Reliability, validity, and responsiveness of the short form 36 (SF-36) health survey questionnaire in patients with Graves' disease

Background: The SF-36 as a generic instrument has been used widely to evaluate health-related quality of life in both healthy subjects and groups of patients. However, no data can be found on its performance in patients with Graves' disease. Hence, this study aimed to assess the reliability, validity, and responsiveness of the SF-36 in a Chinese population of patients with Graves' disease. Methods: A total of 325 patients with Graves' disease completed interviews that included the SF-36. Internal consistency was measured by Cronbach's a and item-scale correlations. The validity of the SF-36 was studied by means of factor analysis and the association of this scale with sociodemographic and clinical variables. A standardized response mean was used to assess the responsiveness of the SF-36 to change. Results: Cronbach's a coefficients surpassed the 0.70 criteria for seven subscales and 0.63 for the SF subscale indicating good internal consistency. The correlation coefficients between items and the remainder of the same subscale ranged from 0.51 to 0.96, which were all higher than with other subscales. A principal components factor analysis with varimax rotation identified eight factors. Hamilton Anxiety Rating Scale (HARS) scores correlated negatively to scores at the physical functioning, general health, social functioning, vitality, and role limitations due to emotional problems SF-36 subscales, while HARS scores correlated negatively only to scores on the mental health SF-36 subscale. No correlations were found between SF-36 scores and the levels of thyroid hormones. Responsiveness to improvements in health status was acceptable overall (standardized response means 0.33-0.88). Conclusion: The Chinese (mainland) version of the SF-36 is a suitable instrument that can be used in patients with Graves' disease.