3.2 Inpatient Rehabilitation

While many ABI victims are discharged directly home or to a long term care facility, many will benefit from discharge to a dedicated inpatient rehabilitation service. These services vary from institution to institution but generally include some type of intensive therapy program for physical, social, behavioral and cognitive difficulties. However, deciding who should receive inpatient rehabilitation remains a major challenge. Patient referral decisions are inherently complex and need to be understood as a dynamic phenomenon shaped by characteristics of the individual. However, they also rely on the interactions and interpretations of health professionals who operate within unique organizational and broader health care contexts 52. These discrepancies are confounded by social and funding issues. For example, in the US patients insured by Medicaid or an HMO were more likely to go to a skilled nursing facility rather than inpatient rehabilitation relative to people with commercial fee-for-service plans 31.  In Canada, patients aged 36 – 45 with more co-morbid conditions are more likely to end up in rehabilitation than those older than 65, rural dwellers, non-English speaking people and people with mental health, alcohol and/or drug problems 53. The diversity of patient needs has also led to the formation of differing systems of rehabilitation. In Calgary, for instance, the Halvar Johnson Centre for brain injured patients has established a program to treat TBI and non-TBI patients in a slow stream rehabilitation program for individuals who may require slightly extended care. According to Cullen 54rehabilitation in Canada, on average, discharges 80% of patients home.

Due to the unique challenges posed by ABI, the structure of inpatient rehabilitation is extremely diverse. Patients are generally rehabilitated in one of two centers; a general rehabilitation unit or a coordinated multidisciplinary neurorehabilitation unit. Some argue that an effective rehabilitation service requires a multidisciplinary team, which includes nursing care, physician monitoring, psychologist and social work intervention, physiotherapists, occupational therapists, and speech language pathologists among other things 55. In reality, differences in care often amount simply to the availability of neuro-rehabilitative beds and facilities. Limited resources mandate decisions regarding which patients will most benefit from inpatient rehabilitation compared to community-based programs.

Debate also exists about appropriate targets of rehabilitative care. Traditional rehabilitation models in other disciplines such as stroke, spinal cord, and polio have focused on orthopedic and neuromotor impairments 12. Brain Injury rehabilitation initially followed a similar path until focus on cognitive and behavioral remediation 56as well as coma stimulation 12gained recognition. The greater emphasis on skill development in rehabilitation has not resolved the uncertainty regarding which patient groups are best suited to inpatient care versus community-based programs. Patients in need of skill application training are increasingly being discharged to community based services while inpatient rehabilitation has focused more on intensive, short term physical or cognitive rehabilitation 57.  Furthermore, some inpatient facilities are recognizing the need to divide patients into different streams during rehabilitation. At the Toronto Rehabilitation Institute, patients have been streamed into a Neurocognitive group and a Neurophysical group since 2002 54. Patients in the Neurophysical stream showed similar FIM gains in a significantly shorter length of stay when compared to similar patients before streaming began.

Inpatient rehabilitation typically begins when a patient is medically stable enough to be transferred out of acute care and into a dedicated rehabilitation unit for a defined period of interdisciplinary rehabilitation. There is a great deal of variability in the length, type, and intensity of services provided in programs throughout the world. As such, we delineate the evidence supporting the various aspects of treatment for inpatient care delivery.

What evidence is there that inpatient rehabilitation improves the outcomes of ABI patients?

  1. There is no Level 1 evidence (from at least one RCT) as to the efficacy or lack thereof of ABI rehabilitation units.
  2. There is Level 3 evidence that over a quarter of patients admitted to inpatient rehabilitation experience good outcome or moderate disability six months post-injury as measured by the GOS
  3. There is Level 4 evidence that inpatient rehabilitation significantly improves functional outcome, as measured by the FIM.

Sahgal and Heinemann 58conducted a pre-post study on 189 patients with TBI admitted to a National Institute on Disability and Rehabilitation Research-Designated Center in the USA. Using a locally developed functional rating scale as the main outcome measure, the authors noted improvements in the patients for self-care and mobility after discharge from the comprehensive multidisciplinary program.

Two case series evaluated patients’ functional outcome after discharge from inpatient rehabilitation. Both used the Functional Independence Measure (FIM) as one of their main outcome measures and both noted significant improvements for patients on FIM measurement 59,60.

Is there any evidence that readmitting a patient over one year after suffering an acquired brain injury would result in functional outcomes?

  1. There is Level 2 evidence that readmission to inpatient rehabilitation at more than 12 months post-injury is related to statistically significant improvement at discharge for over 50% of patients.

