Walking After Stroke: What the LEAPS Trial Says About High-Tech Rehabilitation

A machine is not the whole treatment

Supported treadmill systems can make stepping practice possible while reducing some of the demands of bearing body weight. Their visual sophistication can create an assumption that they must outperform less elaborate exercise. The scientific question, however, is comparative: do they improve meaningful walking outcomes more than a realistic alternative for the population being treated?

The LEAPS trial addressed that question in people recovering from stroke. It did not compare a machine with neglect or no human support. Its home-exercise comparison was managed by a physical therapist. That distinction matters because a report saying “treadmill no better than home exercise” can otherwise sound as if independent, unassessed exercise was equivalent to supervised rehabilitation. That was not the tested claim.

For further context, see planning goal-directed rehabilitation.

What the study tested

Duncan and colleagues enrolled 408 participants and compared locomotor training beginning earlier or later after stroke with a progressive home exercise program. The treadmill intervention included body-weight support and a structured therapeutic approach. Participants differed in the severity of their walking impairment, which the trial considered in its design.

At one year, 52 percent of all participants had improved functional walking ability. The trial did not find a significant advantage for early or late locomotor training over the home-exercise comparison on its primary outcome. Groups also showed similar improvements across several secondary measures, including walking speed and balance. Improvement was real at the group level, but the specialized intervention did not establish superiority.

For further context, see why the comparator matters for rehabilitation technology.

Why a null superiority result still matters

The finding challenges the assumption that more elaborate equipment necessarily creates better recovery. It can support discussions of access and resource use, especially when transportation or specialized facilities are difficult to obtain. It does not establish that every home program is equally effective or that trained supervision can be removed. The comparator's structure and clinical management were part of the evidence.

Nor does the result prove that no individual can benefit from supported stepping. A group-average comparison cannot identify every person's best option. A clinician may have reasons to use equipment to enable practice or manage a specific limitation. Those reasons should be described honestly as individualized judgment rather than proof that the device will outperform a well-designed alternative.

Safety belongs in the headline too

LEAPS reported a higher frequency of dizziness or faintness during treatment in the locomotor-training groups than in the home-exercise group. Among participants with severe walking impairment, multiple falls were more common with early locomotor training than in the other groups. These findings do not mean treadmill rehabilitation is inherently unsafe, but they prevent a simplistic assumption that supported equipment removes all risk.

Walking rehabilitation involves cardiovascular tolerance, balance, attention and judgment as well as leg movement. A session can appear successful while leaving a person exhausted or unsafe later. Ask how adverse events are monitored, who assists with transfers and whether the plan accounts for falls outside treatment. Safety information is part of comparative effectiveness, not an optional disclaimer placed after the promised benefit.

Choose an outcome that reflects the goal

Walking speed is useful, but it is only one dimension of mobility. A person may care about reaching the bathroom safely, crossing a street, managing steps or walking far enough to shop. These goals differ in endurance, balance and environmental complexity. The same average speed can coexist with different levels of independence and fall risk.

Consider a hypothetical patient who can walk down a clinic corridor but avoids leaving home. The treatment discussion should identify why: fatigue, fear, uneven surfaces, transport barriers or another limitation. Increasing treadmill time may or may not address that obstacle. A plan that includes real-world assessment and appropriate aids can be more relevant than pursuing a device-based metric that does not capture the person's difficulty.

Questions before committing to a program

Ask how the proposed service resembles the intervention studied, what alternative is available and how progress will be reviewed. Clarify supervision, total travel and treatment burden, costs and the response to fatigue or dizziness. If a provider cites LEAPS, the explanation should acknowledge the lack of superiority rather than borrowing the trial's credibility while omitting its main result.

New stroke symptoms require emergency assessment, not exercise adjustment. For ongoing recovery, the durable message is hopeful but bounded: people improved across the trial's treatment groups, and specialized technology was not the only pathway to gains. Appropriate professional care, realistic comparisons and individualized safety planning matter more than the assumption that the most impressive machine must produce the most meaningful recovery.

Sources

• Body-weight-supported treadmill rehabilitation after stroke.
• NINDS: Stroke Recovery — background, not replication

Editorial disclosure: It is educational information, not individual medical advice. Prepared for October 30, 2026 using established research; it does not assert current regulatory status or latest findings.