Rented e-bicycles more dangerous than e-scooters in cities

Chalmers University of Technology

Facebook

XLinkedIn

Email

 

image: 

For those who want to rent the safest vehicle, thus far the e-bicycle has seemed to be the best choice. But a recently published study from Chalmers University of Technology, which compared these alternatives in a more equitable way than previous research, has shown that this is not true – on the contrary, the e-scooter is safer according to the study’s results.

view more 

Credit: Chalmers University of Technology, Mia Halleröd Palmgren

E-scooters have often been identified as more dangerous than e-bikes, but that picture changes when they are compared on equal terms. A recently published study from Chalmers University of Technology, Sweden, shows in fact that the crash risk is eight times higher for e-bikes than for e-scooters, calculated based on the trip distance with rental vehicles in cities. This surprising result provides a better basis for cities to make decisions on how much to facilitate different types of micromobility.

E-scooters have become an increasingly common, and often hotly debated, means of transport in cities. The media’s reporting on these crashes has been intense at times, and a number of previous studies have identified e-scooters as far more dangerous than e-bicycles and ordinary bicycles.

But according to a new study from Chalmers University of Technology in Sweden, it is not as simple as that.

“Previous studies have often compared apples with oranges,” says Marco Dozza, Full Professor in Active Safety and Road-User Behaviour at Chalmers. “They have lumped together e-bicycles with ordinary bicycles, and haven’t taken into account where, how and how much these vehicles are used – or whether they are rented or privately owned. When we took all these factors into account, we found that e-scooterists actually have a lower rate of crashes than e-cyclists.”

GPS data contributed to equitable comparison

The study, which is published in the Journal of Safety Research, is based on a unique data set from trips using rented e-bicycles and e-scooters in seven European cities: Gävle in Sweden, Berlin and Düsseldorf in Germany, and the UK cities of Cambridge, Kettering, Liverpool and Northampton.

The researchers analysed 686 crashes involving e-scooterists and 35 involving e-cyclists. The high number of crashes involving e-scooters reflects that they were used much more frequently than e-bicycles. But their crash risk was actually much lower – regardless of whether the risk was calculated on the basis of the number, duration, or distance of the trips.

“When we calculated using trip distance, it turned out that e-cyclists were eight times more likely to have a crash than e-scooterists. It’s a result that surprised us,” says Marco Dozza.

This is the first time that a study of this kind has been able to compare micromobility in such a detailed and equitable way, and from so many countries and cities. A key to being able to do the study in this way was the use of GPS data. This made it possible to measure what is termed ‘exposure’ – which refers to how much a vehicle is actually used – with greater precision than previously.

All vehicles in the study were rented and used in city centres, which makes the comparison more equitable than previous studies that have often mixed together urban and rural settings, or mixed rented vehicles with privately owned vehicles.

Safety of e-scooters grossly underestimated

Despite their results, the researchers stress that they should not be seen as definitive proof that e-scooters are safer than e-bicycles. Uncertainties remain, such as under-reporting of crashes and differences in the way these vehicles are used.

“But what we can say is that previous studies have grossly underestimated the safety of e-scooters in relation to e-bicycles,” says Marco Dozza. “This in turn could have consequences for how cities regulate and plan micromobility. In some cities, attempts are being made to steer micromobility towards e-bicycles, which are considered to be better because previous research has created the idea that all types of cycling are safer than all types of e-scootering,” he adds.

“Now that it turns out that isn’t correct, decision-makers may need to think again. The results might also affect consumers’ decisions if they have rented e-bicycles instead of e-scooters because they believed it’s safer,” he says.

According to the researchers, future analyses of crash risk should always include GPS data and precise information about how the vehicles are used. They would also like to see additional comparable data sets from other parts of the world; in particular, data sets that include more e-bicycle journeys in order to improve statistical reliability.

“With more detailed data, we can make better decisions about transport for the future. And to achieve that, it’s important that we compare apples with apples,” says Marco Dozza.

 

More about the research

The study only compares e-scooters with e-bicycles, unlike previous studies where e-bicycles and ordinary bicycles were lumped together in the same group. It is also the first study to also include several other important factors in the comparison: ownership, geographical location, usage, and exposure.

