Stroke patients do not need to be inactive in the first two-weeks after stroke: results from a stroke unit focused on early rehabilitation

Stroke patients do not need to be inactive in the firsttwo-weeks after stroke: results from a stroke unitfocused on early rehabilitation

Torunn Askim1*, Julie Bernhardt2,3, Anne Dahle Løge4,5, and Bent Indredavik1,5

Background Although stroke unit care with early rehabilita-tion is recommended for most stroke patients, the content ofthe treatment given to acute stroke patients differs a lot.Aim The primary aim of this study was to describe the activ-ity patterns of acute stroke patients managed in a strokeunit focused on early rehabilitation.Methods Patients admitted to the stroke unit at St. OlavsHospital, Trondheim, with the diagnosis of stroke wereasked to participate. Patients were eligible if they were lessthan 14 days post stroke and did not receive palliative care.Patients were observed every 10 mins from 8:00 am to 5:00pm, and activity, location, and who helped them wererecorded.Results A total of 117 patients (mean age 78·7 years; 56%male) were included. In total, patients spent 30·3% of time inbed, 46·4% of time in sitting out of bed, and 19·9% of timein higher motor activities such as transferring, standing,walking, or climbing stairs. Patients with mild, moderate,and severe stroke spent 79·5%, 59·2%, and 31·0% ofobserved time sitting out of bed or engaged in higher motoractivities, respectively.Conclusion This study shows that it is possible for acutestroke patients to spend most of the active day out of bedand to engage in higher motor activities up to 20% of the

time. It also shows that it is feasible for severely affectedstroke patients to stay out of bed in the early phase afterstroke.

Key words: acute stroke, enriched environment,multidisciplinary care, physical activity, stroke unit treatment

Introduction

Stroke unit care is recommended as the first choice of care forall patients suffering from acute stroke. Based on several ran-domized controlled trials, an effective stroke unit seems tocomprise several elements, such as a systematic approach toassessment and diagnosis, early and active management by amultidisciplinary team, including goal setting and dischargeplanning (1). There is also growing evidence for the benefit ofvery early mobilization, i.e. starting mobilization out of bedwithin 24 h after onset of stroke. This may be an importantcomponent of acute stroke unit care (1–3), although largertrials are needed before reliable evidence of benefit or harmcan be determined.

It is not only time to commencement of mobilization that isof interest. The amount of motor activity during the earlyphase after stroke is also likely to be important (1,4). Recentobservational studies have shown that the amount of motoractivity of acute stroke patients treated in different Europeanand Australian acute stroke units who do not focus on earlyrehabilitation varies. Patients managed in these units wereoften ‘inactive and alone’ (5,6). For this reason, the amount ofphysical activity of patients treated in stroke units focused onearly rehabilitation should be of special interest to explore.

Comprehensive stroke unit care has been described as offer-ing an ‘enriched environment’, and it is hypothesized that thismay explain the good outcome achieved by patients managedin these units compared to general ward care (2). In animalstroke model studies, enriched environments have beenshown to improve functional and cognitive outcome (7,8).Animals housed in an enriched environment have larger cages,social housing (more than one animal), with opportunity forvaried activity including exercise; this contrasts with standardhousing which has single animals in small, rather empty cages(7). A stroke unit that provides early active rehabilitation,

Correspondence: Torunn Askim*, Department of Neuroscience, Facultyof Medicine, NTNU, 7491 Trondheim, Norway.Email: [email protected] of Neuroscience, Faculty of Medicine, NorwegianUniversity of Science and Technology, Trondheim, Norway2Stroke Division, Florey Neuroscience Institutes, Heidelberg, Vic.,Australia3Faculty of Health Sciences, La Trobe University, Melbourne, Vic.,Australia4Clinical Service, St. Olavs Hospital, Trondheim, Norway5Stroke Unit, Department of Medicine, St. Olavs Hospital, Trondheim,Norway

Conflict of Interest: None declared.

