Table 3.
Multivariable logistic regression models predicting odds of holding supportive attitudes toward tobacco control, by exposure to tobacco-specific media, knowledge of the negative effects of tobacco, and smoking status
| Response indicating supportive attitudes toward tobacco control |
|||
| Tobacco companies should not be allowed to sell a harmful product (N = 2,256) | There are not enough restrictions on where people can smoke cigarettes (N = 2,252) | Government should be involved in where people can smoke cigarettes (N = 2,248) | |
| OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Exposure to neutral or antitobacco informationa,b (ref. no exposure) | |||
| News: Efforts to ban smoking in public places | 0.85 (0.65–1.16) | 0.81 (0.62–1.06) | 0.93 (0.71–1.20) |
| News: Dangers to kids | 1.17 (0.95–1.46) | 1.08 (0.87–1.35) | 1.13 (0.92–1.40) |
| Ads: Efforts to ban smoking in public places | 1.40** (1.07–1.81) | 1.15 (0.89–1.50) | 1.02 (0.79–1.32) |
| Ads: Dangers to kids | 1.41** (1.13–1.77) | 1.01 (0.80–1.26) | 1.06 (0.85–1.32) |
| Exposure to protobacco advertisinga,c (ref. no exposure) | |||
| Ads for cigarettes in newspaper | 1.03 (0.84–1.29) | 1.04 (0.85–1.29) | 0.95 (0.77–1.17) |
| Ads for cigarettes in magazine | 0.84 (0.68–1.04) | 1.20 (0.98–1.48) | 0.92 (0.76–1.13) |
| Ads for cigarettes on Internet | 0.78 (0.58–1.04) | 0.76 (0.57–1.02) | 0.88 (0.67–1.18) |
| Knowledge (ref. disagree, neutral, or don’t know) | |||
| Smoking has been proven to cause lung cancer | 1.06 (0.84–1.34) | 1.59*** (1.26–2.01) | 2.38*** (1.92–2.97) |
| Secondhand smoke causes lung cancer in nonsmokers | 2.11*** (1.62–2.74) | 2.66*** (1.97–3.59) | 1.89*** (1.46–2.46) |
| Smoking around a baby increases its chance of dying from SIDS | 1.66*** (1.37–2.01) | 1.39*** (1.15–1.67) | 1.34** (1.11–1.61) |
| Smoking status (ref. current smoker) | |||
| Never-smoker | 1.52** (1.20–1.93) | 3.89*** (3.02–5.03) | 1.66*** (1.32–2.09) |
| Former smoker | 1.35* (1.02–1.78) | 2.93*** (2.17–3.95) | 1.29* (1.02–1.70) |
Note. American Smoking and Health Survey 2 —2003. ORs, and 95% CIs for fitted logistic regression models that describe respondent odds of holding supportive attitudes toward tobacco control, by exposure to tobacco-specific media messages, knowledge of the negative effects of tobacco, and smoking status, controlling for education, income, race/ethnicity, age, sex, and state-specific indicator variables. All estimates are weighted. OR = odds ratio. Bold values represent statistical significance at p < .05.
Questions representing exposure to neutral, antitobacco, and protobacco advertising were asked only of respondents who reported at least some media exposure in the past 30 days (e.g., television, radio, newspaper, magazine, or Internet).
In past 30 days, saw news coverage or advertising about efforts to ban smoking in public places, such as restaurants, bars and cocktail lounges, and about the dangers of children being around cigarette smoke.
In past 30 days, saw advertising or promotions for cigarettes or other tobacco products in newspaper, magazine, or on the Internet.
*p < .05, **p < .01, ***p < .0001.