Logistic regression analysis for smoker agreement with the statement ‘tobacco companies should be more tightly regulated’ (all the results weighted and adjusted for the complex design)
| Variables | aOR (95% CI)* | ||
| Model 1 (demographics) | Model 2 (+ socio-demographics) | Model 3 (+ mental health, smoking-related beliefs and behaviours) | |
| Hosmer–Lemeshow χ2=7.29, df=6 (p=0.295)† | Hosmer–Lemeshow χ2=4.00, df=8 (p=0.857)† | Hosmer–Lemeshow χ2=8.85, df=8 (p=0.355)† | |
| Demographic | |||
| Age (35–49 vs 50+ years) | 1.00 (0.70 to 1.45) | 0.97 (0.67 to 1.41) | 0.66 (0.42 to 1.02) |
| Age (<35 vs 50+ years) | 0.91 (0.62 to 1.35) | 0.87 (0.59 to 1.30) | 0.57 (0.34 to 0.95) |
| Gender (female vs male) | 1.35 (1.00 to 1.82) | 1.33 (0.98 to 1.80) | 1.35 (0.94 to 1.95) |
| Mâori vs European/other (includes Asian) | 1.42 (1.04 to 1.94) | 1.43 (1.03 to 1.98) | 1.58 (1.08 to 2.32) |
| Pacific vs European/other (includes Asian) | 1.79 (0.98 to 3.30) | 1.63 (0.88 to 3.02) | 2.09 (0.93 to 4.70) |
| Socioeconomic‡ | |||
| Area deprivation quintiles (increasing deprivation) | – | 0.97 (0.87 to 1.10) | 1.03 (0.89 to 1.19) |
| Individual deprivation using NZiDep (any deprivation vs nil) | – | 1.04 (0.75 to 1.45) | 1.06 (0.71 to 1.58) |
| Financial stress: unable to pay any important bills on time | – | 0.72 (0.39 to 1.34) | 0.97 (0.46 to 2.02) |
| Financial stress: not spending on household essentials | – | 2.02 (1.37 to 2.98) | 1.69 (1.05 to 2.72) |
| Mental health, smoking-related beliefs and behaviours‡ | |||
| Psychological distress Kessler 10-item scale (K10) score (continuous scale) | – | – | 1.01 (0.98 to 1.05) |
| Heaviness of smoking index (alternate version) | – | – | 0.94 (0.87 to 1.02) |
| Awareness of smoking harm (7-item scale) (α=0.69) | 1.54 (1.02 to 2.31) | ||
| Concern around smoking impact on health and quality of life in the future (2-item scale) | – | – | 1.10 (0.87 to 1.39) |
| Strength of intention of quitting (4-point scale) | – | – | 1.06 (0.59 to 0.91) |
| Overall attitude to smoking (5-point scale, high score is more positive towards smoking) | – | – | 0.73 (0.59 to 0.91) |
| Self-exempting beliefs (3-item scale, high score means stronger such beliefs) (α=0.60) | – | – | 0.70 (0.55 to 0.90) |
| Awareness of social denormalisation of smoking (3-item index) (α=0.50) | – | – | 1.11 (0.99 to 1.24) |
Bolded results are statistically significant (at the p < 0.05 level).
↵* The adjusted ORs (aORs) represent those odds for agreement with the statement ‘tobacco companies should be more tightly regulated’ relative to other responses to this question. The aORs in models 1, 2 and 3 are adjusted for the demographic and key socio-demographic variables (eg, deprivation) and model 3 for mental health, heaviness of smoking and quitting intentions (see the Methods section).
↵† Not adjusted for complex sample design.
↵‡ See the Methods section and an online Methods Report 26 for further details on these measures.