Table 2.
Logistic regression analysis of “interest among smokers in trying smokeless tobacco products” by sociodemographic and other characteristics (all results weighted and adjusted for the complex design)
| Variables |
AOR (95% CI)a |
|||
| Model 1 (demographics) | Model 2 (+ sociodemographics) | Model 3 (+ smoking beliefs and behavior) | Model 4 (Model 3 but with each sociodemographic variable considered in isolation) | |
| Demographic | ||||
| Age (years) | ||||
| 35–49 vs. <35 | 1.20 (0.83–1.74) | 1.16 (0.80–1.69) | 1.15 (0.75–1.75) | Variableb |
| 50+ vs. <35 | 1.49 (0.99–2.25) | 1.47 (0.97–2.24) | 1.37 (0.85–2.22) | Variableb |
| Gender | ||||
| Women vs. men | 0.99 (0.72–1.36) | 1.05 (0.77–1.45) | 0.93 (0.65–1.34) | Variableb |
| Māori vs. European | 1.71 (1.23–2.37) | 1.58 (1.13–2.22) | 1.51 (1.04–2.20) | Variableb |
| Pacific vs. European | 1.32 (0.74–2.38) | 1.11 (0.61–2.01) | 1.17 (0.52–2.61) | Variableb |
| Asian vs. European | 1.57 (0.67–3.67) | 1.87 (0.84–4.15) | 2.47 (0.96–6.37) | Variableb |
| Sociodemographicc | ||||
| Area deprivation quintiles (increasing deprivation) | — | 1.08 (0.95–1.23) | 1.06 (0.91–1.23) | 1.05 (0.91–1.21) |
| Individual deprivation using NZiDep (any deprivation vs. nil) | — | 0.85 (0.60–1.19) | 0.76 (0.53–1.11) | 0.80 (0.56–1.15) |
| Financial stress: Unable to pay any important bills on time | — | 1.21 (0.70–2.10) | 1.08 (0.58–2.00) | 1.08 (0.59–1.97) |
| Financial stress: Not spending on household essentials | — | 1.50 (1.04–2.15) | 1.12 (0.73–1.73) | 1.11 (0.73–1.69) |
| Smoking behavior and beliefs | ||||
| Heaviness of smoking index (alternate version)d | — | — | 1.03 (0.95–1.12) | Variableb |
| Concern around smoking impact on health and quality of life in the future (2-item scale)e | — | — | 1.44 (1.17–1.78) | Variableb |
| Strength of intention of quitting (4-point scale)f | — | — | 0.99(0.82–1.21) | Variableb |
Note. AOR = adjusted odds ratio.
The AORs represent those saying “yes” to interest (see Table 1) relative to “no” and excluding those in the “maybe/don't know” group (though those who “contest the proposition” were included in the “no” group). The AORs in Models 2, 3, and 4 are adjusted for the demographic and key sociodemographic variables (i.e., deprivation), Model 3 for smoking behavior and beliefs, and Model 4 considers each SES measure and financial stress measure separately (see the Methods section).
In this model, each sociodemographic variable was considered in isolation from the other deprivation and financial stress variables. The results for the other variables are not shown, as these varied for each analysis.
See Methods section for further details on these measures.
The “Heaviness of Smoking Index” has been developed by others, and we used the “alternative version” utilized by others (Borland et al., 2004). This is calculated as the square root of the daily cigarette consumption minus the natural logarithm of time to first cigarette of the day. The specific equations are detailed in an online Methods Report (Wilson, 2009).
The “concern around smoking impact” scale was based on the following two questions: (i) “How worried are you, if at all, that smoking WILL damage YOUR health in the future?” and (ii) “How worried are you, if at all, that smoking WILL lower your quality of life in the future?” For both these questions, the response options were: “Not at all worried,” “A little worried,” “Moderately worried,” “Very worried,” “Refused,” and “Can’t Say.” The Cronbach's alpha for this scale was relatively high (α = 0.78).
The “concern strength of intention of quitting” scale has been used by other ITC Project workers (Young et al., 2007). The question was: “Now we would like to ask you some questions on any thoughts you might have had about quitting smoking. IF you decided to give up smoking completely in the next 6 months, how sure are you that you would succeed?” The response options were: “Within the next month”; “Within the next 6 months”; “Sometime in the future, beyond 6 months”; “Or are you not planning to quit”; “Refused”; and “Can't Say.”