Skip to main content
. 2021 May 25;26(1):429–441. doi: 10.1007/s11325-021-02405-0

Table 4.

Subgroup analysis of the pooled results in depression levels between SLE patients with good and poor sleep quality

Variables No. of studies Pooled SMD (95% CI) P value Heterogeneity
I2 (%) P value
Corticosteroids use
   < 100% 2  − 0.11 (− 2.05, − 0.18) 0.020 81.6 0.020
  100% 2  − 1.55 (− 2.67, − 0.42) 0.007 73.6 0.052
Measurement of depression
  CES-D 2  − 1.79 − 2.34, − 1.24)  < 0.001 0 0.341
  HAM-D 1  − 1.05 (− 1.61, − 0.49)  < 0.001 0 -
  HADS-D 1  − 0.66 (− 1.13, − 0.20) 0.005 0 -
Diagnostic criteria of SLE
  1982 and 1997 ACR 1  − 0.66 (− 1.13, − 0.20) 0.005 0 -
  1997 ACR 2  − 1.31 (− 1.86, − 0.75)  < 0.001 39.9 0.197
  1982 ACR 1  − 2.21 (− 3.24, − 1.18)  < 0.001 0 -
Region
  Asia 3  − 1.07 (− 1.60, − 0.54)  < 0.001 63.4 0.065
  Africa 1  − 2.21 (− 3.24, − 1.18)  < 0.001 0 -
Quality
  High risk 2  − 0.82 (− 1.19, − 0.45)  < 0.001 6.9 0.300
  Low risk 2  − 1.79 (− 2.34, − 1.24)  < 0.001 0 0.341

CES-D, Center for Epidemiological Studies Depression Scale; HAM-D, Hamilton Rating Scale for Depression; HADS-D, Hospital Anxiety and Depression Scale-Depression; ACR, American College of Rheumatology; SMD, standardized mean difference; CI, confidence interval