Objective: To validate results obtained from using the Action Research Arm Test (ARAT) at tables of a common height for persons who have experienced a stroke.
Design: Each subject was tested three times with the ARAT while sitting at three different tables: a table specially designed for the test and two generally available tables similar in height to the standard table. The patients were randomly and equally assigned to three different raters and to three different tables in accordance with a counterbalanced design. All evaluations were completed within a two-day period.
Subjects: Sixty-one patients who had had only one stroke (mean age 63.3 years; median time since stroke onset 81 days; mean ARAT score administered at the standard table 33.8) participated in this study.
Results: The intraclass correlation coef”cient (ICC) for the total scores obtained using the ARAT at the different tables was 0.99, indicating very high agreement. The ICCs were also very high in each of the subscales.
Conclusion: The results of this study support the use of the ARAT at ordinary tables roughly similar in height to that specially designed for use with the ARAT.