Clinics GRP Friday Research Update | Friday 2 October 2026
Rehabilitation research often reports improvement in strength, walking speed, balance or endurance. Those measures matter, but they are only part of the story. A person can walk further in a six-minute walking test and still walk very little at home. They can improve their stepping reactions in a laboratory and still fall in the community. They can become stronger without necessarily returning to the activities that strength was meant to support.
Several papers we reviewed this week come back to that distinction. The question is not only whether rehabilitation changes a test result. It is whether the change carries over into daily life.
That has implications for how we think about falls, stroke rehabilitation, chronic pain and even the way we use activity trackers with older adults.
Key points
- Improved walking or balance on a test does not always mean more activity or fewer falls in everyday life.
- Exercise can reduce injurious falls in community-dwelling older adults, but rehabilitation still needs individual assessment.
- Stroke rehabilitation research supports looking at both walking capacity and what happens outside treatment sessions.
- Exercise for pain can begin at a manageable dose and progress with the person’s response and goals.
Exercise is reducing more than fall risk factors
One of the more useful recent developments in falls research is a shift towards outcomes that patients actually care about: injury.
An updated systematic review brought together 29 randomised controlled trials involving 6,367 older adults living in the community. Exercise was associated with fewer injurious falls and fewer serious injurious falls. The estimated reduction was about 13% for injurious falls and 33% for serious injurious falls, with moderate certainty in those findings. The review also found possible reductions in falls requiring medical attention and fractures, although the evidence for those outcomes was less certain.
That is an important distinction.
For years, much of falls rehabilitation has been discussed through measures such as balance scores, gait speed, strength or a person's calculated risk of falling. Those measures help us understand capability, but the outcome that matters most may be whether a fall results in a fracture, hospital presentation or significant loss of function.
At Clinics GRP, this strengthens the case for exercise-based falls management, but it does not support a single "falls exercise". Effective rehabilitation is usually broader than that. Strength, walking, balance, stepping reactions, mobility and the demands of the person's own environment all need to be considered.
It also changes what we should follow over time. If reducing the consequences of falls is one of the person's goals, it makes sense to record not only whether they fell, but whether they were injured, required medical treatment or experienced a meaningful change in function.
Read the systematic review on fall-related injuries
Better balance does not guarantee fewer falls
A recent randomised trial provides a useful counterpoint.
Older adults completed six weeks of reactive stepping training: exercises designed to improve the ability to make a rapid step when balance is unexpectedly disturbed. Participants improved several measures of stepping performance, including maximum step length and reaction time.
What did not clearly improve was the outcome that ultimately matters: falls. The study did not find a significant reduction in unexpected laboratory falls or real-world falls over the following 12 months.
That does not make reactive stepping unimportant. It tells us that improving one part of the balance system may not be enough on its own.
Falls happen in messy circumstances. Someone may be turning while talking, stepping from a kerb, walking across an uneven lawn, carrying shopping, responding to a pet underfoot or getting up quickly at night. Vision, attention, strength, sensation, reaction time, medication and the environment may all be involved at once.
Good rehabilitation therefore has to move beyond mastering an exercise. We need to consider whether the capability being developed can be used when the person's real environment becomes unpredictable.
Stroke rehabilitation is exposing a gap between capacity and everyday activity
The same issue is appearing very clearly in neurological rehabilitation.
A systematic review of 31 studies after stroke found that exercise improved walking speed and walking endurance. These are meaningful rehabilitation gains. Yet the change in the amount people actually walked during the day was much smaller.
Interventions that deliberately addressed behaviour and daily activity tended to produce larger increases in step count.
This gives clinicians two different questions to answer:
- What is this person capable of doing?
- What are they actually doing in their everyday life?
They are not interchangeable.
A person may regain enough capacity to walk 400 metres but rarely walk more than 50 metres at a time at home. The limiting factor may no longer be cardiovascular fitness or leg strength. It could be confidence, fatigue, habit, fear of falling, pain, a difficult environment, transport, lack of opportunity or simply the absence of a reason to walk further.
Continuing to increase physical capacity without understanding that gap may eventually produce diminishing returns.
For that reason, we increasingly see value in combining clinical testing with simple measures of real-world activity where it is useful. That might involve step counts, walking bouts, activity logs or another practical measure that tells us what happens between appointments.
The number itself is not particularly interesting. The relationship between the person's capacity, activity and goals is.
Read the review of walking capacity and real-world activity after stroke
How much rehabilitation can people after stroke tolerate?
An Australian stroke feasibility trial explored a very different question: how much rehabilitation can people actually do?
Participants in the HiWalk programme were offered up to three hours of group-based walking practice each day for three weeks. They attended approximately 91% of available sessions and completed a substantial number of walking repetitions. The participants were, on average, 58 years old.
The study does not establish that three hours of walking practice is the ideal dose, nor that every person recovering from stroke should receive such an intensive programme. The effectiveness evidence is still developing.
