Table 2.
Associations between Zinc Intake and Current Depressive Symptoms1
| Women | Men | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total Women | Stratified by SSRI Use2 | N=1,545 245 cases |
||||||||
| N=2,163 508 cases |
No n=1,893 390 cases |
Yes n=270 118 cases |
||||||||
| Zinc Intake | Median intake, mg/d | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Dietary, quartile | Women | Men | ||||||||
| 4 (High) | 13.1 | 13.3 | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| 3 | 10.9 | 11.0 | 1.48 (1.07, 2.06) | 0.02 | 1.36 (0.94, 1.95) | 2.30 (1.02, 5.17) | 0.04 | 1.41 (0.88, 2.23) | ||
| 2 | 9.3 | 9.5 | 1.62 (1.17, 2.45) | 0.004 | 1.44 (1.00, 2.07) | 0.05 | 2.99 (1.31, 6.83) | 0.01 | 1.22 (0.75, 2.00) | |
| 1 (Low) | 7.6 | 7.7 | 1.76 (1.26, 2.45) | 0.001 | 1.61 (1.12, 2.32) | 0.01 | 3.48 (1.34, 9.07) | 0.01 | 1.08 (0.66, 1.74) | |
| P-trend | 0.004 | P-trend | 0.02 | P-trend | 0.02 | P-trend | 0.80 | |||
| Supplemental | ||||||||||
| None | 0 | 0 | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| 0.1–15 mg/d | 15.0 | 15.0 | 0.89 (0.69, 1.14) | 0.94 (0.71, 1.23) | 0.79 (0.39, 1.57) | 0.003 | 0.86 (0.59, 1.28) | |||
| >15 mg/d | 50.0 | 50.0 | 0.65 (0.36, 1.17) | 0.93 (0.51, 1.68) | 0.11 (0.03, 0.46) | 0.01 | 1.68 (0.93, 3.03) | |||
| P-trend | 0.03 | P-trend | 0.58 | P-trend | 0.001 | P-trend | 0.47 | |||
| Total, quartile | ||||||||||
| 4 (High) | 26.8 | 26.8 | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| 3 | 14.8 | 14.9 | 1.13 (0.80, 1.60) | 0.94 (0.64, 1.36) | 2.90 (1.15, 7.34) | 0.02 | 0.80 (0.48, 1.32) | |||
| 2 | 10.5 | 11.0 | 1.22 (0.87, 1.71) | 0.98 (0.68, 1.42) | 3.29 (1.38, 7.86) | 0.007 | 1.12 (0.72, 1.76) | |||
| 1 (Low) | 8.7 | 8.6 | 1.47 (1.06, 2.05) | 0.02 | 1.18 (0.83, 1.69) | 4.75 (1.98, 11.4) | 0.0005 | 1.11 (0.69, 1.78) | ||
| P-trend | 0.008 | P-trend | 0.26 | P-trend | 0.0005 | P-trend | 0.51 | |||
Multivariate models adjusted for age (5-year categories), race/ethnicity, socioeconomic status, BMI (categorical), physical activity, smoking status (never, former, current), total energy intake (quintiles) and any antidepressant/antipsychotic medication use. Models for women additionally adjusted for cardiac disease and arthritis/rheumatism. Models for men additionally adjusted for diabetes, prostatitis, American Urological Association Symptom Index score for lower urinary tract symptoms, and alcohol intake (g/d). P-values for zinc intake categories are listed if P≤0.05.
The association between supplemental or total zinc intake and depression was significantly stronger among women using SSRI medications (supplemental zinc P-interaction=0.008; total zinc P-interaction=0.01). Including the interaction term between supplemental zinc and SSRIs in the model for dietary zinc among all women did not appreciably alter the association between dietary zinc and depression. No statistically significant interactions were observed for the association between dietary zinc and antidepressant medications in women or men (e.g., SSRI P-interaction=0.4).