Validation of the Russian-language version of Ankylosing Spondylitis Quality of Life (ASQoL) questionnaire

Abstract

Objective: validation of the Russian-language version of Ankylosing Spondylitis Quality of Life (ASQoL) specific questionnaire for patients with ankylosing spondylitis (AS). Materials and methods. Тhe study included 100 patients with a definite diagnosis of AS according to the modified New York criteria of 1984. Patients were mostly men (58.5%), average age – 38.6±10 years, disease duration – 10.3±7.0 years, median disease activity by ASDAS-CRP – 2.32 [1.45; 2.94]. At the initial visit and after 3 months, all patients underwent a standard rheumatological examination, with assessment of the tender joint count (TJC) and swollen joint count (SJC) out of 44 joints assessed, the number of inflamed entheses using MASES index, spinal mobility using BASMI index. Assessment of patient reported outcomes included an assessment of spinal pain, night spinal pain, spinal pain during the day using a numerical rating scale, patient’s global assessment of health, assessment of functional status using BASFI index and health related quality of life (HRQoL) using EQ-5D, SF-36 and ASQoL questionnaires. To determine reliability of ASQoL questionnaire, a test-retest analysis and determination of internal consistency by calculating the Cronbach's α coefficient were used. The reliability of the questionnaire was analyzed in 39 patients. To assess the validity of the ASQoL questionnaire, i.e. its ability to reliably measure its inherent characteristics, construct and criterion validity were determined. The sensitivity of the questionnaire was assessed after 3 months during therapy in 79 patients. Results and discussion. According to the test-retest analysis, there was no difference between the initial assessment and re-assessment after 3 days (p>0.05), the Cronbach α coefficient was 0.884, indicating high reliability. Construct validity analysis was carried out using factor analysis and the “known groups” method. 3 main factors were identified: pain, physical health, emotional health. The “known groups” method showed that in patients with high disease activity according to ASDAS-CRP and BASDAI, HRQoL was significantly worse than in patients with low disease activity (р<0,05). To calculate criterion validity, the relationship of ASQoL to “external criteria” and interchangeability with general questionnaires SF-36 and EQ-5D were evaluated by a correlation analysis. It was shown that this questionnaire is highly interchangeable. To determine sensitivity of ASQoL, its changes were analyzed in relation to the effect of therapy after 3 months. Significant changes in the subscales of the questionnaire were observed in patients who achieved low disease activity according to ASDAS-CRP during therapy, reflecting its good sensitivity. Conclusion. The ASQoL has good psychometric properties and is able to reflect changes in a patient's health status over time along with disease activity.