It challenges the assumption that people living with neurological disease necessarily require small doses of rehabilitation. Whether the same approach is feasible and beneficial for older adults needs direct study.
Some people may tolerate considerably more repetition when training is well organised, adjusted for fatigue and supported appropriately. For selected patients, that raises an interesting future possibility: short periods of intensive rehabilitation followed by a return to normal daily activity and consolidation.
The right dose will still depend on the person. Age alone should not determine it.
Read the HiWalk feasibility trial
Exercise for chronic pain: the dose does not always need to be large
A large umbrella review has also added useful context to exercise for pain, including musculoskeletal rehabilitation.
Researchers examined evidence from 157 systematic reviews, representing 2,736 randomised controlled trials and more than 221,000 participants. Across a wide range of painful conditions, exercise was associated with reduced pain. The review rated the overall evidence as moderate certainty.
Aerobic exercise, resistance training, Pilates, yoga, tai chi and other exercise approaches all showed benefit in different settings. One finding is particularly relevant to older adults who are worried about beginning exercise while they are sore: pain improvement did not depend on starting with very high exercise intensity.
That does not mean light exercise is always the best prescription. It means we do not need to wait for pain to disappear before movement becomes useful, and we do not need to push someone hard simply to make the exercise therapeutic.
This fits with another recent review of chronic low-back pain. Higher exercise doses did not appear to produce proportionately larger benefits, although the certainty of that dose-response evidence was low.
Read the review of exercise dose for chronic low-back pain
Our interpretation is fairly practical. Start at a dose the person can manage consistently, monitor the response, and build from there according to function and goals. Exercise can be part of treating pain, rather than something that begins only once pain has been treated.
Read the exercise and pain umbrella review
Activity trackers are useful only when the information is useful
Wearable technology is increasingly finding its way into rehabilitation. That includes watches, pedometers and other devices capable of measuring steps or activity.
A recent qualitative study asked older adults what made them continue using these devices. People were more likely to persist when the technology was convenient, understandable and clearly connected with their health. Technical difficulty, cost and a lack of perceived usefulness made continued use less likely.
That sounds obvious, but it matters.
There is little value in asking someone to collect thousands of data points because the technology allows us to do it. If we use activity monitoring in rehabilitation, the measurement should answer a clinical question.
Suppose a person's walking endurance improves significantly during treatment, but their daily walking remains unchanged. That discrepancy tells us something useful. It may direct the next conversation towards confidence, fatigue, pain, routine, the home environment or opportunities for activity.
In that situation, the wearable has not provided the treatment. It has helped us identify where the treatment needs to go next.
Read the study of older adults using activity monitors
Medication still belongs in the falls conversation
Falls are rarely explained by a single factor, and medication remains an important part of the assessment.
A newly published systematic review examined 19 studies involving more than 183,000 participants taking dual orexin receptor antagonists, a newer class of medication used to treat insomnia. Across the available evidence, these medications were not associated with a statistically significant increase in falls or fractures.
The finding needs to be treated cautiously. The number of fall events was relatively small and older adults were under-represented in parts of the evidence base. It would therefore be premature to conclude that these medications are risk-free or protective against falls.
There is a more useful lesson for rehabilitation.
"Sleeping tablet" is not a particularly informative medication history. Different drug classes act differently, and their effects on alertness, blood pressure, cognition and fall risk may not be equivalent.
For someone experiencing falls or near-falls, knowing what medications they take — and when they take them — can help the clinical team identify questions that should be discussed with the person's GP or pharmacist.
Patients should not stop or change prescribed medication on the basis of falls research without discussing it with their prescriber.
Read the review on insomnia medication, falls and fractures
How Clinics GRP is interpreting this week's research
There is no new exercise technique in this week's research that we think every older adult suddenly needs.
There is something more useful emerging across the evidence.
We need to become better at distinguishing capacity from performance.
Strength matters, but we also need to know what the person can use that strength for. Walking endurance matters, but we also need to know whether the person is walking. Better stepping reactions matter, but we need to know whether those reactions transfer into the situations in which balance is actually lost.
That does not reduce the importance of clinical measures. It gives them context.
For us, rehabilitation should progressively answer four questions:
- What can this person currently do?
- What do they want or need to be able to do?
- What is preventing that from happening now?
- Is the rehabilitation changing their life outside the treatment session?
Sometimes the answer will be more strength. Sometimes it will be more repetition, better balance, pain management, confidence, a change in the environment or simply an opportunity to practise the activity that matters.
A test tells us something about the person. It should never become the whole person.
Older Adult Rehabilitation Research
This article is part of the Clinics GRP Older Adult Rehabilitation Research series, published each Friday. We review emerging evidence in falls, balance, neurological rehabilitation, musculoskeletal conditions, exercise and healthy ageing, then explain what we think is relevant to clinical care and everyday life.
Research findings apply to populations, not automatically to an individual. Decisions about rehabilitation, medication and health conditions should be made with the health professionals involved in your care.