Table 4.
Covariates associated with daily tobacco use in Kenya
| Daily tobacco use | Crude Odds Ratioa | Adjusted Odds Ratioa | ||
|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Sex | ||||
| Female | 1.00 | 1.00 | ||
| Male | 7.16 (5.52, 9.28) | < 0.001 | 7.48 (5.34, 10.48) | < 0.001 |
| Age group | ||||
| 18–29 | 1.00 | 1.00 | ||
| 30–39 | 1.79 (1.38, 2.31) | < 0.001 | 1.35 (0.87, 2.08) | 0.181 |
| 40–49 | 2.33 (1.77, 3.06) | < 0.001 | 1.39 (0.86, 2.23) | 0.176 |
| 50–59 | 2.81 (2.08, 3.81) | < 0.001 | 2.57 (1.61, 4.11) | < 0.001 |
| 60–69 | 3.12 (2.16, 4.52) | < 0.001 | 1.36 (0.74, 2.51) | 0.324 |
| Education level | ||||
| No formal education | 1.00 | 1.00 | ||
| Primary complete | 0.77 (0.60, 1.00) | 0.050 | 0.28 (0.18, 0.43) | < 0.001 |
| Secondary and above | 0.35 (0.26, 0.47) | < 0.001 | 0.12 (0.07, 0.20) | < 0.001 |
| Residence | ||||
| Urban | 1.00 | 1.00 | ||
| Rural | 1.11 (0.91, 1.36) | 0.288 | 0.63 (0.46, 0.85) | 0.002 |
| Occupation | ||||
| Unemployed | 1.00 | 1.00 | ||
| Employed | 1.14 (0.93, 1.38) | 0.200 | 0.58 (0.45, 0.76) | < 0.001 |
| Ever used alcohol | ||||
| No | 1.00 | 1.00 | ||
| Yes | 5.35 (4.28, 6.69) | < 0.001 | 2.54 (1.85, 3.49) | < 0.001 |
| Episodic alcohol drinking | ||||
| No alcohol | 1.00 | 1.00 | ||
| Binge drinking | 8.03 (6.49, 9.93) | < 0.001 | 0.77 (0.30, 1.96) | 0.588 |
| Non-heavy drinking | 3.41 (2.34, 4.97) | < 0.001 | 0.52 (0.06, 4.59) | 0.560 |
| Wealth band | ||||
| Poorest | 1.00 | 1.00 | ||
| Second | 0.76 (0.58, 1.00) | 0.054 | 0.91 (0.58, 1.41) | 0.665 |
| Middle | 0.76 (0.57, 1.01) | 0.061 | 0.56 (0.33, 0.93) | 0.026 |
| Fourth | 0.70 (0.53, 0.94) | 0.018 | 0.69 (0.41, 1.15) | 0.150 |
| Richest | 0.38 (0.28, 0.52) | < 0.001 | 0.47 (0.26, 0.86) | 0.014 |
| Marital status | ||||
| Not married | 1.00 | 1.00 | ||
| Married | 1.10 (0.87, 1.41) | 0.429 | 0.73 (0.53, 1.02) | 0.066 |
| Formerly married | 2.03 (1.48, 2.77) | < 0.001 | 1.41 (0.91, 2.17) | 0.120 |
Key: aAll sociodemographic variables (except occupation) were included in final regression models if found to be statistically significant. This was true except for occupation given the original coding of the variable in the survey that was not felt to be meaningful for our study. The variable sex was maintained in the three models, even though it was only found to have a statistically significant relationship with daily tobacco use given the hypothesized importance of the role of sex on tobacco use