The Chartered Society of Physiotherapy The Chartered Society of Physiotherapy


View your shopping cart.

Stroke rehabilitation

Spotlight on research: stroke rehabilitation

Find out more about research into stroke rehabilitation.

Anatomy of a study

Constraint induced movement therapy is an innovative approach in stroke rehabilitation. It is based on the theory of ‘learned non use’, elucidated by American scientist Edward Taub more than 20 years ago. Basically, when patients first try to use their stroke affected limb and fail, they stop there instead of trying again.


Research: The EXCITE trial (extremity constraint-induced movement therapy evaluation).

Lead investigator: Steven Wolf, departments of rehabilitation medicine and medicine, Emory School of Medicine, Atlanta, Georgia. Professor Wolf has published more than 170 research papers. He also teaches and spends about a fifth of his time in clinical practice.


Publication: Wolf SL, Winstein CJ et al. ‘ Retention of upper-limb function in stroke survivors who have received constraint-induced movement therapy: the EXCITE randomised trial.’ Lancet Neurology 2008; 7: 33-40.

CIMT aims to overcome this pattern by forcing use of the affected side while restraining the unaffected limb, for example in a sling. The patient is trained and encouraged to use the arm repetitively and intensively, generally for a period of two or three weeks.

The EXCITE trial (extremity constraint-induced movement therapy evaluation) wanted to know whether a two-week course of CIMT would improve outcomes for recovering stroke patients and whether that improvement would be maintained longer term. This year they published research in Lancet Neurology showing for the first time that CIMT has significant long-term benefits for stroke patients: improvements seen in an earlier trial were retained at 24 months.

The team’s research focused on an approach to CIMT that is more intensive than other variations defined in the literature and found in clinical practice. In EXCITE, patients rehearsed functionally relevant and repetitive tasks, including shaping procedures, with the paretic limb for up to six hours every weekday. The entire course of therapy lasted for two weeks. The less affected hand and wrist were restrained during most waking hours.

With other types of forced-use therapy, including modifications of CIMT, not studied in EXCITE, patients might have no formal training, less support from the therapist, or a longer overall treatment period of up to 10 weeks.

The primary outcome measure was impairment of upper limb function. Two instruments were used to assess this endpoint: the Wolf motor function test and the motor activity log.

As a secondary outcome measure, health-related quality of life was assessed using the stroke impact survey, a structured interview covering five domains: strength; memory and thinking; activities of daily living; social participation; and overall physical function. For the upper limb function tests, improvements seen at 12 months were not eroded at 24 months.

Moreover, the average change was in the direction of an additional treatment effect. Significant further gains were seen in the strength components of the WMFT. All of the quality of life domains showed continued improvement. All patients doubled their grip strength, while those higher functioning patients showed further gains in overall arm strength.

Patients in this study showed substantial improvement in both functional use of the upper paretic limb and in quality of life, two years after the initial intervention. Thus, the authors concluded, CIMT has persistent benefits.

What does it tell us?

The key message from this large high-quality study is that constraint-induced movement therapy significantly improves the arm function and quality of life of stroke patients. Other studies have shown CIMT is effective, but this is the first to show the benefits are maintained and increase further over two years.

Patients received intensive therapy, six hours a day, five days a week for two weeks in an outpatient setting. It may not be feasible to deliver CIMT in exactly the same way, but it can be modified and applied in a community setting, as contributors to this article show.

A review in Physiotherapy (see below: Further information) also identifies factors relevant to implementing CIMT in clinical practice, including group versus individual therapy.

Patients' values, beliefs and preferences should be at the centre of rehabilitation services; and indeed, most funders now recommend patient involvement in all stages of research. It’s interesting to note that many patients don’t like the idea of CIMT. However, patient and carer behavioural contracts promoted compliance in this study – a tip for putting research into practice. Patients must be informed of benefits and risks.

Research to evaluate the cost effectiveness of CIMT is urgently needed. A good piece of research generates further important questions and can justify further funding. Lead researcher Steven Wolf comments that his study has been a springboard for more than 30 additional papers, and the seven centres in his trial own their data and continue to undertake further related research. Look out for more of their work.

