Table 2.
Multivariable logistic regression analysis of factors independently associated with recent cross-border injection drug use among people who inject drugs in San Diego (N=471)
| Variable1 | Adjusted Odds Ratio (95% CI) | p-value |
|---|---|---|
| Age (per 1-year increase) | 0.95 (0.93–0.98) | <0.001 |
| Identify as Hispanic or Latino/a2 | 0.43 (0.23–0.82) | 0.010 |
| Concerned about risk of violence when travelling to Mexico2 | 0.57 (0.33–0.99) | 0.046 |
| Injected cocaine ≥ weekly (last 6m)2 | 3.98 (1.03–15.44) | 0.046 |
| Ever lived in Mexico2 | 2.50 (1.39–4.50) | 0.002 |
| Know at least one PWID who lives in Mexico2 | 6.09 (3.35–11.07) | <0.001 |
1
Variables not included in model due to collinearity: Duration of injecting (collinear with age), duration of residence in San Diego (collinear with age), and have family and/or friends in Mexico (collinear with having ever lived in Mexico);
2
Yes vs. no
Missing data excluded
Abbreviations: PWID = people who injects drugs
Hosmer-Lemeshow Goodness-of-fit p=0.4198