Table 3.
Bivariate analysis* of ever e-cigarette use among current smokers only (N=772; 82 e-cigarette users [10.6%])
| Covariate | OR* | 95% CI for OR | p-value |
|---|---|---|---|
| Age (years) | 0.93 | (0.85, 1.02) | 0.132 |
| Male vs. Female | 1.95 | (1.22, 3.12) | 0.005 |
| Year in school | 0.155 (df=3) | ||
| Freshman (reference) | - | - | - |
| Sophomore | 1.46 | (0.77, 2.78) | 0.250 |
| Junior | 1.30 | (0.69, 2.45) | 0.425 |
| Senior/5th yr. | 0.68 | (0.32, 1.43) | 0.305 |
| Race & Ethnicity | 0.524 (df=4) | ||
| White (reference) | - | - | - |
| Asian/Pacific Islander | 2.16 | (0.45, 10.4) | 0.336 |
| Black | 0.41 | (0.05, 3.11) | 0.389 |
| Hispanic | 1.08 | (0.24, 4.80) | 0.920 |
| Other | 1.88 | (0.69, 5.10) | 0.216 |
| Mother’s education | |||
| Some college or less (reference) | - | - | - |
| College degree or higher | 1.41 | (0.88, 2.26) | 0.147 |
| Father’s education | |||
| Some college or less (reference) | - | - | - |
| College degree or higher | 1.05 | (0.66, 1.66) | 0.850 |
| Greek vs. Non-Greek | 1.22 | (0.67, 2.22) | 0.514 |
| Off-campus vs. On-campus residence | 0.93 | (0.59, 1.47) | 0.750 |
| Sensation seeking | 1.75 | (1.24, 2.48) | 0.002 |
| Lifetime hookah use: Yes vs. No | 1.48 | (0.85, 2.60) | 0.166 |
| Current binge drinking: Yes vs. No | 1.85 | (1.01, 3.36) | 0.045 |
| Current marijuana use: Yes vs. No | 1.60 | (0.98, 2.61) | 0.060 |
| Lifetime other illegal drug use: Yes vs. No | 1.74 | (1.10, 2.76) | 0.019 |
| E-cigarette harm perception | <0.001 (df=3) | ||
| As harmful as cigarette (reference) | - | - | - |
| Less harm than cigarette | 1.30 | (0.70, 2.42) | 0.412 |
| More harmful than cigarette | 3.67 | (1.18, 11.4) | 0.025 |
| Don’t know | 0.23 | (0.10, 0.52) | <0.001 |
| Cigarette smoking quit intentions Not thinking of quitting vs. Any intentions |
1.13 | (0.68, 1.87) | 0.633 |
| Type of Institution | |||
| Public (reference) | - | - | - |
| Private | 1.59 | (0.85, 2.96) | 0.144 |
OR is odds ratio (unadjusted for other covariates); Each variable is the only covariate present in each bivariate mixed-effects logistic regression model presented above.