Two other case series assessed functional outcome after inpatient rehabilitation using the Glasgow Outcome Scale (GOS) and Barthel Index (BI) respectively.  In the former, 35% of subjects experienced good outcome or moderate disability at six months post-injury, as measured by GOS scores 61.  In the latter, 53% of patients readmitted to inpatient rehabilitation at more than twelve months post-injury showed statistically significant improvement (p = 0.0001) on BI scores from readmission to discharge 62.

3.2.1 Intensity of Inpatient Rehabilitation

While patients are undergoing rehabilitation the amount of therapy provided to them is potentially an important factor in promoting neurological and functional recovery.  We review the evidence for increased intensity in this section.

What evidence is there that increasing rehabilitation intensity influences outcomes?

  1. Based on the finding from a single RCT, there is Level 1 evidence that increasing rehabilitation intensity reduces length of stay.
  2. There is Level 4 evidence that patients with a long length of stay who receive high-intensity rehabilitation fair better on the Rancho Scale at discharge than those who receive low-intensity rehabilitation.
  3. Based on the findings from a single RCT, there is Level 1 evidence that intensive rehabilitation improves functional outcome as measured by FIM and GOS scores, at two and three months post-injury, but not necessarily at six month and beyond.
  4. There is Level 2 evidence that therapy intensity predicts motor functioning but not cognitive gain.
  5. There is a reciprocal relationship between cognitive function and community integration.
  6. There is Level 4 evidence that earlier time from injury onset to rehabilitation admission results in improved functional outcomes.

Two RCTs focused on inpatient rehabilitation of ABI patients 63;64. Both studies assessed the effects of increasing therapy intensity levels.  In the Shiel et al. 64study, patients in the intervention group received additional therapy from a health care professional (a rehabilitation nurse at one centre and an occupational therapist at the other) who provided these extra services as necessary. Shiel et al. 64found that patients showed improvements on discharge both the FIM+FAM measures; however these improvements may be related to the size of the rehabilitation facility and the amount of staffing available to the patients. The study authors noted that patients in the larger facility received more intensive therapy over a shorter period of time and saw significant gains.  n contrast, patients in the intervention group at the smaller center actually experienced a longer length of stay than their control counterparts.

In the second RCT, conducted Zhu et al. 65, subjects were randomly assigned to either four hours (study group) or two hours (control group) of rehabilitation per day. Functional outcome was determined by monthly Glasgow Outcome Scale (GOS) and Functional Independence Measure (FIM) scores. The authors found that more subjects in the study group than in the control group achieved full FIM scores and good GOS scores at two and three months post-injury; however at the sixth month time period, despite initial improvements obtained in rehabilitation, the control group had made significant gains and were seen to be “catching up” 65. Both studies noted a trend towards improvements in functional gains with increased intensity but recommend future study into more long term effects.

Two studies examined the efficacy of the intensity of rehabilitation in relation to the length of hospital stay 64;66. In both studies, one a prospective RCT and the other a case series, increased rehabilitation intensity resulted in decreased length of stay.


Spivack et al. 67, conducted at study looking at the combined effects of rehabilitation intensity and inpatient rehabilitation length of stay. In their comparison of patients who had a long length of stay and received either low-intensity or high-intensity rehabilitation, the latter group fared better on the Rancho Scale outcome measure at discharge.

Semlyen et al. 68compared coordinated multidisciplinary inpatient rehabilitation to single discipline therapy provided in a local district hospital. Patients treated in the multidisciplinary hospital showed greater improvement in Barthel, FIM and Newcastle Independence Assessment Form scores and maintained improvement at 24 months. However, the authors point out some methodological concerns. Patients were non-randomly divided between the two groups which resulted in less severe injuries in the single discipline group as well as shorter LOS. This may have resulted in a ceiling effect for these patients that could have hindered their recovery gains.

A multicenter, prospective, nonrandomized study also assessed the relationship between therapy intensity and functional outcome.  While rehabilitation intensity was found to predict motor functioning at discharge (p<.001), it did not predict cognitive gain (p<.05) 55. Intensive and structured cognitive rehabilitation therapy (group and individual) has been reported to cause significant improvements in client reported satisfaction when compared to standard multidisciplinary rehabilitation 69.  The intensive rehabilitation program participants showed significant effects on their cognitive functioning as demonstrated on their improvement on standard neuropsychological tests69.

In all of the studies identified, trends towards improved function after multidisciplinary inpatient rehabilitation were seen. Several study authors noted that they saw no ceiling effect associated with increased intensity of therapy. However, all of the authors indicated concerns about outcome measurement tools. There seems to be consensus regarding the need for a more accurate, ABI specific measure of functionality.

