Commentary on Arias M & Smith L (2007) Early mobilization of acute stroke patients. Journal of Clinical Nursing 16, 282–288


  • Fiona Ellery BA,

  • Julie Bernhardt PhD

Fiona Ellery, Neuroscience Trials Australia, Level 2, Neurosciences Building, Austin Health, 300 Waterdale Road, Heidelberg Heights, Melbourne, Vic. 3081, Australia. E-mail:

We read with interest the paper by Arias and Smith (2007) in which they report the views and knowledge of 99 health professionals about early mobilisation, and their level of consensus about how early mobilisation should be implemented. The results of this survey highlight the lack of agreement between stroke care professionals about what constitutes early mobilisation and who should decide when it (and if) it should begin.

We are pleased to know that the authors conclude that there exists a ‘need for an evidence base that defines what early mobilisation is, and provides guidance on the frequency, duration and intensity of specific activities’. Although there is something to be gained from seeking the views of those in the field, to make advances in patient care, clinicians often need to be convinced of benefit before change will occur. Indeed, attempting to gain consensus from clinicians, no matter how experienced, as a means of progressing knowledge or making real changes in clinical practice usually fails. Even when guidelines exist to express expert opinion (National Clinical Guidelines for Acute Stroke Management 2003), change in clinical practice is slow or non-existent (Hammond et al. 2005). In the case of early mobilisation, where considerable controversy exists about what constitutes early mobilisation (Diserens et al. 2006, Bernhardt et al. 2007), the importance of developing an evidence base becomes even more pressing.

Our group has been studying very early mobilisation for the past six years. In 2006, the phase II safety and feasibility study was completed and the results of this trial are now published (Bernhardt et al. 2006). We are now one year into the final phase of the AVERT study – an international, multi-centre randomized controlled study of 2104 patients (Bernhardt et al. 2006). This trial examines not only the possible benefits and harms of very early mobilisation, but also the cost-effectiveness of the intervention.

We define very early mobilisation as ‘functional mobilisation out of bed commencing within 24 hours of onset of stroke’. Passive movements are not considered mobilisation in our trial. This protocol was developed in consultation with Indredavik et al. (1991) following many years of clinical research with early mobilisation in the acute stroke setting.

In preliminary work conducted to support development of the clinical trial, we found that nurses and physiotherapists play a major role in initiating and supporting mobilisation (Bernhardt et al. 2004). This seems to be the case in the UK as well as here in Australia. Consequently, the early mobilisation intervention in our trial is delivered by a nurse/physiotherapist team. Either member of the team can initiate a mobilisation, encouraging a multidisciplinary team approach. Importantly, the very early mobilisation protocol needed to be pragmatic in design, so as to be able to incorporate centres with different care and staffing models.

It was interesting to note that most clinicians surveyed in the study by Arias and Smith believed that patients should be fully conscious and medically stable before mobilisation. We require patients to be able to at least react to verbal commands; however, they do not have to be fully conscious. The inclusion of these patients in our study is based on the clinical experience of Indredavik in Trondheim, Norway (Indredavik B, 2003, St Olavs Hospital, Trondheim, Norway pers. comm.). Medical stability is evaluated during mobilisations according to specific guidelines using physiological parameters. The setting of safety criteria based on the Norwegian experience has worked well within the structure of our clinical trial. We have experienced no difficulties (ethical or practical) in having the trial protocol accepted by numerous ethics committees, nor have there been any safety concerns reported to date.

We have adhered to good clinical practice guidelines (TGA 2000) and the National Statement (NHMRC 1999) for our trial. These local and international ethical and quality standards for the design, conduct and reporting of a clinical study will result in reliable trial results – essential if clinical change is to be initiated. It is of course important to note that our model of very early mobilisation may not prove to be the most appropriate. Setting the criteria to determine what very early mobilisation is and trialing these criteria in a rigorous manor will provide results which we hope will more accurately guide clinicians in their further work.

It is only once we have hard evidence of the benefit of very early mobilisation, defining how much, what sort and how early mobilisation should take place, that clinical change will happen. Qualitative work relating to the clinicians’ ongoing knowledge of evidence-based research surrounding early mobilisation is certainly beneficial. It will greatly assist us to understand where further work is required when trying to implement new practice based on evidence, and will be important when observing advancements in the clinical care of the stroke patient.