The reliability of the Functional Assessment Measure (FIM+FAM) is an important issue with its increased use in the measurement of neurological disability and rehabilitation outcome. Although the Motor items have good reliability ratings, the Cognitive items are more difficult to complete and their reliability is not as good. This study tests the suggestion that this might be due to the Cognitive items being more abstract. A keyword from each of four Motor items was compared with a keyword from four Cognitive items. Abstractness was measured by measuring the 'imageability' of each keyword. The Motor items were found to have a significantly higher mean imageability rating than the Cognitive items. Thus, there is support for the suggestion that abstractness contributes to the poorer reliability of the Cognitive items. These results led to the proposal that the reliability of the Cognitive items might be improved by various methods of increasing the tangibility of these measures (e.g. subdivision of broad categories of disabilities, enhancing item descriptions, training raters to increase their recognition of relevant observations, and using specific assessment tasks to elicit relevant behaviours).

Training in administration of the NART improves raters' accuracy and use of the pronunciation guide. This offers an alternative to the suggestion of improving the NART's reliability by replacing lower reliability words and therefore would avoid the need to re-standardize a modified test.

This article summarises experience of evaluation gained over 9 years of providing cognitive-behavioural rehabilitation . Measures of impairment are not particularly useful as indicators of outcome but do contribute to rehabilitatio n planning. Measurement of disability is central to rehabilitatio n but current measures have a limited ability to assess cognitive disability despite their relative strength in assessing physical problems. Goal setting is an integral part of the rehabilitatio n process but has limited utility in the evaluation of outcome. Although handicap is important as an index of outcome, this is difficult to measure whilst the person is resident in a rehabilitation centre. The use of modified measures and methods of assessment have helped to compensate for some of the limitations . However, new or revised measures are needed to allow repeated measurement and to enable satisfactory assessment of change in cognitive disabilities.There are several aspects of a service that influence the selection of methods for assessment. These include historical factors, the design and aims of the programme, the constitution of the rehabilitation team, and of course, the nature of the patients' difficulties and goals.Our Brain Injury Rehabilitation Service was initially set up at Ticehurst in 1989 as an inpatient and day patient unit for those with cognitive and Requests for reprints should be sent to Michael Oddy, Ticehurst House Hospital, Ticehurst, Wadhurst, East Sussex TN5 7HU, UK.

No abstract

No abstract