VR Treadmills, Pilates, and the Evidence on Fall Prevention

A virtual-reality treadmill can improve balance tests more than the same treadmill training without the virtual layer. That is interesting. It does not yet show that the device prevents falls—the outcome older adults and clinicians actually care about.

For fall prevention, the best direct evidence still supports conventional multi-component exercise: programs that combine balance and resistance training, sometimes with walking or aerobic work. A systematic review of trials in ambulatory, community-dwelling adults aged 65 and older reported fall-risk reductions of up to 50% when programs met the World Health Organization target of at least 150 minutes of moderate activity per week. Technology-enhanced exercise, Pilates, and dual-task training can improve mobility and balance measures, but they have not matched that hard-endpoint evidence in healthy older adults.

The VR treadmill result is real, but the endpoint matters

In a 2024 randomized controlled trial of C-Mill virtual/augmented-reality treadmill training, 60 adults aged 60 to 70 trained for six weeks, three times weekly. Both groups received the same specialized treadmill gait training plus 30 minutes of conventional exercise each session. The difference was that one group used VR/AR gamification and the other used the unprojected version of the treadmill protocol.

Both groups improved significantly after training on the Timed Up and Go (TUG), the Biodex Balance System fall-risk index, and limits of stability. The VR/AR group improved more on all three measures. Four weeks after training stopped, both groups still retained statistically significant gains, and the VR/AR group retained its relative advantage.

That active-control design deserves credit. It controls for treadmill exposure and exercise volume better than a comparison with no exercise would. The four-week follow-up also tests whether gains outlast the last session, a question many short exercise trials never ask.

Still, the trial did not measure fall incidence, injurious falls, or fall-related hospital visits. Its "fall-risk index" is a balance-system measure, not a count of falls in daily life. The machine is also specialized and expensive, and both trial arms used it. The study therefore tested gamified versus unprojected gait training on the same platform; it did not test whether a C-Mill beats a community strength-and-balance class, Pilates, or a home program.

Old-school exercise has the outcome that counts

The 2023 systematic review of exercise for fall prevention focused on ambulatory, community-dwelling adults aged 65 and older. Its included multi-component programs combined resistance and balance work, with or without aerobic exercise and walking. These programs consistently reduced fall rates, and WHO-compliant doses were associated with reductions of up to 50%.

"Up to" is not an average effect, and the review's narrative synthesis does not make every program interchangeable. Training had to be maintained, and the review excluded VR/exergames, Tai Chi, Otago, yoga, and dance by design. It cannot therefore tell us whether a technology-enhanced program prevents more falls than conventional multi-component exercise. But it does establish the standard: if a newer modality claims superiority, it should eventually beat—or at least match—a program with recorded falls as an endpoint.

Exercise dose also matters more than novelty. In a 2021 trial of balance exercises during tooth brushing, healthy older adults did not improve balance or muscle strength despite daily repetition. A balance challenge has to be substantial and progressive enough to change capacity. A game console cannot negotiate that requirement away.

Pilates improves fall-risk factors, not proven fall rates

Pilates has a more modest but useful evidence base. In a 2022 randomized trial, 61 healthy older adults with a mean age of 70.08 years attended Pilates twice weekly for 12 weeks or remained in a passive-control group. Pilates produced significant between-group improvements in step time and double-support time. The Functional Reach Test showed a significant time-by-group interaction, which supports a Pilates-specific improvement in dynamic anterior balance after accounting for change over time.

The same trial found within-group changes in TUG and mediolateral balance, but those changes did not reach between-group significance. It recorded no reduction in fall incidence during its 12-week window. At this dose, Pilates can improve selected balance and gait measures; the study does not establish that it prevents falls.

A 2021 systematic review and meta-analysis of Pilates trials reached a similar conclusion. Pilates reduced fear of falling and improved functional mobility compared with inactive controls. Comparisons with active exercise produced smaller and inconsistent advantages, and the review found insufficient evidence that Pilates reduces fall incidence in healthy older adults.

Fear of falling still matters. It can lead people to restrict activity, which can worsen deconditioning. But lower fear is not the same outcome as fewer falls. The distinction keeps the conclusion honest.

Exergames can beat active comparators on balance tests

A 2022 systematic review and meta-analysis limited its analysis to randomized trials in healthy adults aged 60 and older that compared exergames with active training such as strength, Tai Chi, or balance exercise. Exergames produced statistically significant advantages for dynamic balance and fall-related outcomes over traditional physical training.

That is a stronger comparison than a couch-versus-console study. Yet the authors reported substantial heterogeneity between trials, the advantage varied by outcome, and the available studies did not identify a clean minimum effective dose. The pooled findings support exergames as a plausible way to train balance; they do not establish a universal effect size or a prescription that fits every older adult.

A separate six-week exergame randomized trial found improvements in physical and affective physical-literacy domains in older adults. This supports short-term feasibility and functional engagement, not long-term fall prevention. No technology-enhanced trial discussed here has reported hard fall outcomes as its primary endpoint.

Does the cognitive challenge explain the benefit?

Exergames often ask participants to move, react, attend, and make decisions at once. That resembles motor-cognitive dual-task training. A 2022 meta-analysis of randomized trials found that motor-cognitive interventions improved selected gait and balance outcomes in older adults. Related evidence in stroke populations also finds better gait and balance outcomes with cognitive-motor dual-task training than with single-task approaches, but stroke rehabilitation cannot be treated as evidence for healthy community-dwelling adults.

The mechanism remains unsettled. Trials that directly compared motor-cognitive with motor-motor dual-task training in older adults, and cognitive-plus-instability resistance training with instability resistance training alone in elderly women, complicate the claim that added cognitive demand is the main reason exergames work. Motivation, novelty, adherence, and the physical challenge itself may each contribute. No study here shows that cognitive gains predict fewer future falls better than physical training alone.

What to do with this evidence now

If the goal is preventing falls, choose a sustained multi-component program that includes balance and resistance work. It has the strongest direct evidence and does not require an expensive machine. Pilates is a reasonable accessible option for improving selected balance measures, mobility, and fear of falling, especially for someone likely to keep attending. Exergames and VR treadmill programs may make balance training more engaging and may improve surrogate outcomes beyond some active comparators.

The unanswered question is practical and testable: can a VR or exergame program reduce actual falls at least as well as conventional multi-component exercise? The decisive trial would use an active multi-component control, track fall incidence over a clinically meaningful period, report adherence, and follow participants for at least six months after training ends. Until then, the flashy option has better balance-test data than its predecessor, while conventional exercise has the stronger record on falls.