Funding: Torunn Askim was supported through the Norwegian Fund forPostgraduate Training in Physiotherapy.

DOI: 10.1111/j.1747-4949.2011.00697.x

Research

© 2011 The Authors.International Journal of Stroke © 2011 World Stroke Organization Vol 7, January 2012, 25–31 25

therapeutic, and social interaction with the multidisciplinaryteam and the family may provide a more enriched environ-ment than general ward care. Previous studies conducted inacute stroke units suggest that interactions between patientsand multidisciplinary team members are low (6,9). The strokeunit at St. Olavs Hospital in Trondheim, Norway, is oftendescribed as the ‘gold standard’ stroke unit with good evidenceof benefit (4). It is therefore timely to explore some of thecore components delivered within this particular stroke unitthat promotes high levels of motor activity and interactionbetween stroke team members and the patient.

We hypothesized that:

• Acute stroke patients treated at the stroke unit at St. OlavsHospital in Trondheim would engage in high-level motoractivities such as transferring, standing, walking, and climbingstairs, a significant proportion of the active day from 8:00 amto 5:00 pm

• Nurses and therapists would work together with the patient(joint working practice)

• Even severely affected patients would spend time out of bedin the early phase after stroke, and

• Patients observed early in their stay would receive similaramounts of therapy as those observed later post stroke (i.e.therapy is not restricted to the first few days of hospital stay).

Methods

Population and settings

This cross-sectional observational study included patientsadmitted to the stroke unit at St. Olavs Hospital, TrondheimUniversity Hospital, Norway, between July 2008 and Decem-ber 2009. This 15-bed stroke unit that mainly treat patients atthe age of 60 years and older is a well-established, evidence-based, comprehensive unit, emphasizing a multidisciplinaryapproach and mobilization to standing or sitting position outof bed within the first 24 h after onset of symptoms and there-after motor training according to a task-oriented approach,focusing on independence in activities of daily living (ADL)(10,11). The multidisciplinary team consists of a physician, anurse, a physiotherapist (PT), a speech therapist (ST), and amember from the early supported discharge (ESD) team,which could be a nurse, a PT, or an occupational therapist.Family involvement is also considered an important part ofgood multidisciplinary stroke care.

All patients admitted to the stroke unit with the diagnosisof stroke were eligible for inclusion if time from onset ofstroke was less than 14 days, except for those already includedor with a devastating stroke receiving end-of-life palliativecare. Eligible patients were included if they were able andwilling to sign informed consent. Patients who were not ableto give informed consent were also included if their next of kingave oral consent to participation. This is in keeping withNorwegian consent procedures for patients unable to consentfor themselves.

The study was approved by the Regional Committee ofMedical and Health Research Ethics and Norwegian SocialScience Data Services.

Behavioral mapping

We used a standardized method of observation, behavioralmapping, for this study (9). Every second week, the strokepopulation was screened to determine whether observationwas feasible at that time. A minimum of four and a maximumof eight patients who met inclusion criteria and providedinformed consent on the day before observation were requiredto make observation feasible. Observation was conducted bytrained observers throughout the active day, from 8:00 am to5:00 pm, over a single day. The day of observation alternatedbetween Tuesday, Wednesday, and Thursday, intending toembrace the variability of activity during these days.

Patients were observed every 10 mins with the exception ofa few randomly scheduled breaks. At each observation, patientlocation, activity, and people present were recorded. Observa-tions lasted for approximately one-minute per patient, and thehighest level of activity observed during each observation wasrecorded. The route through the ward remained consistent.When bed rest was prescribed, this was recorded.

Patients were informed that they would be monitoredthroughout the day, and it was emphasized that they should donothing different on the day of observation.

Categories of activity

At each observation, 12 activities could be recorded. The 12activities were grouped into three main activity categories forthe analysis. These were:

• In bed

• Sitting out of bed, and

• Higher motor activity.Table 1 describes how the observed activities were grouped

into the analyzed categories.