Zhu XL, Poon WS, Chan CH and Chan SH. Does intensive rehabilitation improve the functional outcome of patients with traumatic brain injury? Interim result of a randomized controlled trail. British Journal of Neurosurgery 2001: 15(6):464-473.

  • 36 individuals were randomly assigned to one of two groups (intensive treatment group or conventional treatment group.
  • The intensive treatment group received 4hours of therapy 5 days a week while the conventional treatment group received 2hours.
  • Those in the intensive treatment group achieved significantly better outcomes (40 vs 10, p=0.046), at the 2 month evaluation period than the conventional treatment group.
  • At the 6th month time this difference was reduced.
  • Significant differences were not found when looking at the results of the FIM.


Intuitively, it seems reasonable to assume that more therapy will result in more rapid and ultimately greater improvement in recovery from brain injury. Based on the available literature, greater intensity appears to result in quicker recovery and therefore shorter lengths of stay, but not necessarily better outcomes at six months.  More studies are needed in this regard.

3.2.2 Timing of Rehabilitation

List some of the benefits of early admission of an ABI patient to rehabilitation?

  1. Better outcomes overall
  2. Improved functional outcomes
  3. Shorter overall lengths of stay
  4. Decreased overall costs
  5. Higher cognitive levels at home
  6. Greater likelihood of discharge home

Sandhaug et al. 70looked at the benefits of having individuals who had sustained either a moderate or severe TBI participate in a sub-acute rehabilitation program. On average patients were transferred to a specialized rehab program 27 days after being admitted to sub acute rehab. Those with a severe TBI remained in rehab longer than those with moderate injuries and were discharged either to a rehab hospital or nursing homes for further treatment. FIM scores improved significantly (p<0.001) for all regardless of the level of injury. Patients diagnosed with a severe TBI showed a significantly (p<0.001) greater improvement in their overall FIM scores with, the greatest improvement, for both groups, being seen in the motor scores.  Improvement on the FIM-COG subscale was also noted but in both groups the score improved by only 5 points. The authors suggest that the FIM score at admission to rehabilitation, together with the GCS and PTA, were positive predictors of functional level at discharge 70.

Wagner et al. 71examined the proper timing for physical medicine and rehabilitation consultation. Using multivariate analysis, the authors found that when PM&R consultations occurred earlier (< 48 hours after hospital admission) patients experienced significantly better FIM scores with transfers and locomotion and significantly shorter lengths of stay (p = 0.001). 

In the other outcome study, Edwards et al. 72compared 26 patients admitted to inpatient rehabilitation more than 200 days after injury to 264 patients admitted to inpatient rehabilitation less than 200 days after injury.  Discharge BI and FIM scores were lower in the former group than in the latter (11 vs. 14 and 77 vs. 92 respectively). However, the differences were not significant.  Rehabilitation length of stay was also similar for the two groups. 

Mackay et al. 73assessed the timing of inpatient rehabilitation during the earlier phase of recovery in their cohort study. They compared a formalized program (average of 2 days to initiation of therapy) with a non-formalized program (average of 23 days to initiation of therapy) using co-relational analysis. Number of days in coma, length of stay, cognitive levels, and discharge disposition were used as the main outcome measures. Overall, starting rehabilitation early was associated with shorter comas and lengths of stay, higher cognitive levels at discharge, and a greater likelihood of being discharged to home. High et al. 74in a study of TBI patients the authors examined the amount of time that lapsed from diagnosis of injury to the start of rehabilitation and its effect on outcomes of rehabilitation. They found that those who began treatment within six months of their TBI scored higher on the disability rating scale indicating a decrease in their disability. These results were not noted for the other two groups. The supervision rating scale scores decreased for all groups indicating that they required less supervision after admission to rehabilitation and when tested again at follow up post discharge, again a decrease in supervision was noted. When analyzing the results of the community integration questionnaire an increase in scores could be seen from admission to discharge from the program for all groups.

Mackay et al. 73and Cope and Hall 75found that those who were involved in rehabilitation earlier in the recovery stage were discharged from hospital earlier than those who were not involved in the early rehabilitation program.  Aronow 76, found that although there was no statistically significant differences on the individual outcomes there was a cost savings favoring those who were subjected to early interventions.

Tepas et al. 77retrospectively reviewed patient charts to evaluate the effect of delays for admission to rehabilitation on functional outcomes. They report that delays in admission to rehabilitation resulted in significant decreases in total FIM gains as well as reductions in rehabilitation efficiency. These findings were similar to those reported by Kunik et al. 78.  In this study individuals admitted sooner into rehabilitation (<1 week to 3 weeks post insult) were admitted with higher FIM scores (~59.8) than those admitted later (4 week or more post insult -FIM scores (48.29)). Overall those admitted sooner to rehabilitation were released on average 19 days post admission. Those admitted later to rehabilitation were released on average 26 days post admission. Kunik et al. 78suggest that those who are admitted into rehabilitation sooner after injury perform better and faster and their overall cost of stay is less.