• Only rented vehicles were included in the study.
• The locations were limited to highly urbanised city centres using geofencing.
• Usage type was further controlled by comparing e-scooters and e-bicycles from the same rental company.
• Exposure was investigated using three different measures: number, duration, and distance of the trips.

The difference in crash risk between these vehicle types was greatest when trip distance was used as the measure for exposure, when the crash risk was 8.3 times higher for e-bicycles than for e-scooters. But even when using the other two measures for exposure, the crash risk was considerably higher for e-bicycles.

The data in the study comes from GPS data from trips with rented e-scooters and e-bicycles in seven European cities in the years 2022–2023 and includes a total of 686 reported crashes with e-scooters and 35 with e-bicycles. Despite the low number of crashes with e-bicycles, the results of the study are statistically significant when the data from all the cities was weighed together.

The study – Is e-cycling safer than e-scootering? Comparing injury risk across Europe when vehicle-type, location, exposure, usage, and ownership are controlled – was funded by Trafikverket, the Swedish Transport Administration.

 

Previous press releases about Marco Dozza's research on micromobility:

E-scooter crashes mainly caused by reckless driving

Why your e-scooter may pose higher risks than a rental

How e-scooters can safely operate in a city

---

Journal

Journal of Safety Research

DOI

10.1016/j.jsr.2025.06.015 

Method of Research

Observational study

Subject of Research

People

Article Title

Is e-cycling safer than e-scootering? Comparing injury risk across Europe when vehicle-type, location, exposure, usage, and ownership are controlled

Article Publication Date

18-Jun-2025

COI Statement

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Swedish Transport Administration funded the research. The authors have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

 

Ditches as waterways: Managing ‘ditch-scapes’ to strengthen communities and the environment

Virginia Institute of Marine Science

Facebook

WeChatBlueskyMessageWhatsAppEmail

 

image: 

Scientist Chelsea Clifford explores a ditch near Woodville Park in Gloucester, VA. Photo by John Wallace

view more 

Credit: Photo by John Wallace

Ditches are all around: along roads, through neighborhoods, across fields and marshes. These human-made waterways are so common that they can be easy to miss. A new literature review published in Nature Communications Earth & Environment calls on the public to pay more attention to this often neglected resource, one that could advance sustainability goals and benefit local communities with modern ditch management strategies.

In the English language, “ditch” has a bad rap. It evokes images of trash or something that ought to be discarded. That negative connotation and the everywhereness of ditches helps explain why they’ve been understudied and undervalued, according to the authors of the “Lines in the Landscape” article. 

This paper was produced by a collective of self-described “ditchologists”, whose informal talks at conferences and over email eventually culminated in a series of Zoom meetups that led to a more organized research effort. 

“As that group was established and grew, it became apparent that us ditchologists aren’t perhaps quite as rare as we thought,” said co-lead author Michael Peacock of the University of Liverpool. “And so we had the idea to try and get funding to arrange a hybrid meeting to bring forward different views and research backgrounds, and learn from each other what we think ditches ‘do’, and why they matter to us all.”

The dozens of authors involved looked at ditches as waterways from multiple angles: their social and economic aspects, physical and chemical characteristics and the plants and animals which inhabit these ecosystems.

“Ditches aren’t actually boring,” said co-lead author Chelsea Clifford, a former research scientist at William & Mary’s Batten School of Coastal & Marine Sciences & VIMS. “Try taking a closer look at yours.”

The overlooked history of ditches

Ditches are a staple of human history, with archaeologists dating them as far back as 8,000 years ago. They have been built on every continent, even Antarctica. Loosely defined by the paper as “linear constructions which store and/or move water where humans prefer it to go,” these channels can provide irrigation, drainage, power and more.

On the other hand, ditches can also cause harm and environmental damage, like spreading pollutants, proliferating invasive species and emitting greenhouse gases. 

Regulations could help ensure that ditches benefit people and their environments more than harm them, but the paper points out that legislation, including the U.S.Clean Water Act, doesn’t directly address them. That omission, the authors argue, has consequences.

