TABLE 4.
Subgroup analysis of depression for the highest vs. lowest category of dietary beta-carotene intake.
| Stratification | Number of studies | Pooled RR | 95% CI | P-value | Heterogeneity |
| All studies | 8 | 0.63 | 0.55, 0.72 | P < 0.001 | P = 0.31; I2 = 15% |
| Geographical region | |||||
| Asia | 6 | 0.64 | 0.54, 0.76 | P < 0.001 | P = 0.22; I2 = 27% |
| Non-Asia | 2 | 0.62 | 0.50, 0.76 | P < 0.001 | P = 0.28; I2 = 16% |
| Exposure assessment | |||||
| FFQ | 6 | 0.64 | 0.54, 0.76 | P < 0.001 | P = 0.22; I2 = 27% |
| Recall method | 2 | 0.62 | 0.50, 0.76 | P < 0.001 | P = 0.28; I2 = 16% |
| Sex | |||||
| Male | 1 | 0.36 | 0.13, 1.00 | / | / |
| Female | 6 | 0.65 | 0.54, 0.78 | P < 0.001 | P = 0.23; I2 = 28% |
| Population | |||||
| Adolescent | 3 | 0.46 | 0.33, 0.64 | P < 0.001 | P = 0.82; I2 = 0% |
| Middle aged and elderly | 4 | 0.67 | 0.58, 0.78 | P < 0.001 | P = 0.42; I2 = 0% |
| Sample size | |||||
| < 1,000 | 4 | 0.46 | 0.34, 0.62 | P < 0.001 | P = 0.95; I2 = 0% |
| > 1,000 | 4 | 0.68 | 0.59, 0.80 | P < 0.001 | P = 0.37; I2 = 5% |
| Adjustment of BMI | |||||
| Adjusted | 6 | 0.62 | 0.54, 0.72 | P < 0.001 | P = 0.22; I2 = 28% |
| Unadjusted | 2 | 0.68 | 0.46, 1.02 | P = 0.06 | P = 0.33; I2 = 0% |
| Adjustment of energy intake | |||||
| Adjusted | 5 | 0.61 | 0.52, 0.71 | P < 0.001 | P = 0.27; I2 = 22% |
| Unadjusted | 3 | 0.69 | 0.54, 0.89 | P = 0.003 | P = 0.34; I2 = 8% |