Table 1 Classification of categories of observed activities

Observed activities Classification

1. No active motor supine In bed2. No active motor on left side In bed3. No active motor on right side In bed4. Sit support in bed In bed5. Sit support out of bed Sitting out of bed6. Transfer with hoist Sitting out of bed7. Roll and sit up Higher motor activity8. Sit with NO support Higher motor activity9. Transfer with feet on floor Higher motor activity

10. Standing Higher motor activity11. Walking Higher motor activity12. Climbing stairs Higher motor activity

Research T. Askim et al.

© 2011 The Authors.International Journal of Stroke © 2011 World Stroke Organization Vol 7, January 2012, 25–3126

Location and people present

At each observation, patient location and people present wererecorded. In addition to the 11 people categories (e.g. family,nurses, doctors, PT, ST, and so on), six prespecified groupingsof people were used to help describe the multidisciplinaryapproach. These were:

• PT alone

• PT with nurse, ST, ESD team, or family,

• ESD team alone

• ESD team with nurse, ST, PT, or family

• ST alone, or

• ST with nurse, PT, ESD team member, or family.

Other measures

The following variables were assessed at inclusion: age, gender,medical history, time since stroke, Oxfordshire Classificationof Stroke (12) and National Institute of Stroke Scale (NIHSS)(13). Information about time to first mobilization from onsetof stroke symptoms and who helped with first mobilizationwas collected from the patients’ record. It is standard proce-dure to collect this information from all patients admitted tothe stroke unit in Trondheim.

Statistical analysis

A database (Microsoft Access 2007) was designed to auto-matically calculate the highest level of activity in every 10-mininterval. Descriptive statistics were used for the main results.Parametric tests were applied for normally distributedvariables, while nonparametric tests were used for skewedvariables.

Patients were grouped into mild (NHISS score �7), mod-erate (NHISS score 8–16), and severe stroke (NHISS >16)for subgroup analysis. Kruskal–Wallis nonparametric test forseveral independent samples was used to analyze differences intime to first mobilization between the three different severitygroups. Mann–Whitney U-test was used to analyze the differ-ence in amount of therapy given to patients observed withinthe first week after their stroke compared to those observedwithin their second week after stroke.

The effect of age, gender, and stroke severity on motoractivity was analyzed by the use of linear regression with‘higher motor activity’ as the dependent variable.

Results

Over 18 months, 124 patients were included. This makes up23% of all stroke patients admitted to the stroke unit duringthis period. Six patients with incomplete data due to dischargebefore 1:00 pm on the day of observation were excluded, anda further patient was found to not have had a stroke, leaving atotal of 117 patients included in the analysis. The median agewas 80·5 years, ranging from 51 to 94 years (56·4% male).

Seventy-five percent of the included patients underwent theirfirst ever stroke, 42·7% of the patients suffered from a mod-erate or severe stroke (NIHSS score > 7), and 20·4% had expe-rienced hemorrhagic stroke. The median day from onset ofstroke to day of observation was six-days, ranging from 0to 17 days (Table 2). Two patients were observed more than14 days after stroke because time for first sign of symptomswas unknown at inclusion. Although these two patients felloutside the inclusion criteria, they were kept in the analysis.

Out of a total of 6435 observations, 3·4% of the time,patients could not be observed. On 87 occasions, ‘notobserved’ was recorded to protect privacy when the patientswere toileting. A further 133 were ‘not observed’ because the

Table 2 Baseline characteristics

n = 117

AgeMean (SD) 78·7 (9·2)Range 51·0–94·0

Male, n (%) 66 (56·4)First ever stroke, n (%) 88 (75·2)Time since stroke

Mean (SD) days 6·7 (3·7)Range 0·0–17·0

NIHSS scoreMean (SD) 8·2 (7·0)Range 0·0–27·0

Severity groups, n (%)NIHSS � 7 67 (57·3)NIHSS 8–16 31 (26·5)NIHSS > 16 19 (16·2)