“When this lack of regulation is combined with their capacity to carry harmful pollutants and pathogens, whose load is typically elevated, often deliberately concentrated, in areas of high poverty,” the paper reads, “then artificial channels [like ditches] become hotspots for environmental injustice and exploitation.”

Innovations in ditch management

At the same time, the absence of extensive oversight provides an opportunity to experiment and innovate. In the study and maintenance of ditches, there are many areas where creative thinking is needed, such as figuring out how to plot where ditches are.

“Not only do we usually not have very good maps of them,” Clifford said, “recent updates of maps of waters in general have sometimes cut them out.”

Paper co-authors, however, have made progress in charting the vast global network of ditches using the latest technology.

“Technology has advanced from the object-based image analysis methods we used in our 2017 publication to the modern-day machine learning and artificial intelligence methods used in our 2024 article,” said co-author John Connolly of Trinity College Dublin. “My interest in mapping these lines in the landscape and considering their impact on society in general and peatlands in specific has increased – especially since finding our community of ditchologists.”

Ditches can host unique ecosystems, featuring fish, amphibians, and waterbirds, as well as rare species like the fen ragwort flower or ram’s-horn snails, adding to local biodiversity.

But similar to other aquatic ecosystems, ditches can also bear high nutrient loads and take on runoff, including various pollutants like pesticides, trace metals and microplastics. Fortunately, this tendency could be leveraged to better mitigate pollution with appropriate ditch management strategies. Moreover, because ditches are often part of larger waterway networks, they can make it possible to spot water quality issues before they flow downstream.

Clifford said that this predictive property of ditches extends to anticipating sea level rise and its ramifications. “They can be sentinels of what’s to come before the surrounding landscape also changes to marsh, and then to open water,” she said.

The potential environmental benefits of ditches 

This literature review is a major step toward bringing ditches out of obscurity as a subject of study, and the authors shine light on possible paths for future research and practice. Even though the number of ditch-centric regulations is slim, there are guiding documents that could assist in crafting legislation like the European Union’s Water Framework Directive, which advises on “heavily modified and artificial waterbodies."

Part of the challenge with bringing ditches to their full potential as waterways is that their complexity demands multidisciplinary study and coordinated ditch management. While difficult, such synergies are possible if ditches are seen at the landscape scale, as “ditchscapes.”

Ultimately, ditches can be a source of optimism about environmental conservation and restoration, said Clifford.

“If we’re smart and brave and take initiative about what could be,” Clifford said, “and start trying to manage ditches as corridors to the future rather than just relicts of mistakes of the past, they could become symbols of hope too.”

Visit the Nature website for the full article.

---

Journal

Nature Communications

DOI

10.1038/s43247-025-02699-y 

Method of Research

Literature review

Article Title

Lines in the landscape

 

Who shows up in times of need? High school extracurriculars offer clues

A Rutgers study links participation in nonathletic clubs and volunteering in high school to higher levels of altruism later in life

Rutgers University

Facebook

X

Are nerds the caring ones?

High school stereotypes suggest that athletes score more popularity points than marching band members, debaters or leaders in the student council, but research from Rutgers finds that so-called “geeky” activities may do more to cultivate compassion in the long run.

“By their very nature, sports encourage competition and division, pitting people against each other,” said Chien-Chung Huang, a professor at the Rutgers School of Social Work and lead author of the study published in Youth & Society.

“There are other afterschool activities that do a far better job nurturing altruism.”

Extracurriculars have long been linked to higher grades, self-esteem and personal growth. Less understood is whether they foster generosity later in life, particularly during crises.

To explore the broader social impact of extracurricular activities during high school, Huang and colleagues examined data from the long-running Future of Families and Child Wellbeing Study, which has followed thousands of participants nationwide since 1998.

As part of that survey, in 2022, more than 2,700 respondents answered questions related to their participation in high school extracurriculars – activities that sit outside of the required school curriculum – between 2014 and 2018.

About 87% of respondents reported participating in after-school organizations during their high school years. Roughly 70% reported engaging in altruistic behaviors during the COVID-19 pandemic, including whether they ran errands, offered childcare or provided financial support to others.

 

Using statistical modeling, Huang and colleagues then analyzed the two sets of responses.