CIMT is mentioned in the English national stroke strategy, and the evidence for CIMT has been collated and weighed up in the Royal College of Physicians stroke guidelines, which make a specific recommendation for clinicians on what to do. Physiotherapists are expected to remain up to date with new interventions and clinicians need to consider how to best implement the evidence in their practice.

Gabrielle Rankin, CSP research adviser

Reflections on research

The EXCITE trial has been a springboard for more than 30 additional publications and the topics are diverse. Carer quality of life, neurological mechanisms in stroke recovery, changes in the biomechanics of the upper extremities: these are just a few.

Such follow-ups are flourishing because participants at all seven EXCITE centres have real ownership of the data and the freedom to choose additional new paths of inquiry, says lead investigator Steven Wolf.

This open and enthusiastic attitude has been part of the EXCITE trial from the start. It all began in 1996 when Professor Wolf and behavioural neuroscientist Edward Taub made a presentation on the scientific basis of CIMT at the American Physical Therapy Association’s annual meeting.

‘We had no idea how many people would be interested,’ Prof Wolf remembers. ‘There were 500 or 600 people and we couldn’t fit everyone in the room. Over lunch it was suggested we do a clinical trial.’

He persuaded APTA to provide a $6,000 grant to bring potential collaborators together in Atlanta for a planning session. That seed money was eventually parlayed into a $7.5 million research grant.

It was not a quick or easy road, however. ‘It took 10 years to get from that meeting to a publication in JAMA,’ Prof Wolf says. The journey is not over yet.

In particular, he says, two big ideas emerging from EXCITE beg to be explored next. ‘Can we do something in the acute phase and can the intervention be more specific?’

To answer those questions, another trial is gearing up for launch. Joining Prof Wolf as principal investigators are physiotherapist Carolee Winstein at the University of Southern California and Alexander Dromerick of Washington University in St Louis, Missouri. The new study is called the interdisciplinary comprehensive arm rehabilitation evaluation trial, or ICARE stroke initiative. ‘These days it helps to have an acronym,’ Prof Wolf quips.

It also helps to be at the top of your field, he adds. The US government funds medical research on a huge scale but the competition is fierce.

However, even with his numerous research and teaching commitments, Prof Wolf still makes working with patients a priority. And this, in turn, is where he finds the direction for his research. ‘The ideas,’ he says, ‘come from the patients.’

In the clinic

Martine Nadler, principal therapist at the Wolfson rehabilitation centre, London, has thoroughly reviewed the literature on CIMT as part of her contribution to updating the guidelines for treatment of stroke from the Royal College of Physicians. As a practising physiotherapist, she can also see the evidence in the light of practical realities in the NHS.

Patients in the EXCITE trial had up to six hours of therapy every weekday, for a period of two weeks. Those in the control group had ‘customary care’ as delivered in the US, which included many and varied interventions involving much less therapist time and input.

‘What I’d really like to see is research comparing CIMT with a similar dose of conventional physiotherapy,’ Ms Nadler says.

‘Back in 2001 van der Lee [see: Further information] raised the question: Is CIMT something different or more of the same? This question remains to be fully answered.’

Providing this amount of therapist time in the NHS is unrealistic but that doesn’t mean CIMT is not an option, she adds.

Patients might still benefit from modified CIMT – for example using a restraint within a treatment session while focusing on upper limb treatment and short periods of task practice with carer/family support out of sessions.

Only a small proportion of patients fulfil the inclusion criteria (wrist and finger extension of stroke hand) and many patients do not like the idea of CIMT, she says.

They fear, perhaps rightly, that restricting use of their good arm will make them more dependent in daily life. They might be more likely to agree to a less intensive regimen with shorter spells of constraint.

Appraising the evidence

Whatever hat she is wearing – whether she is acting as a clinician, manager or adviser – consultant physiotherapist in stroke Rhoda Allison is keen to see physiotherapy practice placed firmly on an ever-stronger evidence base.

This means physiotherapists need to keep up to date with research in their field, she says. At the very least they will be one step ahead of their patients, who commonly ask about the latest news from the internet.

However, staying informed is only the beginning. All clinical research, even articles published in the top journals, must be viewed with a questioning eye. When approaching any particular piece of research, ‘physiotherapists need to be able to critically appraise it,’ Ms Allison says. They also need the ability to determine how effective an intervention is, and which patients are likely to benefit. In the rehabilitation field, this task can be especially difficult. Many of the studies have small numbers of subjects, so readers need to look for proof of statistical significance and ask what proportion of their patients could benefit in actual practice.

