Search for a command to run...
Electric scooters (e-scooters) have rapidly evolved from urban novelties to mainstream transport modalities since their commercial debut in 2017 [1]. Touted as affordable, convenient and environmentally sustainable, e-scooters have been championed by local governments and micromobility companies as solutions for congestion and urban mobility [2]. Yet, this enthusiasm and rapid uptake have often overshadowed growing concerns regarding safety and health system impacts. This Letter highlights the scale of the problem, outlines deficiencies in current data capture and calls for coordinated prevention strategies including helmet mandates, infrastructure reform, improved surveillance and integration of healthcare insights into policy. In many settings, the rise of e-scooters has been interpreted as a positive shift towards equitable and flexible mobility. Yet, this optimism has overshadowed a growing body of evidence demonstrating that e-scooters are also a significant, underregulated contributor to neurotrauma [1-3]. The implementation of these services has outpaced both regulatory foresight and health system preparedness, culminating in a preventable burden of traumatic brain injury (TBI) now observable across multiple jurisdictions [3, 4]. Nationally representative Australian data remain sparse and fragmented, meaning current estimates likely underestimate the true incidence and cost of e-scooter-related injury. WorkCover Queensland recorded 1321 e-scooter/e-bike injury claims costing over $25 million (2020–2025), while Royal Melbourne Hospital treated 562 riders—most injured from simple falls, with nearly a third involving alcohol, 1 in 10 involving drugs, and about 10% classed as major trauma, often linked to speeds over 20 km/h [5, 6]. However, these figures exclude unprovisioned costs such as uninsured compensation, litigation, and lost productivity, which fall outside statutory insurance and are rarely captured in current health or compensation datasets, meaning the true financial burden is substantially higher. Unlike motor vehicle crashes or high-velocity trauma, e-scooter injuries are largely low-energy impacts occurring in young, otherwise healthy individuals [7, 8]. Nonetheless, the outcomes are often severe [1]. Across international studies, between 4% and 40% of emergency department (ED) presentations for e-scooter injuries involve intracranial haemorrhage (ICH), and up to 12% require intensive care unit (ICU) admission [4]. Intracranial injury remains the most common indication for hospitalisation and critical care escalation [9]. These figures do not simply reflect a niche safety concern, but represent a distinct, modifiable cause of neurosurgical morbidity. The incongruity between injury severity and vehicle design underscores a deeper issue: a public health risk framed as incidental, when it is in fact structural. The mechanisms of injury are strikingly consistent. Helmet non-use is widespread among e-scooter riders, particularly within commercial rideshare schemes that neither provide nor mandate protective gear [4, 10, 11]. Alcohol intoxication is frequently reported, especially during evening and weekend hours when injury incidence peaks [4, 9-11]. Many users lack prior experience with e-scooters, and few receive any form of training [12, 13]. In this context, urban infrastructure acts not as a buffer, but as an amplifier of risk [8, 12]. Riders are routinely forced to navigate congested roadways, uneven footpaths and shared spaces without dedicated lanes, speed regulation or physical separation from traffic [4, 11]. These are not accidental oversights; they are predictable vulnerabilities arising from the absence of coordinated micromobility planning. Neurosurgical consequences are increasingly recognised, though still underrepresented in the literature [1, 2]. One of the largest retrospective series to date, conducted at a level 1 trauma centre, documented a diverse spectrum of injuries requiring neurosurgical input: subdural and epidural haematomas, traumatic subarachnoid haemorrhage, central cord syndrome and vertebral body compression fractures [1]. These injuries, while mechanistically distinct from those typically seen in high-energy trauma, are equally disabling. Many patients require emergency operative intervention, prolonged ICU admission and structured neurorehabilitation. The affected demographic, predominantly young adults, is notable not only for its vulnerability but also for the long-term socioeconomic impact of injury. Despite mounting clinical evidence, the true burden of e-scooter-related TBI remains inadequately quantified. National trauma registries often fail to capture micromobility-specific data, and many cases, particularly those involving mild TBI or post-concussive symptoms, go undocumented. The absence of longitudinal follow-up further obscures outcomes, particularly in patients discharged without inpatient rehabilitation. This data fragmentation reflects broader limitations in how emerging sources of neurotrauma are monitored and addressed. Relevant variables for capture include helmet