Modified Rankin Scale (mRS), n (%)mRS 0–3 33 (28·2)mRS = 4 60 (51·3)mRS = 5 24 (20·5)

Oxfordshire stroke classification, n (%)TACI 24 (20·5)PACI 20 (17·1)POCI 20 (17·1)LACI 27 (23·1)Hemorrhage 24 (20·5)Not classified 2 (1·7)

Side of hemiparesis, n (%)Left 60 (51·3)Right 54 (46·2)Brainstem 3 (2·6)

Prestroke accommodation, n (%)Home alone 60 (51·3)Home with family 54 (46·2)Hostel 3 (2·6)

Prestroke mobility, n (%)Independent no aids 74 (63·2)Independent with aids 37 (31·6)Dependent 6 (5·1)

SD, standard deviation; NIHSS, National Institute Stroke Scale; TACI,total anterior circulation infarct; PACI, partial anterior circulationinfarct; POCI, posterior circulation infarct; LACI, lacunar infarct.

ResearchT. Askim et al.


patients were off the ward for brain imaging, Doppler, echo, orother reasons. Finally, 30 observations were missing becauseof an observer break.

Amount of activity

In total, patients spent 30·3% of the time in bed, 46·4% ofthe time sitting out of bed, and 19·9% of the time in highermotor activities such as transferring, standing, walking, orstair climbing (Fig. 1). Seventeen patients (14·5%), 12 withmild stroke and five with moderate stroke, practiced stairclimbing during the observation time.

Patients with mild stroke spent most of the active day out ofbed, while those with severe stroke were out of bed for aboutone-third of the day (Fig. 2). The time spent out of bed for the

severely affected patients was mostly given as two doses a day(range 1 to 3). The median (range), NIHSS conscious itemscore, measured on the day before observation, for thesepatients was 1 (0–2), indicating slightly impaired awareness.

Stroke severity had a strong effect (P < 0·000) on motoractivity, with less ‘higher motor activity’ in the severelyaffected patients. There was also a trend toward an effect of age(P = 0·079), with the oldest patients being less active. Genderdid not affect motor activity.

The team approach

Forty-seven percent of the patients were mobilized for the firsttime with assistance from two or more nurses, 29·9% of thepatients were mobilized with a PT together with a nurse,

Fig. 1 Mean percentage of observed time spent in different motor categories for all included patients.

Fig. 2 Mean percentage of observed time spent out of bed for patients with mild, moderate, and severe stroke. NIHSS, National Institute of Stroke Scale.



19·7% of the patients were mobilized by a nurse alone, while2·6% of the patients were mobilized for the first time by a PTalone.

In total, patients spent 48·7% of the time with someonepresent, most frequently with nurses (22·5% of the day).Patients spent on average 7·9% of the time with a PT. Out ofthis time, 4·4% of the time was spent with a PT alone, while3·5% was spent with a PT together with other members of thestroke unit team, such as a nurse, a member from the ESDteam, or family (Fig. 3).

There were no differences between the amount of physi-otherapy given to those observed within the first week poststroke (n = 68) compared to those observed within theirsecond week post stroke (n = 49), 5·6% vs. 7·3% of observedtime, P = 0·409.

Time to first mobilization

Data on time from onset to first mobilization after onsetof stroke were available for 103 patients. Mobilization wasdefined as getting out of bed in sitting or standing position forat least 5 mins, supported by a nurse or PT. Fourteen patientswere missing because time for onset of stroke was unknown(n = 11) or time for first mobilization was not registered(n = 3). Median (interquartile range) time from stroke to firstmobilization was 18·1 (7·8–43·0) hours for the whole group.We found a significant difference in time to first mobilizationbetween the different severity groups (P = 0·002), and it wasthe severely affected patients who differed significantly fromthe mildly and moderately affected patients. In total, 57·8% ofall patients were mobilized within 24 h after onset of stroke.Within the different severity groups, 67·8% of the mildly,55·6% of the moderately, and 25·0% of the severely affectedpatients were mobilized within this time frame (Table 3). A

significant proportion of this time is spent on arriving to thehospital, as 72% of all patients were mobilized within 24 hafter admission.

