This Week in Physical Medicine & Rehabilitation — Aug 29, 2026

Welcome to This Week in Physical Medicine and Rehabilitation. This week we're covering 10 notable papers spanning models of rehabilitation service delivery, technology and modality-based interventions for stroke and Parkinson's disease, and the lifespan view of cerebral palsy and prestroke functional reserve. Let's dive in.

We'll start with the paper most likely to reshape how rehabilitation services are positioned within a health system, published in The Lancet. Sigera and colleagues ran RESHAP-ED, an open-label pragmatic randomised trial across five emergency departments in New South Wales, Australia, enrolling nearly fifteen hundred adults with uncomplicated musculoskeletal presentations — soft tissue injuries, neck and back pain, and fractures or dislocations not needing orthopaedic review. Participants were randomised to physiotherapist-led care or to usual physician or nurse practitioner-led care. Mean emergency department length of stay fell by a full hour, from about three and a half hours down to just under two and a half. Follow-up for that primary outcome was complete for every single participant. Adverse events were essentially identical between groups, costs were modestly lower in the physiotherapist-led arm, and the intervention was very likely to be cost-effective. This is the kind of trial that gives rehabilitation clinicians hard evidence for scope-of-practice expansion in acute care, and it is worth reading in full if you are involved in service design or business cases for allied health staffing. [1]

Turning to Parkinson's disease, two papers this week point in opposite directions on the value of adding technology to movement practice. In Archives of Physical Medicine and Rehabilitation, Zhang and colleagues randomised fifty older adults with Hoehn and Yahr stage two to three disease and balance impairment to either aquatic Tai Chi or virtual reality gait training in a Computer Assisted Rehabilitation Environment, forty minutes a session, five days a week, for twelve weeks. Both arms improved on motor scores, Berg Balance, Timed Up and Go, six-minute walk distance and gait symmetry, and those gains persisted to week twenty-four. But the virtual reality group did better at the end of treatment on essentially every clinical and spatiotemporal measure, and retained advantages in balance, cadence, gait cycle time and joint symmetry at follow-up. Only forty of the fifty participants completed, so this is a small trial, but it suggests instrumented treadmill and virtual reality training earns its place when available, with aquatic Tai Chi as a reasonable fallback for patients who cannot tolerate it. [2] Set against that, Clinical Rehabilitation published a systematic review by Baumgart and colleagues asking a cleaner question — does adding biofeedback to practice beat the same dose of practice alone? Across twenty trials and more than eight hundred participants, the answer was essentially no. Berg Balance improved by one and a half points out of fifty-six, which is not clinically meaningful; Timed Up and Go and walking speed showed no benefit at all. The evidence was rated high quality for balance and moderate for mobility, so this is not a matter of waiting for better trials. The reconciliation is important for practice: what helped in the virtual reality trial was probably the intensive, task-specific gait practice itself, not the feedback layered on top. Don't pay a premium for biofeedback as an add-on; do pay for practice volume. [8]

Stroke rehabilitation generated three papers, all modest in size and all worth reading with tempered expectations. In the American Journal of Physical Medicine and Rehabilitation, Yıldız and colleagues randomised thirty-five patients with ischaemic stroke to level treadmill walking, or downhill walking at minus seven and a half percent, or minus fifteen percent, over five consecutive daily sessions. The primary electrophysiological endpoint, the H-reflex to M-wave ratio, dropped after a single session only in the steepest group, and after five sessions in both downhill groups on the paretic side, with no change in the level-walking controls. Two-minute walk and Timed Up and Go improved in all three arms, with larger gains at the steepest incline. But the Modified Ashworth and Berg Balance scores did not budge, and the authors themselves caution that with five days of training, thirty-five patients, and no follow-up, the durability and clinical meaning of this remain genuinely uncertain. Treat it as a mechanistic signal that eccentric loading modulates spinal reflex excitability, not as a protocol to adopt. [3] On spasticity modalities, Topics in Stroke Rehabilitation published a three-arm trial from Uslu and Küçükşen in forty-two patients with plantar flexor spasticity, comparing rehabilitation alone against rehabilitation plus radial extracorporeal shock wave therapy or plus high-intensity laser therapy. Modified Ashworth scores improved at three weeks and held at three months in both modality arms but not with rehabilitation alone, and passive ankle dorsiflexion improved only with laser. Here is the crucial caveat: there were no significant between-group differences on any outcome, and functional measures — Fugl-Meyer, ten-metre walk, Berg Balance — improved in every group including rehabilitation alone. Neither modality demonstrated superiority for function. [4] More encouraging is a meta-analysis in Archives of Physical Medicine and Rehabilitation from Yang and colleagues on focal muscle vibration for upper limb spasticity. Pooling eleven independent samples and 285 participants, they found a moderate reduction in spasticity, consistent across the whole thirty to three hundred hertz range studied, with the most uniform effect at the fingers. Certainty was graded moderate, though Egger's testing suggested small-study effects may be inflating the estimate. Practically, if you are using focal vibration, the frequency you choose within that band appears not to matter much. [6]