use, alcohol or drug involvement, speed, crash mechanism, injury severity and compensation status, which together would allow more consistent surveillance and comparison across jurisdictions. Without standardised reporting frameworks, it is impossible to accurately estimate costs, identify trends or evaluate the effectiveness of prevention strategies. To significantly enhance data collection on head injuries secondary to e-scooter accidents, two key steps are recommended: mandatory reporting of minimum data sets to a central national agency and the introduction of individualised ICD-10-AM subcodes specific to e-scooter injuries. These measures would directly improve the tracking of clinical outcomes, accurately calculate costs by enabling reliable linkage to the Australian Refined Diagnosis Related Groups (AR-DRGs), and provide the robust, evidence-based data necessary for effective regulation. The policy landscape has been similarly inconsistent. In some jurisdictions, sustained public concern has prompted regulatory action [14]. Paris, for example, banned commercial e-scooter rental services in 2023 following multiple fatalities and rising trauma admissions [14, 15]. At least three Australian states have launched formal reviews into injury trends and legal frameworks governing e-scooter use [14]. Yet many other cities continue to expand e-scooter access without meaningful oversight, and few have enacted robust helmet legislation, infrastructure reforms, or safety data mandates. This reflects a broader failure of anticipatory governance, an inability to apply existing transport safety knowledge to novel technologies in a timely manner. The result is a system in which preventable injury has become normalised. Unlike cycling or motorcycling, where helmet use, lane separation and speed limits are increasingly standardised, e-scooter regulation remains fragmented and reactive. The absence of a coordinated response has left responsibility diffused between commercial operators, municipalities and individual riders, with no clear accountability for safety outcomes. Commercial platforms, despite their central role in enabling and collecting user behaviour data, have few obligations to contribute to injury prevention or public health surveillance. Voluntary measures, such as in-app safety prompts or geofencing, are inconsistently implemented and rarely evaluated. Yet the solutions are neither complex nor unprecedented. Helmet mandates, infrastructure adaptations, and targeted rider education are well-established strategies in other domains of transport safety [1, 13, 16, 17]. For instance, a meta-analysis of 21 studies found that mandatory bicycle helmet legislation reduces head injuries by around 20% overall and by up to 55% for serious head injuries [16]. Although these data are derived from cycling populations and the exact effect size for e-scooters is not yet established, the prevention principles are transferable and indicate a likely reduction in head-injury burden. For instance, real-time integration of user behaviour data with injury surveillance could allow dynamic, evidence-informed regulation of high-risk zones. Licensure conditions for rental operators could require contributions to public health data systems and enforce minimum safety standards. Embedding injury metrics into performance indicators would provide both transparency and accountability. The role of the healthcare community is not merely reactive. Clinicians, particularly those in emergency medicine, neurosurgery, trauma care, and rehabilitation, are uniquely positioned to characterise these injuries, identify missed opportunities for prevention, and shape public discourse. Yet this potential remains underutilised. Too often, health professionals encounter these injuries in isolation, without a system for capturing, reporting, or translating experience into policy reform. Greater interdisciplinary collaboration with transport planners, policymakers, and data scientists is urgently needed to integrate neurotrauma prevention into the fabric of urban mobility policy. E-scooter-related TBI is not the collateral damage of progress. It is the foreseeable consequence of regulatory inertia, data fragmentation, and a reluctance to assign shared responsibility for injury prevention. In a domain where the harms are both visible and avoidable, continued inaction cannot be justified. The rise of micromobility offers an opportunity, not just to rethink transport, but to reframe how health systems anticipate and mitigate new forms of trauma. With coordinated action, the balance can shift from reactive care to proactive prevention. But the window for intervention is narrowing. Without a decisive shift in policy, infrastructure, and data integration, the cost will be counted in lives disrupted and futures diminished, one preventable injury at a time. Peter Curpen: conceptualization, writing – original draft. Virginia Newcombe, Ramon Navarro, Ming Lu, and Fatima Nasrallah: conceptualization, writing – review and editing. The authors have nothing to report. The authors declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.