Discussion

As hypothesized, we found that stroke patients treated at theStroke unit at St. Olavs Hospital in Trondheim spent timeengaged in higher motor activities 20% of the time from 8:00am to 5:00 pm, which is considered to be a significant propor-tion of the active day. Even those with severe stroke were outof bed several times a day. Furthermore, evidence of jointworking practice during patient care and ongoing therapistinvolvement over time, consistent with a rehabilitationphilosophy, was confirmed.

Fig. 3 Mean percentage of observed time spent with therapist alone or in team with other professions or family. Categories are not mutually exclusive.PT, physiotherapist; ESD, early supported discharge team member; ST, speech therapist.

Table 3 Time from stroke to first mobilization

All patients (n = 103)Mean (SD) hours 31·5 (36·1)Median (range) 18·1 (1·1–183·5)Number (%) mobilized within 24 h 59 (57·8)

Patients with mild stroke, NIHSS � 7 (n = 59)Mean (SD) hours 20·9 (20·6)Median (range) 13·2 (1·1–94·5)Number (%) mobilized within 24 h 40 (67·8)

Patients with moderate stroke, NIHSS 8–16 (n = 27)Mean (SD) hours 33·5 (36·8)Median (range) 19·2 (3·4–167·0)Number (%) mobilised within 24 h 15 (55·6)

Patients with severe stroke, NIHSS > 16 (n = 17)Mean (SD) hours 65·4 (54·0)Median (range) 67·6 (2·42–183·5)Number (%) mobilised within 24 h 4 (25·0)

SD, standard deviation; NIHSS, National Institute Stroke Scale.



In contrast to stroke units sampled using behavioralmapping in the European Registers Of Stroke (EROS) study(6), where out-of-bed activity ranged from 2% of the day in St.Petersburg, Russia, to 56% of the day in London, UK, wefound that patients managed in the stroke unit in Trondheimspent almost 70% of the active day (8:00 am–5:00 pm) out ofbed or engaged in higher motor activities. Even the mostseverely affected patients spent a significant proportion of theday out of bed. According to published data, the Trondheimunit provides the highest level of activity for stroke rehabili-tation, together with some of the best published outcomeresults; therefore, it is highly unlikely that this policy isharmful, and the ongoing ‘A Very Early Rehabilitation Trial(AVERT)’ (14) will determine whether the activity itself is animportant driver of these excellent outcomes.

In general, bed rest is regarded as one of the main causes forcomplications in critically ill patients (15). This is also evidentfor stroke patients (16), and prevention, early recognition, andmanagement of post stroke complications are regarded asessential aspects of stroke unit care (1). Our results show thatmildly affected patients spent on average 79·3% of the day outof bed, indicating that they spent less than two-hours restingin bed during the day. These results were achieved at the newward at St. Olavs Hospital, which includes a number of passiverecreation areas and a dining area in its design. Patients areencouraged to eat all meals together in the dining room, andto make use of the available spaces away from their bedside.This likely contributed to the low rate of bed rest in those withmild stroke. However, we have previously shown that it ispossible to achieve similar levels of activity in a more tradi-tional ward (5). While the environment has a role to play, webelieve the philosophy of care is equally if not more importantin promoting activity.

A number of guidelines recommend that even severelyaffected stroke patients should be out of bed (17,18). We haveshown that this is possible; however, it is important to notethat this was achieved in a number of episodes throughout theday rather than in a single, extended block of time. Patientswith severe stroke probably spend a lot of energy on simplemotor tasks such as rising up from bed, dressing, and trans-ferring from bed to a chair, and therefore also need moretime for rest between each mobilization compared to the lessaffected patients. Some of the severely affected patients hadreduced consciousness on the day before mobilization. Earlymobilization has been shown to improve blood oxygenationand alertness (19). These physiological changes may beparticularly helpful for those with severe stroke, and it isimportant to consider that reduced consciousness does notnecessarily exclude patients from getting out of bed.

