Use of Outcome Measures in Norwegian Neurological Physiotherapy

Artikkelen ble mottatt 11.12.2005 og akseptert for publisering 19. september 2006. Artikkelen har gjennomgått ekstern manuskriptvurdering i henhold til våre retningslinjer (se www.fysioterapeuten.no/fag). Forfatteren har oppgitt å ikke ha interessekonflikter i forhold til studien.

Abstract
Purpose: To investigate the extent of outcome measures (OMs) use in neurological physiotherapy and to collect experiences from this. Methods: A questionnaire wasdistributed to 115 institutions woith Neurological Rehabilitation Department (NRDs) across Norway. The NRD's head physiotherapists were responsible for completing the questionnaire which containedquestions about type and size of NRD and about OMs use. Results and conclusion: Eighty-eight questionnaires were returned, and of these 74 reported treating neurological patients. OMs were routinely used at60 out of 74 (81 percent) responding NRDs. The Barthel Index, the Motor Assessment Scale and the Berg Balance Scale were most commonly used. OMs were mostly used in stroke patients. The reasons for using them were to indicate change for progression and potential, aid patient motivation and help in the categorisation of patients. Lack of experience was the most requent objection for not using OMs. Key words: Neurology, Physiotherapy, Outcome Measure, National Survey.

n

Introduction
The ability to reliably and validly assess and report the effect of treatment from clinical trials is essential for evidence-based health care (1,2). The increasing demand for evidence-based clinical practice has challenged all service providers to demonstrate methods for reporting effect of treatment, cost-effectiveness and the quality of care within the National Health Service (3-5). Increased patient awareness has also contributed to this process.

The last decade there has been an increase in development of outcome measures (OMs) that aim to quantify the recovery following treatment for patients after stroke (6,7). OMs may be defined as a «measurement tool used to document change in one or more constructs over time». The construct here is recognised as patient characteristics being measured (8).

Physiotherapists working within neurology have previously in general been reluctant to evaluate and measure their practice, mainly because of the major challenge to find measures that condenses all (often multidimensional) aspects of the patient’s clinical status into reliable parameters (9,10). Quality of movement has been a key factor in neurological physiotherapy, but is difficult to quantify since minor changes in function may have a large impact of the patients’ quality of life (11). The increased use of The World Health Organization’s measurement system - the International Classification of Functioning, Disability and Health (ICF) - has helped physiotherapists working within neurology to find appropriate OMs that organize the outcomes according to the two main domains 1) body structure and function, and 2) activity and participation (12).

When OMs are referred to as being standardised, this means that the psychometric or measurement properties (i.e. the sensitivity, reliability and validity) of the tool have strong scientific support. Standardised measures are developed and applied for a wide range of purposes such as to 1) Predict the needs of a patient and hence decide patient oriented treatment goals that are appropriate, measurable, achievable and functional 2) Discriminate between normal and abnormal characteristics and improve meaningful communication within the multidisciplinary team 3) Evaluate treatment effectiveness, existing practice and use of resources in order to prioritise, argue for funding, give an objective analysis of costs/benefits and indicate whether the minimum criteria for national standards are met or not 4) Facilitate research by bridging the gap between clinical practice and the scientific community (2,9,13).

The primary aim of this study was to investigate the extent of OM use amongst physiotherapists working with neurological patients in all Neurological Rehabilitation Departments or rehabilitation centres (NRDs) in Norway. I also intended to collect experiences from this use. The idea for this study arose when our rehabilitation department was in the process of implementing standardised OMs. Answers to the following questions were sought by handing questionnaires to physiotherapists working within neurology:

- At what proportion are Norwegian physiotherapists in NRDs using OMs regularly?
- Which categories of neurological patients are the most commonly tested?
- Which OMs are most commonly used?
- What are the most common reasons for using these OMs?
- What are the reasons for not using OMs?

Material and methods
An Internet search was done resulting in a list of 115 Norwegian NRDs. Then a questionnaire was distributed to all NRDs during the spring 2002. The head physiotherapist of each NRD was responsible for completing the questionnaire and was asked to do this in cooperation with all their physiotherapy colleagues treating neurological patients.

The questionnaire contained eight questions with two sections, one general and one specific part. The general part asked for the respondents working environment; type of hospital (i.e. university or local hospital or rehabilitation centre), numbers of wards and of clinical physiotherapy posts connected to the department treating neurological patients. The specific part addressed the types of OMs used and which categories of patients were most frequently tested. In addition, the respondents were asked to comments on their experience with the OMs and the reasons for not using them on a regular basis. One set of questionnaires were distributed, and all respondents answered anonymously. The questionnaire used pre-coded responses, with an open section under each question in the specific part. Departments who reported using OMs regularly (not specified how often) throughout the rehabilitation process were defined as routine OMs users. Descriptive statistics were used for the data analysis.

