Objective: To translate the Health Assessment Questionnaire Disability Index (HAQ) into a German version, to validate and to compare its properties with two different versions of the Hannover Functional Ability Questionnaire (HFAQ) in a German speaking population.
Methods: The test-retest reliability was tested by Pearson correlation in 32 outpatients of the Department of Rheumatology of the Medizinische Hochschule Hannover. For retesting, the questionnaire was mailed to them 1 week later. To validate the questionnaire it was administered to 110 inpatients in three different hospitals. All patients fulfilled the American College of Rheumatology 1987 revised criteria of rheumatoid arthritis (RA) or the Rome criteria of definitive inactive RA. The internal consistency was measured by Cronbach's coefficient alpha (CCA). To assess criterion validity we compared the HAQ and the two versions of the HFAQ with Keitel's test (KT) and the modified Steinbrocker classification (mSC). Construct validity was assessed by comparing these instruments with different clinical and laboratory variables. A multivariate analysis was used to identify the most important factors that are influencing the HAQ- and HFAQ-scores.
Results: Test-retest reliability of the HAQ was r = 0.94. CCA was 0.91 (HAQ), 0.90 (HFAQ-P) and 0.93 (HFAQ-PR). The KT Pearson correlation coefficients reached r = -0.73 (HAQ), r = +0.74 (HFAQ-P) and r = +0.71 (HFAQ-PR). The mSC correlated r = +0.75 (HAQ), r = -0.72 (HFAQ-P) and r = -0.70 (HFAQ-PR). The correlation coefficients of HAQ/HFAQ-P was r = -0.87 and of HAQ/HFAQ-PR r = -0.88. The correlations between other clinical and laboratory variables reached from r = +/-0.58 (pain/HAQ) to r = +/-0.11 (number of swollen joints/HFAQ-PR). In backward multiple regression analysis 59-64% of the variance of disability measured by the questionnaires was explained predominantly by pain (32-33%) and by range of motion (16-21%).
Conclusion: The German version of the HAQ presented here and the two versions of the HFAQ are reliable and valid instruments for measuring functional disability in a German-speaking population with RA. The construct measured by the HAQ and both versions of the HFAQ showed a high degree of correspondence.