Joint working practice among team members was evidentin this study. On average, PTs spent 43 mins a day (7·9% of thetime from 8:00 am to 5:00 pm) with each patient. Out of these43 mins, the PT worked together with other members of themultidisciplinary team, mainly nurses, 19 mins a day (3·5% ofthe time from 8:00 am to 5:00 pm). These results indicate that

the team approach engenders joint working practice, withteam members working beyond the confines of their ownprofessional knowledge base. Hence, the multidisciplinaryapproach at this unit seems to include important elements ofinterdisciplinary care (20,21). Patient, family, and staff inter-action and engagement in activity also represent key compo-nents of an enriched environment, and if enrichment hasthe same benefit to recovery in humans as it does in stroke-affected animals (8), this may help explain the stroke uniteffect. However, these aspects should be more closely inves-tigated in future studies.

Another important finding was that most patients weremobilized for the first time by one or more nurses or by thenurse together with the PT. Close teamwork over time hasmeant that nurses in this unit feel confident to mobilizepatients on their own. Patients do not need to stay in bed andwait for the PT to be mobilized, and mobilization can continuewhenever needed throughout the day or on weekends. The factthat the nurses do take part in the training of independencein ADL throughout 24 h may explain some of the reasons forthe high activity level at the stroke unit in Trondheim.

The main strength of the present study was the inclusionof an unselected group of stroke patients receiving evidencebased stroke unit care. The mean age was 79 years, and 16·2%suffered from severe stroke, which is very close to the realNorwegian stroke population (22). However, a greater pro-portion than expected (20·5%) suffered from hemorrhages.Observation every second week increases the risk of includ-ing severely affected patients and also patients with longerhospital stay because of hemorrhages, which again increasesthe risk of underestimating the actual amount of physicalactivity.

Another strength is the use of behavioral mapping as anobservation technique. Activity monitoring after stroke by useof body-worn sensors only gives information about the activ-ity level (23), and can be difficult to use in the acute setting. Incontrast, observing the patient every 10 mins using standardrecording of physical activity, people present, and locationprovides richer information about a stroke unit. It is possiblethat we may have missed some physical activity carried outbetween each observation, especially some of the transfer situ-ations for the severely affected patients that only take a fewminutes to carry out but represent important activity forthis group of patients. Nevertheless, we believe the activityreported here is a representative sample of patient activity overthe active day.

In conclusion, this study has helped to unpack the caredelivered in the evidence-based stroke unit at St. Olavs Hos-pital. We have shown that while those patients with severestroke may start their out of bed activity later than those withmild stroke, patients in the first two-weeks after stroke can beout of bed for considerable periods in the day. However, it isimportant to underscore that early and frequent mobilizationis one out of several important components of evidence-basedstroke unit care, and the unique effect of this component is



currently being assessed as part of a large multicentre trial(14). The most appropriate amount of physical activity duringthe acute and subacute phase is still unknown, and futurestudies should assess the association between the amount ofphysical activity and outcome after stroke to further increasethis knowledge.

Acknowledgements

The authors want to thank Nina Pedersen and Turid Aasheimfor their contribution in collecting data. We also want to thankJan Chamberlain and Li Chun Quang at the National StrokeResearch Institute for processing and preparing the data foranalysis. Finally, we want to thank all the stroke patients andtheir families who agreed to take part in the study.

References1 Langhorne P, Pollock A. What are the components of effective stroke

unit care? Age Ageing 2002; 31:365–71.2 Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Treatment

in a combined acute and rehabilitation stroke unit: which aspects aremost important? Stroke 1999; 30:917–23.

3 Bernhardt J, Thuy MN, Collier JM, Legg LA. Very early versus delayedmobilisation after stroke. Cochrane Database Syst Rev 2009; (1)CD006187. DOI: 10.1002/14651858.CD006187.pub2.

