A stone in the road becomes destabilizing when your balance is compromised.
In Sweden, a study spanning nearly a decade of fatal micromobility crashes has surfaced a troubling pattern: nearly half of those who died on electric scooters were profoundly intoxicated, with blood alcohol levels nine times the legal limit, and not one was wearing a helmet. The findings place e-scooter fatalities in a distinct category from cycling deaths—younger riders, nighttime hours, solo crashes, and a culture of underestimated risk. As cities have focused their regulatory gaze on rental fleets, the data quietly points elsewhere: most of the dead were riding their own machines, beyond the reach of any operator's safeguards.
- Nearly half of all fatal e-scooter riders in Sweden were drunk at the time of their crash—at blood alcohol levels so extreme they would qualify as aggravated drunk driving under Swedish law.
- Not a single person killed on an e-scooter was wearing a helmet, even as head injuries emerged as the leading cause of death across every vehicle category studied.
- The crash profile is distinct and alarming: younger riders, alone, at night, on weekends, losing control of a vehicle many assume is too slow to kill.
- Regulations targeting rental scooters—speed caps, nighttime locks—are largely missing the mark, since nearly nine in ten alcohol-related deaths involved privately owned vehicles.
- Researchers are developing sensor-based systems to detect impaired riding in real time, but acknowledge that technology cannot substitute for a fundamental shift in how riders perceive the danger they are in.
Between 2016 and 2024, Swedish researchers analyzed 204 fatal crashes across three categories of light electric and human-powered vehicles. The e-scooter data stood apart. Of the 18 people who died riding e-scooters, eight—44 percent—were intoxicated at the time of impact. The median blood alcohol level among those riders was 1.8 per mille, nine times Sweden's legal driving limit and deep into the range the country classifies as aggravated drunk driving. By comparison, alcohol was involved in 27 percent of fatal e-bike crashes and just 13 percent of fatal bicycle crashes.
The crashes followed a recognizable pattern: weekends, evenings and nights, single-vehicle incidents rather than collisions with cars. The median age of fatally injured e-scooter riders was 47.5 years. And across all 18 deaths, not one rider was wearing a helmet—compared to roughly 25 percent helmet use among fatally injured cyclists and e-bike riders. Head injuries were the leading cause of death in every group.
Researchers from Chalmers University of Technology noted that many riders fatally underestimate the danger of riding while impaired, assuming the vehicle's modest speed makes it forgiving. It does not. Alcohol degrades balance and reaction time; a small pothole or stone in the road becomes a fatal obstacle when reflexes are compromised and there is nothing protecting the skull.
The policy implications are pointed. Public debate and regulation have concentrated on rental scooters, with operators imposing speed limits and nighttime restrictions. But nearly nine in ten alcohol-related e-scooter deaths in the study involved privately owned vehicles—beyond the reach of any rental platform's controls. Senior researcher Marco Dozza is leading work on sensor systems that could detect impaired riding in real time, but he and his colleagues are candid: technology and regulation can only go so far. The deeper problem is cultural—how riders understand, or fail to understand, what these machines demand of them.
Between 2016 and 2024, Swedish researchers examined 204 fatal crashes involving electric scooters, e-bikes, and conventional bicycles. What they found in the e-scooter data was stark: nearly half of all riders killed were intoxicated at the time of impact. The numbers tell a story that goes beyond simple correlation—they suggest a particular vulnerability in how people use these machines when their judgment is compromised.
Of the 18 people who died in e-scooter crashes during this period, eight were under the influence of alcohol. More alarming than the proportion was the severity. The median blood alcohol content among intoxicated e-scooter riders was 1.8 per mille—nine times Sweden's legal driving limit of 0.2 per mille, and well into the territory of what the country classifies as aggravated drunk driving. For comparison, fatal crashes among e-bike riders showed alcohol involvement in 27 percent of cases, and among traditional cyclists, just 13 percent. The gap is not marginal. It suggests something specific about the intersection of impairment and electric scooter use.
Marco Dozza, a professor of active safety and road user behavior at Chalmers University of Technology and senior researcher on the study, framed the problem plainly: intoxication is common among all road users, but it appears to hit e-scooter riders with particular force. Not only were many of the fatally injured riders drunk—their level of intoxication was extremely high. The crashes themselves followed a pattern. They happened mostly on weekends, in evenings and nights, and they were typically single-vehicle incidents rather than collisions with cars. The median age of the riders was 47.5 years old.