Rehabilitation is also complex because so many interventions are offered at once, and it can be difficult to tease out a single variable. As one example, she says, ‘we know that specialist stroke units are effective but we don’t know why.’

Finally, she warns about getting carried away with any one piece of research. Scientific inquiry builds up on itself over time: ‘There’s a lot to be gained from looking at more than one study at a time to build up a picture of what’s most effective.’

The occupational therapist’s view

"What I'd really like to see is research comparing CIMT with a similar dose of conventional physiotherapy"
Martine Nadle

Anna Sharland is a senior occupational therapist in a community rehabilitation team in Bournemouth and Poole primary care trust. Her team has been so impressed by the evidence for CIMT, they are evaluating the intervention in a pilot trial of their own.

‘We looked at the research and decided that based on the evidence we should attempt to implement it within our service,’ Ms Sharland says. All of the data they could find was from hospital- based studies, so they have built an audit into their project from start to finish to determine how well the approach works in a community setting.

Because CIMT puts an intensive requirement on therapists’ time, the team also needs objective data to demonstrate the investment is worthwhile. Ms Sharland’s team plans to evaluate CIMT in 10 patients initially. The intervention includes six hours of CIMT in the patient’s home, every weekday for a period of three weeks. Therapists visit daily to provide treatment, maintain motivation, assess results and ensure development of normal movement patterns.

Performance on a nine-hole peg test (which is timed) is recorded on video at the start of treatment, following three weeks of CIMT and six weeks after it is completed. Ms Sharland believes this step is crucial: ‘There’s not much research that has looked at the quality of movements, which is why we decided to video this activity.’ Seeing the ‘before’ and ‘after’ videos also helps patients appreciate how far they’ve come in a short time. The team also videos a functional activity, using the same time frames, which the patient has identified as wanting to achieve.

Other outcome measures include measures of grip strength, hand and finger extension and a patient self-rating score on the usefulness of their arm. These are carried out on a weekly basis to show the therapists and the patient how well the treatment is working.

All of the data are recorded on a two-sided form, which can be retrieved easily when the time comes to audit the results. Ms Sharland is encouraged that the EXCITE data demonstrates sustained improvement over time. The study has also inspired her to consider following up her own patients for a longer period.

Research into practice

  • If you would like to put the EXCITE findings to work for your own patients but aren’t quite sure about the practicalities, these tips should help
  • make a video of the patient performing a specified task before and after treatment. This gives you objective evidence, motivates everyone involved, and is useful for training other professionals
  • set up a good screening tool and find a way of documenting all your outcome measures
  • it’s critical to have a behavioural contract with the patient. Make sure they understand the benefits, the challenges and what’s required. Are they willing to give it a serious try?
  • the patient must play a proactive role in designing the treatment plan. Ask them to name 10 tasks they would like to practise with their stroke hand
  • don’t worry if you can’t do six hours a day. Try a modified form of CIMT by incorporating restraint and task practice within a treatment session or as part of a home exercise programme
  • aim for a dynamic approach during treatment sessions. Patients can work effectively on their arm and hand function when standing and walking rather than sitting down. Use the environment and ask the patient to walk around the room searching for tasks to perform with their stroke hand, such as carrying a tray, turning off a light switch, turning on taps or bending down to get something out of a fridge

Further information

  • Tuke A. ‘Constraint-induced movement therapy: a narrative review’, Physiotherapy (2008), 94: 105-114
  • Wolf S et al. ‘Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial’, Journal of the American Medical Association (2006), 296: 2094-2014
  • Underwood J et al. ‘Pain, fatigue, and intensity of practice in people with stroke who are receiving constraint-induced movement therapy’, Physical Therapy (2006), 86: 1241-1250
  • Van der Lee J. ‘Constraint-induced therapy for stroke: more of the same or something completely different?’, Current Opinion in Neurology (2001), 14(6):741-744
  • The UK stroke research network
  • National stroke strategy, DH (2007)
  • National clinical guidelines for stroke, Royal College of Physicians, 2004.
  • Management of patients with stroke, Scottish Intercollegiate Guidelines Network, 2002


More from the CSP

Back to top