Table 2.
Mixed effects model of disagreement that the risks of smoking have been exaggerated by survey wave, adjusted for gender, heaviness of smoking, age group, household income, education, ethnic group and occupational group
| Variable | Value | Odds Ratio | 95% CI |
|
|---|---|---|---|---|
| Lower | Upper | |||
| Wave | Wave 1 (ref) | 1 | ||
| Wave 2 | 0.85 | 0.74 | 0.98 | |
| Wave 3 | 1.23 | 1.04 | 1.46 | |
| Gender | Female (ref) | 1 | ||
| Male | 1.75 | 1.40 | 2.18 | |
| Heaviness of | 0 | 1 | ||
| Smoking | 1 | 0.34 | 0.23 | 0.50 |
| Index | 2 | 0.29 | 0.20 | 0.40 |
| 3 | 0.31 | 0.22 | 0.43 | |
| 4 | 0.25 | 0.16 | 0.37 | |
| 5 | 0.22 | 0.12 | 0.40 | |
| 6 | 0.11 | 0.04 | 0.31 | |
| Missing | 0.09 | 0.04 | 0.23 | |
| Age | 16–24 (ref) | 1.00 | ||
| 25–39 | 1.45 | 0.94 | 2.21 | |
| 40–55 | 0.98 | 0.64 | 1.51 | |
| 56 and older | 0.46 | 0.29 | 0.71 | |
| Household income | Under £30 000 (ref) | 1.00 | ||
| £30 000–£44 999 | 1.38 | 1.08 | 1.76 | |
| £45 000 and over | 1.36 | 1.01 | 1.83 | |
| Don't know/prefer not to say | 0.65 | 0.50 | 0.84 | |
| Educational qualification | High school or less (ref) | 1.00 | ||
| Technical, trade school, A levels, community college | 2.00 | 1.55 | 2.59 | |
| At least university degree | 2.96 | 2.30 | 3.81 | |
| Don't know/prefer not to say | 0.76 | 0.48 | 1.22 | |
| Ethnic group | White British (ref) | 1.00 | ||
| White other | 0.69 | 0.42 | 1.13 | |
| Other ethnic group | 0.39 | 0.23 | 0.65 | |
| Prefer not to say | 0.18 | 0.06 | 0.55 | |
| Occupational social group | ABC1 (ref) | 1.00 | ||
| C2DE | 0.65 | 0.53 | 0.80 | |
| Refused or unknown | 1.15 | 0.70 | 1.89 | |
| Constant | 2.20 | 1.23 | 3.95 | |
a
Number of individuals in GEE analysis =6233, number of observations =12 274.