DISABILITY AND REHABILITATION, 1995; VOL. 17, NO. 1, 10-14 The Functional Independence Measure: zy a comparative validity and reliability study D. KIDD, G. STEWART, J. BALDRY, J. JOHNSON, D. ROSSITER, A. PETRUCKEVITCH and A. J. THOMPSON zyx Accepted zyxwvutsr for publication: October 1993 Keywords Functional Independence Measure, Barthel In- dex, neurological disability Summary The majority of measurement scales used to evaluate out- come in rehabilitation are ordinal in nature and consequently statistically valid assessments of change are difficult to make. The Functional Independence Measure (FIM) can be weighted to possess interval properties, potentially allowing more ac- curate analysis of change. In this study the FIM was compared to the Barthel Index (BI) to determine its validity, reliability and ease of use in two groups of zyxwvut 25 patients undergoing neurorehabilitation. The FIM was considered to be more valid than the BI, and equally reliable in the assessment of disabili- ty. When the two disability scores were compared using sub- jective and objective assessment the agreement between them was comparable, although neither was high. Introduction In order to evaluate accurately the efficacy of any therapeutic practice it is necessary to use appropriate measuring tools which are both valid and reliable.'J This is of particular importance in rehabilitation, which is labour-intensive and is often carried out over a pro- longed period of time. The difficulties inherent in clinical evaluation, particularly in relation to the assess- ment of change, have resulted in the development of a wide variety of measurement scales which attempt to give a numerical notation to the clinical situation. Such measurement of clinical change is crucial to the assess- ment of outcome, benefit and cost-effectiveness. Most scales used are ordinal in nature; that is they show a rank order, but it cannot be assumed that the steps on the scale are equal. It follows that it cannot be zyxwvu ~~ Correspondence to: Dr A. J. Thompson, Neurorehabilitation Unit, Institute of Neurology, National Hospital, Queen Square, London WClN 3BG, UK. concluded that a score of zyx 10 is twice that of 5, and this inequality considerably limits statistical analysis and in- terpretation. Hence it is not correct to use the change in an ordinal scale to produce a numerical notation of cor- responding clinical change unless all patients start at the same point. Intervals scales, on the other hand, are ranked in equal proportions, allowing comparison of change, even when starting at different points in the scale, and therefore providing a more objective way of quantifying functional change in groups of disabled patients. The Barthel Index3 (BI) (Appendix 1) is an ordinal scale with a range of 0 to 20 which has been used as a measure of disability in clinical research for many years. Previous studies have stated that it is valid, relatively reliable but not very sensitive to ~ h a n g e . ~ - ~ The Func- tional Independence (FIM) (Appendix 2) was devised in 1984 by a workshop developed by the American Congress for Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation to be used as a universal assessment tool in the Uniform Data System for Medical Rehabilitation (UDS'O). It was developed in order to provide a more comprehensive assessment of disability not only of self- care activities and mobility but also communication and cognitive function. It has a range of 13-126. Initially it, too, was ordinal but has since undergone several refinements - the most notable of which is the weighting of individual scores using Rasch analysis,' which utilizes log odds ratios of probability to make the intervals between individual steps the same, and thus allows the FIM to be used as an interval scale.12 It com- prises 18 items; 13 motor items (self-care and locomo- tion) and five cognitive items (communication and cognition), each of which is scored on a seven-point scale. The motor and cognitive scores have been separated, and each score now has a range of 0-100. The interrater reliability is said to be high.I3 Giving the FIM interval properties allows it to be used 0963-8288195 $10-00 zyxwvuts 0 1995 Taylor & Francis Ltd. Disabil Rehabil Downloaded from informahealthcare.com by Technische Universitaet Muenchen Universitaetsbibliothek on 08/01/13 For personal use only.