Another detail unified all three groups: the absence of head protection. Not a single person killed while riding an e-scooter was wearing a helmet. Among e-bike riders and cyclists, about 25 percent wore one. Head injuries were the leading cause of death across all three categories. Rahul Rajendra Pai, a doctoral student at Chalmers and the study's first author, noted that the numbers make the intervention obvious. A helmet is not foolproof, but it dramatically improves survival odds. Yet almost no one killed on an e-scooter had one on.
Dozza explained the mechanics of the risk. Many people underestimate how dangerous it is to ride an e-scooter while intoxicated, assuming the vehicle is too slow to cause serious harm. But alcohol slows cognition and reaction time. A stone in the road, a pothole—small obstacles become destabilizing when your balance and reflexes are compromised. Without a helmet, a fall that might have been survivable becomes fatal.
The study revealed different crash profiles for each vehicle type. Bicycle fatalities typically involved elderly riders—median age 71—hit by motor vehicles on weekdays in daylight. E-scooter deaths painted a different picture: younger riders, nighttime crashes, single-vehicle incidents, and alcohol as a dominant factor. This distinction matters for policy. A crash pattern that looks like an elderly person being struck by a car requires different countermeasures than a younger rider losing control alone at night while intoxicated.
One finding has particular policy implications: almost nine out of ten alcohol-related e-scooter deaths involved privately owned vehicles, not rentals. Public attention and regulation have focused heavily on rental scooters, with operators implementing speed limits and night-time restrictions. These measures do nothing to address the private scooter market, where most of the fatal crashes occur. Rikard Fredriksson, a senior advisor in vehicle safety at the Swedish Transport Administration and co-author of the study, emphasized that alcohol remains a major problem across Swedish roads—involved in about 20 percent of all fatal crashes. For e-scooters, the figure is more than double that.
Researchers are exploring technological solutions. Dozza is leading a study on how sensor technology in rental scooters might detect impaired riding in real time and intervene before a crash occurs. But he and his colleagues acknowledge that technology and regulation alone cannot solve the problem. The real challenge lies in social norms and rider behavior—the culture around how people use these machines. Training matters. Understanding what a vehicle can and cannot do matters. A helmet always matters. The study suggests that without shifts in how people think about riding while intoxicated, the deaths will continue.
Citações Notáveis
Alcohol intoxication is a common issue among all road users, but it appears to be particularly severe among riders of electric scooters. Not only were many fatally injured e-scooter riders intoxicated, but their levels of intoxication were also extremely high.— Marco Dozza, Professor of Active Safety and Road User Behavior at Chalmers University of Technology
The typical fatal bicycle crash, with an elderly cyclist being hit by a motor vehicle in daylight, may require completely different countermeasures than the typical fatal electric scooter crash, where a younger rider crashes alone at night while intoxicated.— Marco Dozza
A Conversa do Hearth Outra perspectiva sobre a história
Why does alcohol show up so much more in e-scooter deaths than in bicycle deaths? Is it just that e-scooter riders are younger and go out at night more?
That's part of it, yes—the timing and age profile are different. But the researchers found something more specific: the blood alcohol levels among intoxicated e-scooter riders were extremely high, nine times the legal driving limit. It's not just that more of them were drunk. They were much more drunk.
So why would someone ride an e-scooter while that intoxicated? What's the appeal compared to, say, calling a taxi?
The researchers suggest people don't fully grasp the danger. They think the scooter is slow enough that it doesn't matter. But alcohol slows your reaction time and balance. A pothole or a stone becomes a hazard you can't recover from. And if you're not wearing a helmet—which none of the people killed were—a fall becomes fatal.
That's striking. Zero helmets among the dead e-scooter riders. Why such a difference from cyclists, where 25 percent wore them?
The researchers didn't speculate on the why, but the pattern is clear. E-scooter use is newer, less regulated, and perhaps people don't think of it as requiring the same safety gear. It feels casual. But the consequences are identical—head injuries are the leading cause of death in all three groups.
The study mentions that most of these deaths involved privately owned scooters, not rentals. That seems like a policy gap.
Exactly. Rental companies have added speed limits and night restrictions. But those rules don't touch private scooters, which is where most of the fatal crashes happen. The regulations are targeting the wrong pool of vehicles.
Can technology fix this? The researchers mention sensors that might detect impairment.
They're exploring it, and it's promising. But the researchers are clear that sensors and rules won't solve it alone. The real problem is cultural—how people think about riding while drunk, whether they wear helmets, whether they understand what the vehicle can do. That's harder to change with technology.