Comparison of the reliability and validity of depression scales for systemic lupus erythematosus and analysis of depression-related risk factors
10.3969/j.issn.1674-8115.2019.03.015
- Author:
Jiao-Yu LI
1
Author Information
1. Department of Rheumatology, Renji Hospital South Campus, Shanghai Jiao Tong University School of Medicine
- Publication Type:Journal Article
- Keywords:
Depression;
Reliability;
Risk factor;
Scale;
Systemic lupus erythematosus;
Validity
- From:
Journal of Shanghai Jiaotong University(Medical Science)
2019;39(3):304-308
- CountryChina
- Language:Chinese
-
Abstract:
Objective • To estimate the reliability and validity of Beck Depression Inventory (BDI), Self-rating Depression Scale (SDS) and Center for Epidemiologic Studies Depression Scale (CES-D) that used for systemic lupus erythematosus (SLE) to choose a more suitable depression scale, and analyze the risk factors for depression in SLE. Methods • A total of 259 questionnaires were completed by SLE patients with their informed consent in the early stage. Internal consistency, test-retest reliability, construct validity, consistency and correlation of depression outcomes were analyzed. The most suitable scale was used in 70 SLE patients. The results of depression screening and clinical information were collected to analyze the risk factors for SLE with depression. Results • A total of 240 valid questionnaires were collected. The Cronbach's α coefficients of BDI, SDS, and CES-D were 0.899, 0.884, and 0.711, respectively. The total scores of the three scales were correlated with every item (P<0.05). The test-retest coefficients of BDI, SDS, and CES-D were 0.912, 0.635, and 0.728, respectively (P<0.01). The result of validity analysis showed that the KMO (Kaiser-Meyer-Olkin) values of BDI, SDS, and CES-D were 0.888, 0.895, and 0.918, respectively. The total variation of main factor interpretation of BDI, SDS, and CES-D was 63.13%, 57.09%, and 57.89%, respectively. The results of BDI showed 117 patients (48.25%) with depression, while SDS showed 154 patients (64.16%) with depression and CES-D showed 140 patients (58.33%) with depression. The correlation coefficients of the three scales were 0.939-0.971 (P<0.01). κ Values of all three scales were greater than 0.4 (P<0.01). Younger than 30, unemployment and renal involvement were independent risk factors for SLE with depression. Conclusion • With better internal consistency, test-retest coefficients, good brief framework and well-understood items, BDI seems to be a more suitable scale for SLE. Unemployed, renal involved and young SLE patients are more likely to suffer from depression.