In the absence of an RCT looking at the role of early rehabilitation, what is the primary complicating factor in linking early admission to better outcomes?

  1. The primary complicating factor is the tendency to delay the admission of more complicated or more severely involved patients who are not likely to do as well in rehabilitation as those without such complications.

The studies available on the timing of rehabilitation demonstrate that earlier rehabilitation is associated with better outcomes than later rehabilitation. Patients who have had recent brain injuries typically need much greater medical and nursing support in order to meet their basic care requirements. This evidence is consistent with theories of neuronal plasticity, which suggest that challenging the nervous system by means of therapy results in increased neuronal compensation and/or regeneration. However, delayed rehabilitation may reflect more severe or complicated brain injuries. There is an obvious need for an RCT to address this question.

3.2.3 Factors Affecting the Timing of Inpatient Care Etiology and Inpatient Rehabilitation

What evidence is there for the efficacy of inpatient brain injury rehabilitation in different types of acquired brain injured individuals?

  1. There is Level 3 evidence that inpatient brain injury rehabilitation results in significantly greater gains in total FIM change, self-care, and social cognition for patients with TBI than patients with brain tumors.  However, there are no statistically significant differences between the two groups regarding FIM efficiency and length of stay.


In a retrospective, descriptive, case-matched study by O’Dell et al. 79, forty patients with brain tumors were compared with 40 patients with TBI.  They all underwent inpatient rehabilitation on a freestanding brain injury unit.  Change in FIM scores, length of stay, and discharge disposition were used as the main outcome measures.  Overall, the TBI patients made significantly greater gains in total FIM change (34.6 vs. 25.4), self-care (12.3 vs. 8.5), and social cognition (5.2 vs. 3.6).  However, there were no statistically significant differences between the two groups regarding FIM efficiency (1.9 vs. 1.5 FIM points per day) and length of stay (22.1 vs. 17.8 days). See table 3.6 for details. Age and Inpatient Rehabilitation

What is the impact of age on the outcomes of acquired brain injuries?

  1. There is Level 3 evidence that inpatient rehabilitation results in a higher rate of change on functional measures in patients aged 18-54 than patents over the age of 55.

In Cifu et al. 80DRS, FIM, and RLAS scores were compared at inpatient rehabilitation discharge for 50 patients greater than or equal to 55 years of age and 50 patients aged 18 to 54.  In this case-control study, subjects in the latter group showed a higher mean rate of change on functional measures than subjects in the former group. Occupation and Inpatient Rehabilitation

What evidence is there to suggest that inpatient rehabilitation result in a successful return to work?

  1. There is Level 4 evidence, based on the findings of one case series, that inpatient rehabilitation results in successful return to work and return to duty for the majority of military service members.

In describing only one treatment arm of a RCT, Braverman et al. 81evaluated military service members’ return to work and return to duty after multidisciplinary inpatient rehabilitation.  Multidisciplinary inpatient rehabilitation consisted of eight weeks of group and individual therapies geared towards returning the soldiers to duty. The rehabilitation team included a physiatrist, neurologist, neuropsychologist, and occupational therapist. The authors found that of the sixty-seven subjects who participated in the study, 96% and 66% had returned to work and duty respectively at follow-up of one year. Transitional Living Setting and Inpatient Rehabilitation

What evidence is there for benefit of a transitional living setting at the end of inpatient rehabilitation?

  1. There is Level 2 evidence that a transitional living setting during the last weeks of inpatient rehabilitation results in greater independence in activities of daily living than inpatient rehabilitation alone.

In the study by McLaughlin and Peters 82the effects of a transitional living setting during the last weeks of inpatient length of stay were evaluated using cognitive (Rancho) and functional (Barthel) levels as main outcome measures. Results from a follow-up survey showed that patients who participated in both inpatient rehabilitation and a transitional living setting reported greater independence in activities of daily living than patients who received inpatient rehabilitation alone.  


Previous reviews in addition to literature presented here have delineated the extent of knowledge of the efficacy of inpatient rehabilitation, which is limited at best. There is reasonable evidence to support the use of interdisciplinary rehabilitation followed by a transitional living environment to assist in maximizing recovery. This suggests that a gradual return to the community is preferable to a sudden discharge from hospital to home. Not surprisingly, younger patients tend to make greater gains in rehabilitation than their older counterparts.