Table 3. Factors associated with WCRB for ESCC-risk behaviors and poor hygiene, 2013*.
| Variable | WCRB for ESCC-risk behavior a | WCRB for poor hygiene | ||||
|---|---|---|---|---|---|---|
| N = 233 (%) b | Univariate OR c | Multivariate OR c | N = 381 (%) | Univariate OR | Multivariate OR | |
| Age (continuous) | - | 0.98 (0.96–1.01) | 0.99 (0.96–1.01) | - | 0.99 (0.97–1.01) | 0.99 (0.97–1.02) |
| Gender | ||||||
| Female | 66 (28.3) | Ref | Ref | 258 (67.7) | Ref | Ref |
| Male | 167 (71.7) | 0.71 (0.40–1.29) | 0.81 (0.42–1.55) | 123 (32.3) | 1.14 (0.73–1.77) | 0.89 (0.54–1.46) |
| SES | ||||||
| Low | 24 (10.3) | Ref | NA | 43 (11.3) | Ref | Ref |
| Middle | 165 (70.8) | 0.77 (0.31–1.90) | NA | 303 (79.5) | 1.69 (0.89–3.21) | 1.46 (0.74–2.86) |
| High | 33 (14.2) | 0.85 (0.29–2.51) | NA | 25 (6.6) | 4.19 (1.33–13.21) | 3.31 (0.97–11.26) |
| Unknown | 11 (4.7) | - | - | 10 (2.6) | - | - |
| Exposure of risky behavior | ||||||
| Low | 138 (59.2) | Ref | Ref | 165 (43.3) | Ref | Ref |
| High | 95 (40.8) | 0.35 (0.20–0.61) | 0.38 (0.21–0.68) | 211 (55.4) | 0.54 (0.35–0.82) | 0.54 (0.33–0.87) |
| Unknown | 0 (0) | - | - | 5 (1.3) | - | - |
| Previously diagnosed diseases | ||||||
| No | 105 (45.1) | Ref | NA | 212 (55.6) | Ref | NA |
| Yes | 128 (54.9) | 0.73 (0.42–1.27) | NA | 169 (44.4) | 0.83 (0.55–1.26) | NA |
| Score on the “Faith in People” Scale | ||||||
| 0–2 | 148 (63.5) | Ref | Ref | 247 (64.8) | Ref | Ref |
| 3–5 | 85 (36.5) | 0.70 (0.40–1.23) | 0.62 (0.34–1.11) | 134 (35.2) | 0.58 (0.38–0.89) | 0.51 (0.32–0.81) |
| Behavioral change of surrounding people | ||||||
| No | 135 (57.9) | Ref | Ref | 268 (70.3) | Ref | Ref |
| Yes | 98(42.1) | 1.63 (0.93–2.87) | 1.71 (0.95–3.07) | 113 (29.7) | 2.47 (1.50–4.05) | 2.25 (1.32–3.81) |
| Family history of ESCC | ||||||
| No | 182 (78.1) | Ref | NA | 296 (77.7) | Ref | NA |
| Yes | 51 (21.9) | 0.77 (0.40–1.51) | NA | 85 (22.3) | 0.78 (0.48–1.28) | NA |
* Among 410 participants who were engaged in the behavior. The backward-selection method with a significance threshold of 0.1 was used to identify variables included in the final multivariate models. Confidence intervals that do not overlap the null value of 1 were shown in bold.
a ESCC-risk behavior included smoking, alcohol consumption and risky dietary behavior.
b One participant might contribute to multiple observations in these frequencies.
c The GEE logistic-regression model was used.