Table 1. Prevalence and correlates of depression among study participants.
Variables | Depression (CES-D score ≥9) | Univariate logistic regression analysis | Multiple logistic regression analysis | |||
---|---|---|---|---|---|---|
n out of N | % | OR (95% CI)A | P-valueA | OR (95% CI)B | P-valueB | |
Type of MSM | ||||||
MB | 140 out of 200 | 70.0 | 2.73 (1.81–4.11) | <0.001 | 1.86 (1.07–3.24) | 0.029 |
General MSM | 94 out of 204 | 46.1 | 1.00 | 1.00 | ||
Age (years) | ||||||
18–24 | 113 out of 170 | 66.5 | 2.74 (1.65–4.55) | <0.001 | 2.36 (1.12–4.96) | 0.023 |
25–34 | 79 out of 134 | 59.0 | 1.98 (1.17–3.36) | 0.011 | 1.85 (0.94–3.65) | 0.076 |
35–71 | 42 out of100 | 42.0 | 1.00 | 1.00 | ||
Permanent legal residency | ||||||
Shanghai | 31 out of 82 | 38.7 | 0.36 (0.22–0.59) | <0.001 | 0.80 (0.41–1.53) | 0.494 |
Other provinces | 203 out of 322 | 63.0 | 1.00 | 1.00 | ||
Education | ||||||
Middle school or below | 88 out of 148 | 59.5 | 1.00 | 1.00 | ||
High school | 102 out of 161 | 63.4 | 1.18 (0.75–1.87) | 0.482 | 1.37 (0.82–2.26) | 0.229 |
College or above | 42 out of 93 | 45.2 | 0.56 (0.33–0.95) | 0.031 | 0.95 (0.50–1.81) | 0.883 |
Marital status | ||||||
Never married | 181 out of 302 | 60.0 | 1.00 | 1.00 | ||
Currently married | 30 out of 59 | 50.8 | 0.69 (0.40–1.21) | 0.197 | 1.13 (0.58–2.21) | 0.722 |
Divorced or widowed | 21 out of 40 | 52.5 | 0.74 (0.38–1.43) | 0.370 | 1.36 (0.61–3.03) | 0.455 |
Monthly income (RMB) | ||||||
<3000 | 139 out of 229 | 60.7 | 1.00 | 1.00 | ||
3000–4999 | 62 out of 115 | 53.9 | 0.76 (0.48–1.19) | 0.275 | 0.68 (0.40–1.16) | 0.161 |
≥5000 | 32 out of 59 | 54.2 | 0.77 (0.43–1.37) | 0.709 | 0.93 (0.47–1.84) | 0.837 |
Sexual orientation | ||||||
Gay | 145 out of 252 | 57.5 | 0.97 (0.65–1.46) | 0.884 | 1.19 (0.74–1.89) | 0.474 |
Not gay | 88 out of 151 | 58.3 | 1.00 | 1.00 | ||
Social support (SPS score) | ||||||
Mean score (s.d.) | 67.2 (6.62) | – | 0.92 (0.89–0.95) | <0.001 | 0.92 (0.89–0.96) | <0.001 |
OR, CI and P-value were obtained from univariate logistic regression analysis.
OR, 95% CI and P-value were obtained from multiple logistic regression analysis by adjusting for other potential confounding variables listed in the table.