Development and validation of self-management scale for tuberculosis patients

Background Tuberculosis remains a major threat to global public health. Regarding its control, directly observed therapy is not suitable as a global strategy for all tuberculosis patients. Self-management may be an important patient-centered tuberculosis case management supplement to directly observed therapy. However, there is currently no well-established instrument for measuring the self-management of tuberculosis patients. This study aimed to develop and validate a self-management scale for tuberculosis patients. Methods We developed an initial scale based on the tuberculosis health promotion indicators framework developed by our research group. After item analysis and two rounds of exploratory factor analysis, a final version of the scale was developed. A survey of 462 tuberculosis patients was conducted to develop and validate this scale. Cronbach’s α and intraclass correlation coefficients were used to assess reliability, and Pearson’s correlation coefficients were used to evaluate content validity. Fit indices, convergent validity, and discriminant validity were evaluated using confirmatory factor analysis to determine the construct validity of the scale. Results The scale was composed of 17 items in three dimensions (“adherence to treatment behavior,” “transmission prevention behavior,” and “supportive therapy behavior”). These three dimensions explained 76.60% of the variance. Cronbach’s α of the scale was 0.905, and the intraclass correlation coefficient was 0.897. Additionally, Pearson’s correlation analysis showed that each item was strongly correlated with the dimension to which it belonged (r = 0.849–0.915, p < 0.01). Most fit indices (Comparative Fit Index, Normed Fit Index, Incremental Fit Index, Goodness of fit index) reached the recommended threshold, and the average variance extracted values of the three dimensions were higher than 0.5. The values of the square root of the average variance extracted within each dimension were greater than the correlation between dimensions, and all heterotrait–monotrait values were below 0.85. Conclusions The self-management scale for tuberculosis patient demonstrated good reliability and validity and could be used as an instrument to evaluate the self-management of patients. Additionally, it could be used to develop evidence-based self-management interventions and evaluate those interventions.