Results
Out of the 115 NRDs, 88 returned their questionnaires, giving a response rate of 77 percent. Of these, 74 reported to have neurological patients on a regular basis, and the responses of these were further analysed. The questionnaires were completed by the physiotherapists heading the NRDs, representing a total of approximately 340 physiotherapists treating neurological patients. These figures were estimated from the number of physiotherapists reported from each NRD. Approximately 60 percent of the 74 respondents were from university or local hospitals, whereas one-third was from rehabilitation centres (Figure 1). The majority of NRDs were characterised by having all neurological patients on one ward, with one to six physiotherapists connected to their neurological department.

81 percent of the NRDs used at least one standardised OM routinely. Independent of type of NRD, either less than 25 percent or more than 75 percent of the patients were reported to be tested with standardised and/or non-standardised OMs (Figure 2). Of the NRDs testing less than 25 percent of the patients, the Rehabilitation Centres were overrepresented compared to the University and Local Hospitals (data not shown). Of the NRDs testing more than 75 percent of the patients, all but one reported patients with stroke to be the most frequently tested category (data not shown).

Stroke was the dominating diagnosis amongst the most frequently tested patients, being reported by 59 percent of the departments. A few NRDs also reported to use OMs most routinely on patients having Parkinson’s disease (n=4) and multiple sclerosis (n=3).

The most frequently used OMs were the Barthel Index (n=33), the Motor Assessment Scale (n=32) and the Berg Balance Scale (n=18) (Figure 3). Non-standardised and/or in-house developed OMs, meaning modifications of existing standardised OMs, were used at 27 percent of the NRDs. Just three percentages of the NRDs reported using unstandardized OMs only. The Berg Balance Scale, Functional reach and in-house developed OMs were used with patients having Parkinson’s disease and the Berg Balance scale with patients having multiple sclerosis.

The users’ experiences with using standardized OMs were overall positive. Table I gives the percentage proportion of reported reasons for using OMs routinely. Most respondents agreed or partly agreed that OMs were helpful for patient motivation (79 percent), indicated change for progression and potential (98 percent) and were a useful tool for patient categorisation (82 percent). Some respondents also reported using OMs for research purposes (n=3) for communication with the multidisciplinary team (n=3), and due to requirements from the management (n=2).

The two most common reasons for not using OMs were lack of experience in administering them (n=21), and the view that OMs were not sufficiently sensitive to detect qualitative changes within the function and/or activity being tested (n=10) (Figure 4).

Only a few NRDs reported lack of time (n=2) and lack of evidence to support the psychometric properties (n=2) as a reason for not using OMs on a regular basis. Surprisingly, respondents at eight NRDs reported to have gained a new interest in OMs by completing the questionnaire.

Discussion
This survey addressed the use of OMs among physiotherapists treating neurological patients in Norway and showed that a large proportion used OMs regularly. The most frequently used OMs were the Barthel Index, the Motor Assessment Scale and the Berg Balance Scale. The most common neurological group of patient being tested had stroke, followed by Parkinson’s disease and multiple sclerosis, from most to least frequent, respectively.

Cole and colleagues (14) performed the first survey on the use of OMs in generic rehabilitation early in the 1990ies. At that time, the use of OMs was in its infancy, and correspondingly they reported on a number of barriers for use, such as lack of knowledge regarding available instruments and time needed to implement them (14).

In later studies considering neurological rehabilitation, it appeared that OMs were used by approximately 40-85 percent of the responding physiotherapists, though with a large variation in the response rate from which this is derived (15-17). A low response rate has been associated with the tendency by respondents not to react to questionnaires when no OMs were used (17). The high proportion of NRDs reporting to regularly use OMs in this study therefore gives a positive picture of the OM application situation in Norway, although I would rather have seen all departments in this group.

The Barthel Index and the Functional Independent Measure are reported to be the most commonly applied OMs in NRDs in Germany, Italy and the UK (9,16-18) which are in concordance with the result here. In addition to these, the Motor Assessment Scale, the 10m Walk, the Motricity Index and the Nine-hole Peg Test are reported to be commonly used in Ireland and the UK (15,16). In agreement with our results, stroke has been the category of neurological patients who most frequently are being tested with the OMs mentioned above (9,15-18).

Interestingly, the Norwegian NRDs reported using quantitative OMs only, although quality of movement during functional performance is an important aim of the neurological physiotherapy intervention (19). This may reflect the difficulties in describing quality of movement, apart from comparing it to normal movement (19). The quality of movement in different functional tasks is traditionally described qualitatively in writing, together with an analysis of its functional consequences to be communicated within the multidisciplinary team.