4 Stroke Unit Trialists’ Collaboration. Organised inpatient (stroke unit)care for stroke. Cochrane Database Syst Rev 2007; (4) CD000197. DOI:10.1002/14651858.CD000197.pub2.

5 Bernhardt J, Chitravas N, Meslo IL, Thrift AG, Indredavik B. Notall stroke units are the same: a comparison of physical activity patternsin Melbourne, Australia, and Trondheim, Norway. Stroke 2008;39:2059–65.

6 Wellwood I, Langhorne P, McKevitt C, Bernhardt J, Rudd AG,Wolfe CD. An observational study of acute stroke care in four coun-tries: the European registers of stroke study. Cerebrovasc Dis 2009;28:171–6.

7 Johansson BB. Functional and cellular effects of environmentalenrichment after experimental brain infarcts. Restor Neurol Neurosci2004; 22:163–74.

8 Janssen H, Bernhardt J, Collier J et al. An enriched environmentimproves sensorimotor function post-ischemic stroke. NeurorehabilNeural Repair 2010; 24:802–13.

9 Bernhardt J, Dewey H, Thrift A, Donnan G. Inactive and alone: physi-cal activity within the first 14 days of acute stroke unit care. Stroke2004; 35:1005–9.

10 Indredavik B, Bakke F, Solberg R, Rokseth R, Haaheim LL, Holme I.Benefit of a stroke unit: a randomized controlled trial. Stroke 1991;22:1026–31.

11 Indredavik B, Slordahl SA, Bakke F, Rokseth R, Haheim LL. Stroke unittreatment. Long-term effects. Stroke 1997; 28:1861–6.

12 Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classificationand natural history of clinically identifiable subtypes of cerebral inf-arction. Lancet 1991; 337:1521–6.

13 Brott T, Adams HP Jr, Olinger CP et al. Measurements of acute cer-ebral infarction: a clinical examination scale. Stroke 1989; 20:864–70.

14 Bernhardt J, Dewey H, Collier J et al. A Very Early Rehabilitation Trial(AVERT). Int J Stroke 2006; 1:169–71.

15 Brower RG. Consequences of bed rest. Crit Care Med 2009; 37:S422–8.16 Govan L, Langhorne P, Weir CJ. Does the prevention of complications

explain the survival benefit of organized inpatient (stroke unit) care?Further analysis of a systematic review. Stroke 2007; 38:2536–40.

17 Hill K. Australian clinical guidelines for acute stroke management2007. Int J Stroke 2008; 3:120–9.

18 Nasjonale retningslinjer for behandling og rehabilitering vedhjerneslag. Helsedirektoratet 2010. Available at: http://www.helsebiblioteket.no/Retningslinjer/Hjerneslag/Forord+og+innledning

19 Indredavik B, Loge AD, Rohweder G, Lydersen S. Early mobilisation ofacute stroke patients is tolerated well, increases mean blood pressureand oxygen saturation and improves level of consciousness. Cerebrov-asc Dis 2007; 23:65.

20 Barr O. Interdisciplinary teamwork: consideration of the challenges.Br J Nurs 1997; 6:1005–10.

21 Suddick KM, De Souza L. Therapists’ experiences and perceptions ofteamwork in neurological rehabilitation: reasoning behind the teamapproach, structure and composition of the team and teamworkingprocesses. Physiother Res Int 2006; 11:72–83.

22 Indredavik B, Ellekjaer H, Selmer R. Stroke in Norway. Int J Stroke2008; 3:205–6.

23 Gebruers N, Vanroy C, Truijen S, Engelborghs S, De Deyn PP.Monitoring of physical activity after stroke: a systematic review ofaccelerometry-based measures. Arch Phys Med Rehabil 2010; 91:288–97.



Documents

Stroke patients do not need to be inactive in the first two-weeks after stroke: results from a stroke unit focused on early rehabilitation