To Identify the Minimal Number of Outcome Measures for a Rehabilitation Programme For

TO IDENTIFY THE MINIMAL NUMBER OF OUTCOME MEASURES FOR A REHABILITATION PROGRAMME FOR IN-PATIENTS WITH RHEUMATOID ARTHRITIS

Ruby Ho, Desmond Yeung, Carmen Ho, Jo Wong, Josephine Ang, S.K. Lau, Amanda Mak, Ellen Cheng, Leonard Li

(i)Introduction:The Rehabilitation Programme for In-patients with Rheumatoid Arthritis (RA) in MacLehose Medical Rehabilitation Centre was first launched in June 1998. Pilot studies done in the past 2 years indicated that the programme was effective for clients with RA. The purpose of this study is to identify the minimal number of outcome measures for future studies on the same programme.

(ii)Method:Test items demonstrating significant improvement in the pilot studies were pooled together to form the database of the present study. Factor analysis was used to categorise the test items. The item-to-item correlation of the variables within each factor was then studied. Clinical concerns were also taken into consideration in the final selection of test items.

(iii)Summary of Results:Twenty-five factors were identified in the pooled data. Using the 2.4 as a cutoff for the eigenvalue, 4 factors were selected accounting for a cumulative 67.3% of the variance in the data. Having selected the 0.65 as a cutoff for the factor loading, the number of variables retained was found to be 21. Nine variables were found in “Factor One” namely walking endurance, 50 feet walk time, AIMS2 (mobility), AIMS2 (self-care), AIMS2 (household), AIMS2 (work), self-efficacy (function), grip Right and grip Left. Four variables retained in “Factor Two” namely AIMS2 (support from family), self–efficacy (pain), self–efficacy (other symptoms) and Visual Analogue Scale (pain). There were 4 variables left in “Factor Three” namely role performance, role satisfaction, tender joint count and swollen joint count. Four variables belonged to “Factor Four” namely AIMS2 (arm function), AIMS2 (pain), AIMS2 (mood) & Visual Analogue Scale (fatigue). The item-to-item correlation of the variables within each factor studied further reduced the total number of variables to 13.

(iv)Conclusion:Using factor analysis and clinical decision as a guideline for selecting outcome measures, 25 test items showing significant improvement in the pilot studies were now reduced to half. This reduction in number of outcome measures significantly shortens team members’ time in data collection for future studies on the concerned in-patient rehabilitation programme, thus facilitating further studies on the programme.