Cycling into Cognitive Health: Lessons from the U.K. Biobank

By Ellen Feldman, MD

Synopsis: In this large prospective study, active travel (especially cycling) was associated with lower dementia incidence both overall and for specific types of dementia as well as greater hippocampal volume.

Source: Hou C, Zhang Y, Zhao F, et al. Active travel mode and incident dementia and brain structure. JAMA Netw Open. 2025;8(6):e2514316.

Dementia is a growing global health burden projected to affect 139 million people by 2050.¹ Young-onset dementia (YOD) affects nearly 4 million people worldwide and carries higher care demands and earlier mortality compared to late-onset dementia (LOD).^1,2 The 2024 Lancet Commission estimated that nearly half of dementia cases may be linked to modifiable risk factors, with physical activity during midlife noted to be especially protective.³ However, population-level evidence on “active travel” (such as walking and cycling as transportation) remains limited.

Active travel incorporates exercise into a daily routine. Rather than a trip to the gym, this pattern of activity fits seamlessly into leisure time or errands, making it highly accessible. Hou et al conducted a large prospective study to clarify whether travel modes are associated with dementia incidence and brain structure and whether genetic risk (especially apolipoprotein E4 [APOEε4]) modifies these associations.

This study drew on data from the U.K. Biobank, recruiting more than 479,000 adults aged 40-69 years between 2006 and 2010 with a median follow-up of 13.1 years.⁴ Participants reported their most common travel mode (excluding work commute) over the prior four weeks, categorized as:

• nonactive (car, public transportation only);
• walking;
• mixed walking (walking plus non-active mode); or
• cycling and mixed-cycling (cycling alone or combined with any other mode).

The primary outcome was incidence of all-cause dementia, with sub-analysis for Alzheimer’s disease (AD), YOD, and LOD. Dementia cases were identified from hospital records and death registries. Brain structure was evaluated in a subset using magnetic resonance imaging (gray matter and hippocampal volume). Models were adjusted for sociodemographic factors, health behaviors, comorbidities, cognitive function, and genetic risk.

Of the 479,723 individuals in this study (mean age 56.5 years; 54% female) 8,845 developed dementia (1.8%), including 528 with YOD and 3,956 with LOD.

In all models, cycling/mixed cycling was associated with significantly decreased incidence of dementia with a hazard ratio (HR) of 0.81 (95% confidence interval [CI], 0.73-0.91) for all-cause dementia; HR, 0.78 (95% CI, 0.66-0.92) for AD; HR, 0.60 (95% CI, 0.38-0.95) for YOD; and HR, 0.83 (95% CI, 0.75-0.93) for LOD.

Mixed walking was next and was associated with a more modest reduction in dementia (HR, 0.94; 95% CI, 0.89-0.98), while walking alone did not appear protective and was linked with a slightly higher AD risk (HR, 1.14; 95% CI, 1.01-1.29).

When looking at genetic factors, cycling’s protective effect was stronger among non-APOEε4 (HR, for all-cause dementia, 0.74 [95% CI, 0.63-0.87]) compared to carriers (HR, 0.88; 95% CI, 0.76-1.02).

Neuroimaging was conducted in a subset of approximately 38,000 participants. Cycling and mixed-cycling were associated with greater hippocampal volume and higher gray matter volume across multiple brain regions. In contrast, both walking and mixed walking were linked to smaller gray and white matter volumes compared with cyclists, but participants in these groups showed slightly larger hippocampal volumes than those who relied solely on nonactive transport. The nonactive group demonstrated the lowest overall brain volumes.

While this study was large and included long-term follow-up, several limitations should be noted. Travel mode was self-reported for a brief period and may not accurately represent long-term behavior. The cohort lacked diversity, with more than 85% of participants of European ancestry, which may limit generalizability. In addition, the number of YOD cases was relatively small. Finally, the neutral or adverse associations observed for walking could reflect factors, such as lower activity intensity, shorter distance, or reduced cognitive engagement compared with cycling or driving. These potential explanations were not directly measured and remain important areas for future study.

