View commentaries on this research

This is a plain English summary of an original research article

Early, intensive rehabilitation aids recovery and improves outcomes for people with moderate to severe traumatic brain injury.

A review of 11 studies found that starting rehabilitation early, while people were still in intensive care, or offering more intensive treatments helped patients with brain injury regain function compared with usual care. Early rehabilitation often included multisensory stimulation while the patient was still in a coma. The intensive multidisciplinary programmes mostly aimed to help people recover, or compensate for, lost skills.

Half the studies showed a high risk of bias, mainly due to lack of randomisation and use of subjective outcomes. However, results were similar after exclusion of these studies, and the effects were moderate to large. So, despite these reservations, the review suggests that early and intensive intervention may be as important in head trauma as it is already recognised to be in stroke.

Why was this study needed?

There were 348,934 UK admissions to hospital with acquired brain injury in 2013-14, a 10% increase since 2005. People with moderate to severe brain injury may be left with a range of disabilities, including difficulties with mobility, employment, education and social interaction, poor cognitive function and reduced quality of life. Some people may need significant long-term social and medical care, which puts a strain on healthcare resources.

After brain injury, the brain can learn to adapt by reorganising its networks of nerve cells, in order to compensate for the injury and recover some of the affected functions.

Some studies have found better results with earlier rehabilitation and more intensive treatments. This review aimed to summarise the available evidence on the timing and intensity of rehabilitation in people with moderate or severe traumatic brain injury.

What did this study do?

This systematic review and meta-analysis included six randomised and one controlled trials, one quasi-randomised controlled trial and four non-randomised controlled trials, with a total of 497 adults. One study was from the UK. The others were from the US, Iran, China, India, Australia and Nordic countries.

Six studies (286 people) assessed the effect of early neurorehabilitation programmes starting in the trauma unit. Five studies (211 people) assessed intensive programmes (more than 20 hours of therapy a week in trials where intensity was specified). All had a usual care comparator group.

The researchers judged five trials to be at high risk of bias. Results were similar when these trials were excluded. There was variation between the trials in the content and duration of the programmes, the outcomes assessed and when they were assessed.

What did it find?

  • Early rehabilitation interventions were delivered in the trauma or intensive care unit. They included four sensory stimulation programmes (three for patients in comas), an intensive multidisciplinary programme (one trial), and a reality orientation programme (one trial).
    Interventions started after the patient’s bleeding was stabilised, or between one and a median of 12 days after the injury or admission, and lasted between six days and two weeks where stated. Interventions were delivered by a multidisciplinary team, nurses, therapists, or the patient’s family. Usual care was only described for one study, where it involved physical therapy.
    Usual care is likely to have differed from the early interventions in content as well as timing.
    Outcomes were assessed at varying time points, ranging from daily during the intervention to 12 months post-injury.
  • Early rehabilitation had a large positive effect on functional outcomes compared with usual care (effect size [Cohen’s d] = 1.02, 95% confidence interval [CI] 0.56 to 1.47; 6 trials, 286 patients). The outcomes assessed included cognitive functioning and functioning in movement, sensation, language and physical abilities. There was a high level of variability in these pooled results (heterogeneity I2 = 68.2%).
  • The intensive programmes mostly included intensified versions of usual care rehabilitation programmes, delivered at inpatient rehabilitation units or on an outpatient basis. All were delivered by multidisciplinary teams.
    The point at which these programmes were initiated post-injury was not reported. Two studies did not quantify therapy time, but three studies stated intensive programmes entailed four to five therapy hours a day, four to five days a week. Interventions lasted between 4 and 16 weeks where stated.
    Usual care included up to 15 therapy hours per week where stated.
    Outcomes were assessed at varying time points, from the end of the intervention to up to two years post-injury.
  • Intensive neurorehabilitation programmes had a medium sized positive effect on functional outcomes compared with usual care (Cohen’s d = 0.67, 95% CI 0.38 to 0.97; five trials, 211 patients).The outcomes assessed included functional independence, self-care, movement, communication, cognition and work status.

What does current guidance say on this issue?

NICE quality standards on head injury say: “People who are in hospital with new cognitive, communicative, emotional, behavioural or physical difficulties that continue 72 hours after a traumatic brain injury [should] have an assessment for inpatient rehabilitation.”

NHS England (formerly the NHS Commissioning Board) published service specifications for the specialist rehabilitation of patients with highly complex needs. The specifications state that there is "strong evidence" that rehabilitation in specialist settings for people with traumatic brain injury is effective and provides value for money, and that "early transfer to specialist centres and more intense rehabilitation programmes" are cost-effective.

What are the implications?

This review supports the introduction of specialist neurorehabilitation services to provide intensive rehabilitation. It also suggests that an early start to rehabilitation, even before a patient regains consciousness, is likely to be beneficial. Some early multisensory stimulation could potentially be delivered by the patient's family members if given appropriate instruction.

The findings strengthen evidence for NHS England’s service specifications and support a stroke-style service for brain injury, where patients with serious head trauma are taken directly to the closest specialist centre, rather than to the nearest accident and emergency department. The evidence could inform future guidance on the development of traumatic brain injury standards and services.

Citation and Funding

Königs M, Beurskens EA, Snoep L, et al. The effects of timing and intensity of neurorehabilitation on functional outcome after traumatic brain Injury: a systematic review & meta-analysis. Arch Phys Med Rehabil. 2018;99(6):1149-59.

This study was funded by grants from the Daan Theeuwes Foundation and the Netherlands Organisation for Scientific Research.



NICE. Head injury. QS74. London: National Institute for Health and Care Excellence; 2014.

NHS England. NHS standard contract for specialised rehabilitation for patients with highly complex needs (all ages). London: NHS England; 2013.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


  • Share via:
  • Print article


Moderate to severe traumatic brain injury. Defined by the review as Glasgow Coma Scale score of 12 or less, being unconscious for 30 minutes or longer, and/or having post-trauma amnesia lasting an hour or more.Multidisciplinary teams. These teams can include, for example, neurosurgeons, rehabilitation physicians, physiotherapists, occupational therapists, speech therapists, nurses, medical social workers and psychotherapists.Cohen’s D. Is a measure of the size of an effect. It standardises the difference between two means, and therefore allows comparisons where different continuous scales are used. Cohen offered guidelines for interpreting the thresholds in social science:
  • small effect = 0.2
  • medium effect = 0.5
  • large effect = 0.8 or more


Expert commentary

Rehabilitation after traumatic brain injury improves outcomes. However, questions remain over the timing and dose of rehabilitation. There is compelling evidence from animal studies that commencing intensive rehabilitation in the earliest phase of injury when the brain appears to be uniquely sensitive to adaptation, the so-called ‘plasticity window', leads to much better outcomes.This study tests whether the theory holds true in humans. The researchers examined brain injury following trauma, which accounts for around half of all acquired brain injuries, and found clear evidence that commencing rehabilitation very early and in high doses significantly improves outcome.Professor Alan Carson, Consultant Neuropsychiatrist & Major Trauma Rehabilitation Lead, NHS Lothian
Back to top