The study revealed a clear pattern: The more involved students were after school, the more likely they were to help others in times of need. Volunteer service ranked highest in predicting altruism, followed by school clubs, hobbies, religious services and performance groups. Sports – the most popular activity by participation – ranked last for altruistic benefit.

For policymakers and school officials, Huang said, the findings should be a wake-up call to how sports are integrated into student development. While athletes often do better in school, are strong leaders and live healthier lives, the narrow focus on competition means there’s plenty of room for improvement.

“Coaches could bring a component of volunteering into their sports programs,” Huang said. “Maybe teams could do an activity each month where they volunteer at a community center or retirement home. This would bring teams closer together and add an altruistic element.”

Still, Huang emphasized the findings aren’t a criticism of sports, per se, but a reminder that any activity can include opportunities to care for others.

After-school programs of all kinds, including athletics, “can play a critical role in preparing youth to contribute meaningfully during times of societal need,” he said.

---

Journal

Youth & Society

DOI

10.1177/0044118X251350719 

Method of Research

Data/statistical analysis

Subject of Research

People

Article Title

High School Extracurriculars and Altruistic Behavior in Crises: Insights From the COVID-19 Pandemic

 

Training doctors for the digital age: Canadian study charts new course for health education

New study introduces a comprehensive framework to guide digital health training across Canada

JMIR Publications

Facebook

XLinkedIn
MessageWhatsAppEmail

 

image: 

New study introduces a comprehensive framework to guide digital health training across Canada

view more 

Credit: JMIR Publications

(Toronto, September 12, 2025) As Canada’s health care system rapidly adopts digital technologies, a group of Canadian researchers is calling for a major overhaul of health professional education to ensure consistent, outcomes-based training in digital health and informatics competencies. A new article published in JMIR Medical Education by researchers at the British Columbia Institute of Technology and University of Calgary proposes using the Quintuple Aim as a national guiding framework to prioritize the digital health skills health care workers need now and in the future.

The paper, titled “Shaping the Future of Digital Health Education in Canada: Prioritizing Competencies for Health Care Professionals Using the Quintuple Aim” and published by JMIR Publications, argues that current education is fragmented and inconsistent, leaving health professionals underprepared to use tools like telehealth, electronic health records, and data analytics effectively. The proposed framework offers a solution: to align digital health competencies with five key goals—improving patient experience, boosting population health, lowering health care costs, enhancing provider experience, and advancing health equity.

This model helps identify and prioritize core skills such as digital literacy, privacy awareness, user-friendly technology integration, data-informed decision-making, and inclusive access. The authors also emphasize that training programs should include practical, real-world assessments—like simulation exercises and project-based evaluations—to ensure professionals are ready to apply their knowledge in the field.

“Additional professional development opportunities in digital health are essential to support scaled and sustainable change in Canada’s health systems that can truly create opportunities for better outcomes for all,” says author Tracie Risling from the University of Calgary.

While calling for national standards, the article allows room for local adaptation, encouraging educational institutions to customize learning based on regional health needs and resources. The authors also emphasize the importance of collaboration among health care organizations, educational institutions, and technology developers to ensure that programs keep pace with rapid innovation.

Ultimately, this research signals a clear call to action: to prepare Canada’s health workforce for the digital age, the country needs cohesive, forward-thinking education strategies now more than ever.

 

Please read the full article here:

Rees G, Nowell L, Risling T. Shaping the Future of Digital Health Education 

in Canada: Prioritizing Competencies for Health Care Professionals Using 

the Quintuple Aim. JMIR Med Educ. 2025;11:e75904

URL: https://mededu.jmir.org/2025/1/e75904

DOI:  doi:10.2196/75904

 

###

 

About JMIR Publications

JMIR Publications is a leading open access publisher of digital health research and a champion of open science. With a focus on author advocacy and research amplification, JMIR Publications partners with researchers to advance their careers and maximize the impact of their work. As a technology organization with publishing at its core, we provide innovative tools and resources that go beyond traditional publishing, supporting researchers at every step of the dissemination process. Our portfolio features a range of peer-reviewed journals, including the renowned Journal of Medical Internet Research.