Commentary

This large cohort study highlights the importance of daily lifestyle choices in shaping long-term brain health. While the benefits of structured physical activity for cardiovascular and metabolic outcomes are well-established, the association between ordinary travel habits and dementia risk has been less clear. By focusing on “active travel,” particularly cycling, the study reframes dementia prevention from abstract ideal to something that can be built into the rhythms of daily life.

One of the most notable findings was that cycling, whether it was the sole form of transport or in combination with other modes, was associated with a substantially lower risk of dementia across all categories, including YOD, LOD, and AD. This protective effect was further supported by neuroimaging evidence, with cyclists showing increased hippocampal volume, a brain region central to memory and highly vulnerable in AD. The convergence of epidemiological outcomes with brain structured data strengthens the biological possibility of the observed associations.

The results related to walking are more complex. Walking alone was not protective and even showed a modest association with increased dementia risk. This contrasts with the consistent benefits seen with cycling. Several explanations are possible. Walking varies enormously in intensity and duration. Casual or short-distance walking may not be sufficient to trigger the protective mechanisms associated with more vigorous activity. Moreover, cognitive engagement may be a critical factor. Cycling requires navigation, balance, spatial awareness, and sustained effort. All of these features may stimulate hippocampal plasticity. Walking, by contrast may be less cognitively demanding unless combined with other activities. This could explain why mixed walking, which often includes travel by car or other modes requiring higher levels of attention, showed a small reduction in dementia risk compared with walking alone.

The interaction with genetic risk also is significant. Participants without the APOEε4 allele derived greater protective effects from cycling, yet even among carriers, the association remained directionally favorable. This emphasizes that lifestyle modifications, while not fully overcoming genetic predisposition, still contribute meaningfully to risk reduction. For clinicians, this reinforces a central message: A family history of genetic risk should not lead to resignation or inaction. Patients at higher risk may stand to gain the most from interventions that promote brain resilience.

For primary care providers, these findings offer practical points for patient counseling. Many middle-aged patients express interest in dementia prevention, but they struggle to connect abstract recommendations about “staying active” with their daily routines. This study provides a concrete example that incorporating cycling into transportation choices can support brain health. Framing exercise in terms of function, such as “bike to the store” or “cycle to work” or “use active transport when possible,” may resonate more than prescribing gym sessions or generic “more exercise.” Such approaches also align with broader goals of sustainability and cardiovascular fitness, reinforcing the multifaceted benefits of active travel.

The neuroimaging findings can be particularly useful for patient communication. Some patients may find it compelling to hear that cycling is associated with a larger hippocampus, the brain’s memory center. Relating behavior to visible structural benefits can make prevention strategies feel more tangible and motivating.

Of course, practical barriers must be acknowledged. Not all patients have access to safe cycling infrastructure, and some may face physical or cultural barriers. Future studies may want to incorporate other forms of active, cognitively engaging physical activities.

Finally, while the observational design cannot prove causality, the consistency across dementia incidence, subtypes, and brain imaging suggests that the associations are unlikely to be spurious. The study scale, long-term follow-up, and robust adjustments further increase confidence in its findings. Future research in more diverse populations and with objective activity tracking should help clarify these relationships.

Overall, this study provides compelling evidence that how people travel has implications for brain health. For primary care professionals, it underscores the value of integrating active travel into discussions of dementia prevention and healthy aging. The key message is not simply to “walk more” but to encourage patients toward more vigorous, engaging, and sustainable forms of daily activity. Cycling emerges as a promising and accessible option that may not only reduce dementia risk but also can support broader goals of cardiovascular health and mobility.

Ellen Feldman, MD, works for Altru Health System, Grand Forks, ND.

References

1. Alzheimer’s Disease International. Dementia statistics. https://www.alzint.org/about/dementia-facts-figures/dementia-statistics/
2. He Q, Wang W, Zhang Y, et al. Global burden of young-onset dementia, from 1990 to 2021: An age-period-cohort analysis from the global burden of disease study 2021. Transl Psychiatry. 2025;15(1):56.
3. Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024;404(10452):572-628.
4. U.K. Biobank. Health research data for the world. https://www.ukbiobank.ac.uk