Table 3.
Subgroup analysis of the correlation of vitamin D supplementation on depression scores in older adults.
| Number of studies (sample size) | Pooled SMD (95% confidence interval) | I2 (%) | The p-value (one group difference) | |
|---|---|---|---|---|
| Baseline serum vitamin D level | ||||
| Vitamin D insufficiency or deficiency (<30 ng/mL) | 6 (679) | −0.33 [−0.63, −0.02] | 74 | 0.03 |
| Vitamin D was adequate (> 30 ng/mL) | 2 (427) | −0.05 [−0.24, 0.14] | 0 | 0.58 |
| Vitamin D levels were not tested | 3 (20455) | 0.01 [−0.02, 0.04] | 0 | 0.39 |
| Intervention to supplement the vitamin D level | ||||
| <2000 IU/d | 5 (2773) | −0.03 [−0.10, 0.05] | 0 | 0.5 |
| ≥2000 IU/d | 6 (18788) | −0.21 [−0.39, −0.03] | 83 | 0.03 |
| The duration of follow-up | ||||
| Less than 12 Months | 7 (19561) | −0.07 [−0.18, 0.05] | 69 | 0.26 |
| 12 Months of disease | 4 (2042) | −0.06 [−0.14, 0.03] | 0 | 0.21 |
| Sex | ||||
| The study was limited to women only | 5 (2588) | −0.19 [−0.44, 0.05] | 75 | 0.11 |
| Study gender balance | 6 (18973) | 0.01 [−0.02, 0.04] | 66 | 0.51 |
| Baseline depression | 3 (475) | −0.21 [−0.65, 0.23] | 81 | 0.35 |
| Baseline health, with no confirmed disease | 7 (21135) | 0.01 [−0.02, 0.04] | 0 | 0.54 |
| Latitude | ||||
| High-latitude | 3 (1955) | −0.04 (−0.15, 0.06) | 7 | 0.40 |
| Low-latitude | 8 (19606) | −0.14 (−0.27, 0.00) | 76 | 0.04 |
| Depression scale | ||||
| PHQ-9 | 2 (16899) | −0.44 [−1.39, 0.52] | 94 | 0.37 |
| PANAS | 1 (181) | −0.09 [−0.38, 0.20] | NA | 0.54 |
| GDS | 3 (394) | −0.27 [−0.73, 0.20] | 77 | 0.26 |
| CES-D | 1 (151) | 0.08 [−0.24, 0.40] | NA | 0.61 |
| WHO-5 | 1 (183) | −0.23 [−0.52, 0.06] | NA | 0.13 |
| SF-12, SF-36 | 3 (3753) | −0.03 [−0.09, 0.04] | 0 | 0.44 |
| Sensibility analysis | ||||
| High risk studies excluded | 10 (2258) | −0.14 [−0.26, −0.01] | 73 | 0.03 |
| Studies with shorter duration (<6 months) | 7 (19561) | −0.07 [−0.18, 0.05] | 69 | 0.26 |
| Fixed effect model | 11 (21561) | 0.00 [−0.02, 0.03] | 70 | 0.88 |