Table 3.
Effect of the treatment on depressive symptoms, analyzed with logistic regression (total sample) and ANCOVA (subsample with symptoms) (intention-to-treat).
| Baseline | Two-Year Follow-up | Model 1 a | Model 2 (Fully Adjusted) b | |||||
|---|---|---|---|---|---|---|---|---|
| Logistic regression (total sample, N = 2588) | N with GDS-15 ≥ 5 (%) | N with GDS-15 ≥ 5 (%) | OR c (95% CI) | p | OR c (95% CI) | p | ||
| 1.1 (0.8, 1.5) | 0.56 | 1.1 (0.8, 1.5) | 0.45 | |||||
| B-vitamins | 101 (7.0) | 112 (8.6) | ||||||
| Placebo | 99 (6.8) | 111 (8.5) | ||||||
| ANCOVA (subsample GDS-15 ≥ 5, N = 161) | Median GDS-15 score (IQR) | Median GDS-15 score (IQR) | Mean change (95% CI) | F | p | Mean change (95% CI) | F | p |
| 0.57 | 0.45 | 0.36 | 0.55 | |||||
| B-vitamins | 6 (5–8) | 5 (3–7) | 1.4 (0.7, 2.2) | 1.5 (0.7, 2.2) | ||||
| Placebo | 6 (5–8) | 4 (3–6) | 1.8 (1.1, 2.5) | 1.8 (1.1, 2.5) | ||||
a Adjusted for baseline values of the respective outcome variable; b additionally adjusted for age, sex, homocysteine, and study center; c odds ratio of having clinically relevant depressive symptoms (GDS-15 ≥ 5) after two years. GDS-15: Geriatric Depression Scale, 15-item version.