Table 2. Hazard ratios of suicides by depressive symptoms in Korean middle-aged men (n = 10,238).a .
| Unadjusted analysis | Adjusted analysisb | |||||
|---|---|---|---|---|---|---|
| Categories of depressive symptoms (score) | No. of suicidesc | p-value | HR (95% CI) | p-value | HR (95% CI) | |
| Total BDI score (0–63) | Five-score increase | 41 | <0.001 | 1.4 (1.2–1.5) | 0.01 | 1.2 (1.1-1.5) |
| 2-group analysisd | No-to-moderate (0–30) | 14 | 1.0 (Reference) | 1.0 (Reference) | ||
| Severe (31–63) | 27 | <0.001 | 5.4 (2.8–10.6) | 0.004 | 3.4 (1.5–7.7) | |
| 5-group analysisd | No (0–13) | 3 | 1.0 (Reference) | 1.0 (Reference) | ||
| Mild (14–21) | 6 | 0.51 | 1.6 (0.4–6.8) | 0.69 | 1.4 (0.3–6.1) | |
| Moderate (22–30) | 5 | 0.70 | 1.4 (0.3–6.3) | 0.94 | 0.9 (0.2–5.0) | |
| Severe (31–39) | 12 | 0.009 | 5.4 (1.5–19.4) | 0.12 | 3.3 (0.7–14.4) | |
| Extreme (40–63) | 15 | <0.001 | 10.2 (3.0–35.0) | 0.05 | 4.6 (1.0–21.5) | |
| Trend teste | 41 | <0.001 | 1.9 (1.5–2.5) | 0.009 | 1.6 (1.1–2.2) | |
Notes.
- BDI
- Beck Depression Inventory
- CI
- confidence interval
- HR
- hazard ratio
a
Hazard ratios were calculated using a Cox proportional hazards model over multiple imputed data.
b
Adjusted for age at enrollment, smoking status, drinking status, body mass index, self-rated health, marital status, educational status, household monthly income and participant group.
c
Number of suicides may not match those of the other groups of depressive symptoms due to rounding of averages over multiple imputed data.
d
Cut-off score based on quartiles and the last decile.
e
Five categories of Depressive symptoms were analyzed as ordinal variables.