Do Bike Helmets Work? Not 85%, but Still Yes

The Short Version: Yes, bike helmets work. The often-cited claim that helmets reduce head injuries by 85% has been discredited, but modern meta-analyses consistently show they cut head-injury risk by about half and serious head-injury risk by 60–70%.
Those figures come from multiple meta-analyses pooling 40 to 55 studies of real-world crash data. The benefit is greatest for severe injuries like skull fractures and intracranial bleeding. In 2013, NHTSA and the CDC officially stopped citing the inflated 85% figure after it was challenged under the Data Quality Act.

My neighbor told me the other day that bike helmet usefulness is “exaggerated”, that the research is “mixed.” I’ve heard this one before. It’s the kind of argument that sounds solid if you’ve skimmed a headline or two, and it hangs around because nobody likes being told what to put on their head.

The thing is, the research isn’t mixed. Not when you actually read it. So I pulled up the studies, the real numbers, and the track record of what happens when a cyclist’s head meets asphalt with and without a layer of EPS foam and a polycarbonate shell between the two.

Let’s take a look at what the data actually says.

What Does the Research Say About Bicycle Helmet Effectiveness?

The most thorough look at whether bicycle helmets work came from Jake Olivier and Prudence Creighton, who published a systematic review and meta-analysis in the International Journal of Epidemiology in 2017. They pulled together 40 studies covering more than 64,000 injured cyclists and combined the results using statistical models that accounted for differences between the studies. The results were pretty clear.

Helmet use was associated with a 51% reduction in head injury risk, a 69% reduction in serious head injury, a 33% reduction in facial injuries, and a 65% reduction in fatal head injury. For neck injuries, the data showed no statistically significant effect in either direction, which puts to rest the old claim that helmets somehow make neck injuries worse.

A separate meta-analysis by Alena Høye, published in Accident Analysis & Prevention in 2018, looked at 179 effect estimates drawn from 55 studies spanning 1989 to 2017. Her findings were broadly consistent: a 48% reduction in head injury, 60% in serious head injury, and 53% in traumatic brain injury (TBI). She also found no indication that the results were skewed by publication bias or by a failure to control for confounding variables.

Then in 2023, a systematic review published in Scientific Reports (Nature) pulled all of this together and stated it plainly: “Wearing a helmet while cycling is beneficial, regardless of age, crash severity, or crash type.” That’s about as close to a settled question as you’ll find in injury prevention research.

Where Did the 85% Number Come From?

If you’ve ever looked into bike helmet statistics, you’ve probably come across the claim that helmets reduce head injury risk by 85% and brain injury by 88%. Those numbers were everywhere for decades, cited by government agencies, safety organizations, and well-meaning parents alike. But the story behind them is more complicated than most people realize.

The figures originated from a 1989 case-control study by Thompson, Rivara, and Thompson, published in the New England Journal of Medicine. The study collected data from cyclists who showed up at Seattle-area hospitals after crashes. Only 7% of those with head injuries were wearing helmets, compared to 24% of those without head injuries. From that comparison, they estimated the 85% protective effect.

The problem was that later efforts to replicate those results consistently found weaker effects. A case-control study relies on comparing cases (injured cyclists) to controls (other cyclists), and the methodology is better at showing whether something has an effect than at precisely quantifying that effect. The study’s own authors revised their estimate down to 69% in a 1996 follow-up.

By 2013, the situation had become untenable. The Washington Area Bicyclist Association (WABA) formally challenged federal agencies under the Data Quality Act, asking them to stop citing the 85% figure. Both the National Highway Traffic Safety Administration (NHTSA) and the Centers for Disease Control and Prevention (CDC) agreed to remove the estimate from their websites.

The point being, helmets absolutely do work, but the real-world protection is closer to 50% for head injuries and 65% for fatal head injuries (per the Olivier and Creighton meta-analysis), not the inflated 85% that became a kind of safety mantra. Ironically, overclaiming the benefit may have done more harm than good, because it gave skeptics an easy target.

Can a Bike Helmet Prevent a Concussion?

This is where we need to be upfront, because the honest answer isn’t what most people expect. The Consumer Product Safety Commission (CPSC) has stated explicitly that no bicycle helmet has been proven to prevent concussions. That’s not a criticism of helmets. It’s a reflection of what current bicycle helmet safety standards were designed to do.

Standards like CPSC 16 CFR 1203 and EN 1078 test for linear (straight-down) impacts. They measure whether the helmet keeps peak acceleration below a g-force threshold (300g for CPSC, 250g for EN 1078) that would cause skull fractures and severe traumatic brain injury. They do that job well. But concussions are strongly associated with rotational and linear acceleration, especially in oblique impacts, even at relatively low speeds.

Technologies like MIPS (Multi-directional Impact Protection System) are designed to address this gap. A low-friction slip plane inside the helmet allows 10 to 15 millimeters of movement during an angled impact, redirecting some of that rotational energy away from the brain. Lab testing published in the Annals of Biomedical Engineering showed MIPS-equipped helmets produced 16.8% lower peak rotational acceleration and 49.3% lower brain injury risk metrics at low-speed impacts compared to conventional helmets.

