Abstract
Objective: To investigate the interrater reliability of the TEMPA in adults with traumatic brain injury (TBI).
Participants: Twenty adults with upper limb dysfunction after TBI who were participating in inpatient rehabilitation.
Design: The TEMPA assessment was videotaped for each participant and 5 physical therapists independently rated these video recordings.
Main outcome measure: Functional rating and speed of execution for each item of the TEMPA.
Results: Intraclass correlation coefficients (ICC) for the speed of execution and functional rating components of the TEMPA ranged from 0.898 to 1.000.
Conclusion: The excellent interrater reliability supports the use of the TEMPA in adults with TBI.
THE Test Evaluant la performance des Membres superieurs des Personnes Agees (Upper Extremity Performance Evaluation Test for the Elderly, or TEMPA) is a clinical assessment tool developed by Desrosiers and colleagues 1 for evaluating upper extremity function in people with neurosensorimotor deficits. The TEMPA incorporates performance of unilateral and bilateral real-life tasks while maintaining a standardized testing environment to measure both speed of execution and quality of movement patterns. 1 Speed of execution is timed for all tasks that the subject can perform entirely, without assistance or task modification (ie, the rater needs to make a decision about whether or not to time each task). Movement quality is assessed using functional ratings and task analysis. The functional ratings can be added to produce a single total functional rating score, or subtotals for unilateral and bilateral tasks.
Previous studies have used the TEMPA in the elderly 1-3 as well as in people with stroke, 4,5 multiple sclerosis, 6 and traumatic brain injury (TBI). 4,7 "Excellent" interrater reliability (intraclass correlation coefficient (ICC) values above 0.75 8) is reported for speed of execution of individual TEMPA tasks (ICC ranging from 0.77 to 1.00) and the total functional rating (ICC 0.98) in elderly people with a range of different living arrangements and upper extremity deficits. 1 Interrater reliability in people with stroke is also excellent for total speed of execution for all tasks combined (ICC 0.99) and for the total functional rating (ICC 0.98). 5 Task analysis ratings demonstrated lower interrater reliability, particularly for the bilateral tasks, with some ICC values falling in the "poor" range (ICC values less than 0.40 8). 5
After TBI, the reliability of the TEMPA is less clear as interrater reliability has been evaluated only in a sample of people with different neurological diagnoses (N = 32) that included some people with TBI. 7 While excellent interrater reliability was reported for the speed of execution (ICC 0.91), only 2 raters were used to establish reliability and the total time to complete all tasks, rather than the time for individual tasks, was analyzed. A more systematic evaluation of the interrater reliability of the TEMPA in the TBI population is required before routine clinical use can be justified.
The aim of this study is to investigate the interrater reliability of the functional rating and speed of execution components of the TEMPA in adults with TBI. For the functional ratings, the subtotals for unilateral and bilateral tasks were analyzed in addition to the total functional rating. The speed of execution for each task (and for each hand for unilateral tasks) as well as the judgement as to whether to time the task or not were evaluated.