Finally, three papers push us toward thinking about function across the whole lifespan rather than the episode of care. Also in Archives, Shi and colleagues harmonised four population-based ageing cohorts from China, the United States, England and continental Europe, following nearly seventy-five hundred stroke-free adults who went on to report an incident stroke. They built a multidimensional prestroke rehabilitation reserve score, and each standard deviation lower reserve was associated with roughly a quarter higher risk of having at least two mobility or self-care limitations after the stroke. In absolute terms, adjusted risk rose from about a fifth of patients in the high-reserve group to just over a third in the low-reserve group — a gap of nearly seventeen percentage points, consistent across all four cohorts. The authors are explicit that this describes group-level vulnerability and is not an individual prediction or triage tool, and the cohorts lacked harmonised stroke severity and rehabilitation dose data. Still, it is a strong argument for prehabilitation research in vascular risk populations. [7] In PM and R, Barbuto and colleagues retrospectively studied 139 children with cerebral palsy admitted to an inpatient rehabilitation unit after selective dorsal rhizotomy, single-event multilevel orthopaedic surgery, or intrathecal baclofen pump placement. A small set of variables — pattern of limb involvement, age, aetiology, Gross Motor Function Classification System level, and admission WeeFIM score — explained eighty-six percent of the variance in discharge function, while surgical procedure and admission mobility subscore predicted length of stay. Notably, cognition at admission mediated the relationship between admission mobility and self-care and the functional gains achieved, which argues for cognitive screening as part of setting realistic postoperative goals with families. [9] And rounding this out, a perspective piece in the American Journal of Physical Medicine and Rehabilitation from Strader and colleagues makes the case that cerebral palsy — affecting an estimated fifty million people globally, most of them now adults — should be reframed in physiatry residency training as a lifelong condition with progressive multisystem, metabolic, musculoskeletal and cellular change, despite a nonprogressive index brain injury. Their point that cerebral palsy education remains concentrated in paediatric subspecialties while the patient population ages is one most of us recognise from clinic. [10]

One negative trial deserves a brief mention before we close. Clinical Rehabilitation published a three-arm sham-controlled trial from Ferrari and colleagues in ninety people with knee osteoarthritis, testing twenty minutes daily of elastic bandage knee compression over four days. No between-group difference in pain or physical function reached the minimal clinically important threshold at any timepoint out to twenty-four weeks. Global rating of change favoured compression, which likely reflects the sensation of the bandage rather than a genuine treatment effect. Short-course compression is not superior to sham. [5]

If you only have time for one paper this week, make it the RESHAP-ED trial in The Lancet. It is a rigorously conducted, fully followed-up randomised trial showing that rehabilitation professionals can safely and cost-effectively take primary responsibility for a large slice of emergency musculoskeletal care — evidence with immediate implications for workforce and service planning.

Here are the key takeaways from this week in Physical Medicine and Rehabilitation. First, physiotherapist-led emergency department care for uncomplicated musculoskeletal presentations cut length of stay by an hour with no safety signal and lower costs. Second, in Parkinson's disease the therapeutic ingredient appears to be intensive task-specific practice, not the biofeedback attached to it — high-quality evidence shows added biofeedback confers no meaningful benefit. Third, for post-stroke spasticity, focal muscle vibration has moderate-certainty support for the upper limb with frequency choice apparently not critical, while shock wave and high-intensity laser for plantar flexors reduced tone but did not beat rehabilitation alone on function. Fourth, downhill treadmill walking modulates spinal reflex excitability and may accelerate early walking gains, but the trial was too small and too short to change practice. Fifth, prestroke functional reserve strongly tracks post-stroke disability at the population level, and in paediatric cerebral palsy admission cognition mediates functional gain — both arguments for looking beyond the presenting motor problem. And finally, short-course elastic bandage compression for knee osteoarthritis was no better than sham.

That's your roundup for This Week in Physical Medicine and Rehabilitation. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.