Psychometric properties of the Tampa Scale of Kinesiophobia (TSK-11) among older people with chronic pain

Abstract

Objectives: The study aimed to test the construct validity, factor structure and reliability of the 11-item version of the Tampa Scale for Kinesiophobia (TSK-11, Swedish version) among older people (65+) with chronic pain. Design: Methodological study. Subjects: 433 participants with chronic pain (mean age 74.8, 65–98 years) completed postal questionnaires. 264 of the participants completed a test–retest assessment. Methods: Construct validity was evaluated through corrected item-total correlations. Convergent validity was analyzed by correlations with activity/activities of daily living (ADL) dependence, pain intensity and physical activity (all of which are constructs related to kinesiophobia according to fear-avoidance theories). Factor structure was tested through confirmatory factor analysis. Reliability was assessed with Cronbach’s α and test–retest reliability, analyzed by intra-class correlation coefficient (ICC) and weighted κ coefficient analysis. Results: Evidence of convergent validity was shown by significant positive correlations with activity/ADL dependence (r = 0.20) and pain intensity (r = 0.31), and a significant negative correlation with physical activity (r = −0.38). Confirmatory factor analysis showed that both one- and two factor-solutions were possible. Cronbach’s α coefficients ranged between 0.74 and 0.87. Test–retest analysis showed strong agreement regarding ICC (r = 0.75, 95% confidence interval 0.64–0.82). The weighted κ coefficients for the individual items showed fair to moderate reliability. Conclusion: The Swedish version of TSK-11 had acceptable construct validity, factor structure, and reliability and, hence, can be considered suitable for older people with chronic pain.