The UK has no central register of trampoline park injuries. What exists is ambulance callout counts obtained by journalists, single-hospital studies like the one above, and RoSPA estimates built from emergency department sampling. That means three real limitations. Nobody can state a national park-injury total with confidence. Injury rates (per visit, per jump-hour) mostly come from the industry's own attendance figures, which outsiders can't verify. And the park-versus-garden comparison is muddied by exposure: parks concentrate many jumpers into supervised hours, gardens spread fewer jumpers across unsupervised years, and the studies can't fully control for that.
So the defensible summary is narrower than either camp wants: park injuries are frequent enough to be a serious operational concern, they skew more severe than garden trampoline injuries when they happen, and the true denominators are unknown. Anyone quoting you a precise national injury rate is quoting a guess.
What changed after 2017
The 2017 figures were the pressure that produced the framework the industry now runs on. RoSPA and operators developed a minimum safety standard, which matured into PAS 5000:2017, the BSI specification for trampoline park construction and operation, alongside IATP membership requirements. We covered the full framework in our trampoline park standards guide. The honest caveat is that no equivalent national dataset has been published since to prove the standards bent the injury curve; what operators have instead is their own incident data, which brings us to the practical half of this post.
What the numbers mean for how you run a park
Enforce one jumper per trampoline like it's the whole job, because statistically it nearly is. If around 60% of injuries involve multiple users, single-bounce enforcement is the highest-value safety intervention available to you, worth more than any padding upgrade. It's also the rule that dies quietly on busy days when courts are full and monitors are tired: the exact drift our peak season guide is about.
Design and staff for lower-limb fractures. The injury pattern is landings: ankles and tibias, concentrated where jumpers meet surfaces or each other. That tells you where monitors stand, what your briefing emphasises (no double-bouncing, land on two feet, walk don't run between courts), and why your first aiders need to be genuinely comfortable immobilising a leg. Our first aid guide covers sizing that provision.
Take the adults seriously. A mean casualty age of 20 says the risk conversation can't be pitched at children only. Adult sessions, parents jumping “just for a minute”, and works outings are where the heavier, faster bodies meet the same springs.
Build your own dataset, because the national one doesn't exist. The absence of central data cuts both ways: nobody can benchmark you, and nothing will warn you except your own records. A park logging every incident and near-miss with location, time, session type and mechanism owns something the entire UK research community lacks: a complete injury picture for one venue. Patterns in that log (one court, one session type, one time of day) are your early-warning system, and the discipline connects straight into the RIDDOR duties you already carry for the serious cases.
Reading injury headlines like an operator
When the next “trampoline injury surge” story lands (and it will, usually in August), three questions sort the signal from the noise. Does the figure separate parks from garden trampolines, or lump them together? The 13,000-a-year estimate is all trampolining; quoting it as a parks number inflates the sector's count several-fold. Is there a denominator? A raw injury count without visit numbers can't tell you whether risk rose or attendance did. And is the comparison severity or frequency? Parks look worse on severity (the fracture comparison), while gardens rack up frequency across millions of unsupervised back-garden hours. A journalist has two hours to file. You have the incident log of an actual venue. Read accordingly, and when a local paper calls you for comment, these three questions are also your answer.
Where SafePlay fits
SafePlay is where that private dataset lives without anyone maintaining a spreadsheet: incidents and near-misses logged from a phone in the moment, court inspections and padding checks timestamped in Daily Checks, monitor training current in Staff Compliance, and the whole picture reviewable when you actually have time to think. The national data may be patchy. Yours doesn't have to be.
Sources and data vintage: ambulance callout and fracture-comparison figures as reported by BBC News (2018) from 2017 data; annual injury and cost estimates from RoSPA with RCEM and Oxford University Hospitals; single-park clinical figures from the peer-reviewed study linked above (data 2014 to 2015). No newer national dataset had been published at the time of writing.