None of the NRDs reported using any self-reported multidimensional OMs (e.g. the SF-36 and the Stroke Impact Scale). These generic OMs may give additional and valuable global information regarding the perceived health-related quality of life. Also in previous studies of rehabilitation after stroke there has been a tendency towards measuring only the physical aspect of physiotherapy interventions (20,21). However, all three most commonly used OMs in this study appear to adequately fulfil the most important quality criteria for selecting OMs (8).

First, the OM must be specific for assessing the construct and domain of interest (12). The preferred OMs tend to differ across environments where different neurological treatment approaches are preferred as they are being supported on different theoretical foundations (11). The Barthel Index, the Motor Assessment Scale and the Berg Balance Scale are related to the ICF domain body structure and function and measure the activity level. They still quantify quite different aspects of the patient’s functional characteristic. If the Barthel Index, which tests basic activities in daily living, is used to monitor more specific physiotherapy outcomes such as balance, there is a risk that changes in progress might not be documented, due to the scales insensitivity to small changes in functional status, as well as its ceiling effects (22-24). Still, in terms of prognosis the Barthel’s subscale concerning urinary incontinence is a major prognostic factor in stroke patients (25). The Motor Assessment Scale and the Berg Balance Scale are more closely related to important constructs of physiotherapy, measuring movement and balance, respectively.

Secondly, the psychometric properties of each OM need to be documented for the chosen neurological patient category (26). The three most routinely used OMs in this study have all been found reliable, valid and sensitive to change for patients after stroke, and have been validly translated into Norwegian (14,22). Also the clinical applicability of the three OMs seems good as they require little equipment, are easy to use, and to score.

Finally, the multidimensional influences of the chosen OM – that the OM might not only reflect the result of physiotherapy – is a highly relevant issue and is probably the most difficult to account for (20,27). Function in everyday activities is a key end-result, but will be influenced by a number of social, medical, psychological and environmental factors (11,27,28). It is therefore important to consider these influences when analysing results from effect-studies, and bear in mind that the selected OM might not be sensitive enough to measure change due to physiotherapy and hence the intervention may be misjudged and undervalued (15). The possibility of finding one single OM gold standard that takes all characteristics into consideration is unlikely. Physiotherapists therefore seems obliged to compromise, or to use a test battery including OMs that are generic (e.g. the Stroke Impact Scale) and more physiotherapy specific (e.g. the Motor Assessment Scale).

The respondents’ lack of experience with OMs as well as viewing their sensitivity as inadequate were the main reasons for not using OMs in this study as in earlier studies (14,16,29). To overcome these barriers in the future, it is important that Norwegian standards for neurological physiotherapy include evidence-based guidelines on which OMs to apply. Separate guidelines should be developed for different neurological patient groups according to the different ICF domains. Utilizing identical OMs in similar settings will facilitate comparisons between treatment methods and outcomes on a national level. These guidelines should incorporate both generic rehabilitation OMs and more physiotherapy specific OMs.

Most of the NRDs in this survey used standardised OMs routinely, but still about a quarter of the NRDs reported an additional use of non-standardised OMs. However, just three percentages of the NRDs used such OMs only, which is in agreement with previous studies (9,15,16,18). Use of non-standardised OMs should, if possible, be avoided since they make comparisons between different interventions and institutions more difficult.

When interpreting the findings of this study, a number of points should be born in mind. The physiotherapist responding to this questionnaire might have acted differently when giving their answers. Although the physiotherapist heading each NRD was asked to complete the questionnaire in cooperation with each clinical physiotherapist, it may have been difficult to summarise all individual answers in a representative way. Obviously there is also the risk that the questionnaire was completed by one physiotherapist alone without discussion with his or her colleagues. Furthermore, no information was collected regarding the characteristics of the clinical physiotherapists (e.g. clinical experience, year of graduation).

In conclusion, this study showed that OMs were routinely used in most NRDs, with the Barthel Index, the Motor Assessment Scale and the Berg Balance Scale being the most commonly used. These solidly evidence-supported OMs were mostly applied with stroke patients. Reported reasons for using OMs were to indicate change, to aid patient motivation and to categorise patients. The national guidelines should be further developed and include suggestions on which generic and physiotherapy specific OMs to use for the different categories of neurological patients.

Acknowledgements:
This study was supported by the Department of Physiotherapy, Haukeland University Hospital, Bergen, Norway. I would also like to thank all physiotherapists who contributed to this study by completing and returning the questionnaires. Thanks also to the referees for comments of great value.