There are caveats to this, though. A 2020 study in Traffic Injury Prevention found that when a biofidelic scalp layer (one that behaves more like real human skin and hair) was added to the test headform, the performance difference between MIPS and non-MIPS helmets largely disappeared. The science is still evolving, and if you want to go deeper on this question, we’ve written about concussion prevention specifically.

Do Helmets Make Cyclists Take More Risks?

This is the risk compensation argument, and it goes something like this: if you put a helmet on a cyclist, they feel safer, so they ride faster, take more chances, and the helmet’s protective benefit gets cancelled out by riskier behavior. It sounds logical. It even makes intuitive sense. But does the evidence back it up?

Not really. In 2019, Esmaeilikia and colleagues published the first systematic review specifically examining bicycle helmet use and risk compensation, in Transportation Research Part F. They searched five research databases and identified 23 studies that met their inclusion criteria.

Of those 23 studies, 18 found no evidence that helmet use was associated with riskier cycling behavior. Three produced mixed results, and only two (both from the same research lab) supported the risk compensation hypothesis. In fact, many of the studies found the opposite: helmet wearers tended to cycle more cautiously than non-wearers.

The risk compensation argument also tends to come up in countries like the Netherlands and Denmark, where cycling rates are very high and helmet wearing is very low. The reasoning goes that these countries are proof you don’t need helmets, but the reality is more nuanced. Both countries have invested heavily in separated cycling infrastructure, lower urban speed limits, and a cycling culture where motorists expect and look for cyclists.

Their low injury rates reflect infrastructure and driver behavior, not evidence that helmets are unnecessary. The question of whether to mandate helmets is genuinely contested (we’ll get to that shortly), but it’s a separate question from whether helmets themselves work.

How Many Cyclists Get Hurt Without a Helmet?

Let’s look at the numbers, because this is where the abstract question of “do bike helmets work” becomes very concrete. In the US alone, emergency departments treat an estimated 80,000 cycling-related head injuries per year, according to the CPSC. Head injuries account for roughly one-third of all cycling emergency department visits, two-thirds of cycling-related hospital admissions, and three-quarters of all cycling deaths.

In 2023, NHTSA recorded 1,166 cyclist fatalities on US roads, the highest number ever recorded. That’s an 87% increase from the low point of 623 deaths in 2010. The League of American Bicyclists noted that each of those deaths costs society approximately $10.5 million in medical expenses, lost wages, and lost quality of life.

And here’s the stat that should make you stop and think. According to IIHS analysis of 2024 FARS data, 68% of bicyclists killed in crashes were not wearing a helmet. Only 13% were confirmed to be wearing one (the remaining cases were unknown). The vast majority of fatalities involved riders age 20 and older, male cyclists accounted for nearly 90% of deaths, and most occurred in urban areas.

Helmets are more effective in single bicycle crashes (falls, hitting a curb, losing balance) than in collisions with motor vehicles, simply because a motor vehicle collision involves forces that can exceed what any bicycle helmet is designed to handle. But even in motor vehicle collisions, helmets significantly reduce the odds of a fatal head injury.

What Do Mandatory Helmet Laws Tell Us?

This is where the debate gets genuinely complicated, and where common sense has to step in alongside the data. Australia introduced mandatory all-age helmet laws in the early 1990s and saw significant reductions in cycling head injuries. Similar laws in parts of Canada produced comparable results. In the US, 21 states and Washington D.C. have mandatory helmet laws, but all of them apply only to children.

On the other side of the argument, countries like the Netherlands and Denmark have deliberately chosen not to mandate helmets, partly because of the “safety in numbers” effect: more cyclists on the road means drivers become more accustomed to watching for them, which can reduce collision rates overall. The concern is that helmet laws discourage some people from cycling, which shrinks the cycling population, which paradoxically might increase per-cyclist risk.

The question remains, though, whether any of that changes the fundamental physics of what happens when your head hits the ground. Regardless of what you think about helmet legislation, the effectiveness data doesn’t change based on which country’s law you’re looking at.

A helmet absorbs impact energy through its EPS (Expanded Polystyrene) foam liner the same way in Amsterdam as it does in Austin. Whether a government should force you to wear one is a policy debate. Whether wearing one protects your head is a physics question, and that question has been answered pretty convincingly.

So, Do Bicycle Helmets Work?

Yes. The evidence across decades of research, across multiple countries, and across every credible meta-analysis conducted in the last thirty years says the same thing: wearing a bicycle helmet substantially reduces your risk of head injury, serious head injury, and fatal head injury. Not by 85% (that number was overblown), but by roughly 50% for head injuries and 65% for fatal head injuries. Those are still very good odds for something that costs as little as $20 and weighs less than a pound.

There are things helmets won’t do. They won’t prevent concussions (at least not with current mandatory testing standards, though technologies like MIPS and testing programs like the Snell Memorial Foundation’s voluntary certification are pushing in that direction). They won’t protect you in every conceivable crash scenario. And they absolutely won’t work properly if they don’t fit your head correctly.

So, my neighbor’s statement about helmet usefulness being exaggerated was an exaggeration in itself. Which brings us back to the main question of whether helmets actually work. And the answer, backed by more than 64,000 cyclist injury records and the combined weight of every major research institution that has looked at this question, is yes. They do. Common sense says wear one.