 

To learn more about JMIR Publications, please visit jmirpublications.com or connect with us via Twitter, LinkedIn, YouTube, Facebook, and Instagram.

Head office: 130 Queens Quay East, Unit 1100, Toronto, ON, M5A 0P6 Canada

Media contact: communications@jmir.org

The content of this communication is licensed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, published by JMIR Publications, is properly cited.

---

Journal

JMIR Medical Education

DOI

10.2196/75904 

Method of Research

Commentary/editorial

Subject of Research

People

Article Title

Shaping the Future of Digital Health Education in Canada: Prioritizing Competencies for Health Care Professionals Using the Quintuple Aim

Article Publication Date

8-Sep-2025

 

OpenAI, DeepSeek, and Google vary widely in identifying hate speech

Neil Fasching and Yphtach Lelkes of the Annenberg School for Communication have found dramatic differences in how large language models classify hate speech, with especially large variations for language about certain demographic groups, raising concerns

University of Pennsylvania

Faceboo

XLinkedInWhatsAppEmail

With the proliferation of online hate speech—which, research shows, can increase political polarization and damage mental health—leading artificial intelligence companies have released large language models that promise automatic content filtering. “Private technology companies have become the de facto arbiters of what speech is permissible in the digital public square, yet they do so without any consistent standard,” says Yphtach Lelkes, associate professor in the Annenberg School for Communication.

He and Annenberg doctoral student Neil Fasching have produced the first large-scale comparative analysis of AI content moderation systems—which social media platforms employ—and tackled the question of how consistent they are in evaluating hate speech. Their study is published in Findings of the Association for Computational Linguistics.

Lelkes and Fasching analyzed seven models, some designed specifically for content classification and others more general: two from OpenAI and two from Mistral, along with Claude 3.5 Sonnet, DeepSeek V3, and Google Perspective API. Their analysis includes 1.3 million synthetic sentences that make statements about 125 groups—including both neutral terms and slurs—ranging from ones about religion to disabilities to age. Each sentence includes “all” or “some,” a group, and a hate speech phrase.

Here are three takeaways from their research:

The models make different decisions about the same content

“The research shows that content moderation systems have dramatic inconsistencies when evaluating identical hate speech content, with some systems flagging content as harmful while others deem it acceptable,” Fasching says. This is a critical issue for the public, Lelkes says, because inconsistent moderation can erode trust and create perceptions of bias.

Fasching and Lelkes also found variation in the internal consistency of models: One demonstrated high predictability for how it would classify similar content, another produced different results for similar content, and others showed a more measured approach, neither over-flagging nor under-detecting content as hate speech. “These differences highlight the challenge of balancing detection accuracy with avoiding over-moderation,” the researchers write.

The variations are especially pronounced for certain groups

“These inconsistencies are especially pronounced for specific demographic groups, leaving some communities more vulnerable to online harm than others,” Fasching says.

He and Lelkes found that hate speech evaluations across the seven systems were more similar for statements about groups based on sexual orientation, race, and gender, while inconsistencies intensified for groups based on education level, personal interest, and economic class. This suggests “that systems generally recognize hate speech targeting traditional protected classes more readily than content targeting other groups,” the authors write.

Models handle neutral and positive sentences differently

A minority of the 1.3 million synthetic sentences were neutral or positive to assess false identification of hate speech and how models handled pejorative terms in non-hateful contexts, such as “All [slur] are great people.”

The researchers found that Claude 3.5 Sonnet and Mistral’s specialized content classification system treat slurs as harmful across the board, whereas other systems prioritize the context and intent. The authors say they are surprised to find that each model consistently fell into either camp, with little middle ground.

Yphtach Lelkes is an associate professor of communication in the Annenberg School for Communication, co-director of the Polarization Research Lab, and co-director of the Center for Information Networks and Democracy.

Neil Fasching is a doctoral candidate in the Annenberg School for Communication and member of the Democracy and Information Group.

This research was supported by the Annenberg School for Communication.

---

DOI

10.18653/v1/2025.findings-acl.1144 

Method of Research

Computational simulation/modeling

Subject of Research

Not applicable

Article Title

Model-Dependent Moderation: Inconsistencies in Hate Speech Detection Across LLM-based Systems