References
1. Enderby P, Kew E. Outcome measurement in physiotherapy using the World Health Organisation’s classification of Impairment, Disability and Handicap: a pilot study. Physiotherapy 1995; 81(4): 177-80.
2. Ottenbacher KJ, Hinderer SR. Evidence-based practice: methods to evaluate individual patient improvement. Am J Phys Med Rehabil 2001; 80(10): 786-95.
3. Orchard C. Comparing health care outcomes. BMJ 1994; 308(6942): 1493-6.
4. Mawson S. What is the SF-36 and can it measure the outcome of physiotherapy? Physiotherapy 1995; 81(4): 208-12.
5. Stokes M. Principles of physiotherapy assessment and outcome measures. London: Mosby International; 1998.
6. Poole JL, Whitney SL. Assessment of motor function post-stroke: a review. Phys Occup Ther Geriatr 2001; 19(2): 1-22.
7. Turner P. Evidence-based practice and physiotherapy in the 1990s. Physiotherapy Theory and Practice 2001; 17(2): 107-21.
8. Finch E, Brooks D, Stratford P, Mayo N. Physical rehabilitation outcome measures, 2nd ed. Canada: Lippincott, Williams & Wilkins; 2002.
9. Chesson R, MacLeod M, Massie S. Outcome measures used in therapy departments in Scotland. Physiotherapy 1996; 82(12): 673-9.
10. Bentley J. A physiotherapy perspective on the dilemmas of outcome measurement in severe and complex brain injury rehabilitation. Physiotherapy 2001; 87(11): 593-99.
11. Partridge C, Edwards S. The bases of practice – neurological physiotherapy. Physiother Res Int 1996; 1(3): 205-8.
12. ICF International Classification of Functioning, Disability and Health. Geneva: World Health Organization 2001 [cited 2003 May 20]. Available from: http://www.who.int/icf
13. Crombie I, Davies H, Abraham S, Florey C. The Audit Handbook: improving health care through clinical audit. Chichester: John Wiley & Sons; 1993.
14. Cole B, Finch E, Gowland C, Mayo N. Physical rehabilitation outcome measures. London: Williams & Wilkins; 1995.
15. Stokes E, O’Neill D. The use of standardised assessments by physiotherapists. Br J Ther Rehabil 1999; 6(11): 560-65.
16. Turner-Stokes L, Turner-Stokes T. The use of standardised outcome measures in rehabilitation centres in the UK. Clin Rehab 1997; 11: 306-13.
17. Torenbeek M, Caulfield B, Garrett M, Van Harten W. Current use of outcome measures for stroke and low back pain rehabilitation in five European countries: first results of the ACROSS project. Int J Rehabil Res 2001; 24(2): 95-101.
18. Haigh R, Tennant A, Biering-Sørensen F, Grimby G, Marincek C, Phillips S et al. The use of outcome measures in physical medicine and rehabilitation within Europe. J Rehabil Med 2001; 33(6): 273-8.
19. Pomeroy VM, Pramanik A, Sykes L, Richards J, Hill E. Agreement between physiotherapists on quality of movement rated via videotape. Clin Rehabil 2003; 17(3): 264-72.
20. Lewinter M, Mikkelsen S. Therapists and the rehabilitation process after stroke. Disabil Rehabil 1995; 17(5): 211-6.
21. Wade DT. Editorial. Describing rehabilitation interventions. Clin Rehabil 2005; 19(8): 811-8.
22. Wade DT. Measurement in neurological rehabilitation. Oxford: Oxford University Press; 1992.
23. Lennon S. Using standardised scales to document outcome in stroke rehabilitation. Physiotherapy 1995; 81(4): 200-7.
24. Duncan PW, Jorgensen HS, Wade DT. Outcome measures in acute stroke trials. Stroke 2000; 31: 1429-38.
25. Wade DT, Langton Hewer R. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry 1987; 50(2): 177-82.
26. Hobart JC, Lamping DL, Thompson AJ. Evaluating neurological outcome measures: the bare essentials. J Neurol Neurosurg 1996; 60: 127-30.
27. Salter K, Jutai JW, Teasell R, Foley NC, Bitensky J, Bayley M. Issues for selection of outcome measures in stroke rehabilitation: ICF Participation. Disabil Rehabil 2005; 27(9): 507-28.
28. Mawson SJ. Measuring physiotherapy outcome in stroke rehabilitation. Physiotherapy 1993; 79(11): 762-65.
29. Pollock A, Legg L, Langhorne P, Sellars C. Barriers to achieving evidence-based stroke rehabilitation. Clin Rehabil 2000; 14(6): 611-7.