Tobacco Industry Denormalization Beliefs in Hong Kong Adolescents

Tobacco Industry Denormalization Beliefs in Hong Kong Adolescents Abstract Background Tobacco industry denormalization (TID) seeks to expose the industry’s misconducts. Research on TID beliefs, meaning negative attitudes toward the tobacco industry (TI), may inform TID programs, but was limited to western populations. We investigated TID beliefs and their association with smoking and sociodemographic characteristics in Hong Kong adolescents. Methods In a school-based cross-sectional survey of 14214 students (mean age 15.0 years, 51.5% boys), TID beliefs were assessed by two questions: (1) whether the TI was respectable and (2) whether the TI tried to get youth to smoke, each with four options from “definitely no” to “definitely yes.” Smoking susceptibility and behaviors were also assessed. Sociodemographic characteristics included age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. Results Of all students, 77.6% considered the TI not respectable and 56.6% believed that the TI tried to get youth to smoke. Stronger TID beliefs were inversely associated with smoking susceptibility and behaviors. For example, students considering the TI definitely not respectable (vs. definitely yes) were 56% (95% confidence intervals [CI] = 45% to 66%), 49% (95% CI = 41% to 56%), and 53% (95% CI = 36% to 65%) less likely to be susceptible to smoking (among never-smokers) and be ever- and current smokers, respectively. Of all correlates examined, only younger age and having no co-residing smoker were associated with TID beliefs. Conclusions Substantial proportions of Hong Kong adolescents did not hold TID beliefs, but those who did were less likely to smoke. Our results suggest that TID programs may help reduce adolescent smoking. Implications The TI’s misconducts and responsibility for the tobacco epidemic were not well known by Hong Kong adolescents. TID beliefs in this population were inversely associated with smoking. These findings suggest that TID programs in local adolescents may be of value. The investigation into the correlates of TID beliefs found that socioeconomic status and school-level smoking prevalence were not associated with TID beliefs. This suggests that local TID programs targeting adolescents in general, for example, mass-media campaigns, may be more appropriate than those targeting particular schools or selected groups of adolescents. Introduction Confronted by compelling scientific evidence of the harms of tobacco, the tobacco industry (TI), since the 1950s, has resisted government regulation of its products using various tactics. These tactics include buying scientific expertise to create controversy over tobacco harm, hiring lobbyists to influence policy making, and corrupting government officials.1,2 Article 12 of the World Health Organization Framework Convention on Tobacco Control supports raising public awareness of the TI’s responsibility for the tobacco epidemic and exposing its deceptive and manipulative tactics.3 Such intervention, known as tobacco industry denormalization (TID), has been the focus of several antitobacco campaigns in the United States.4 Most of the TID campaigns targeted adolescents, and some were effective in reducing adolescent smoking prevalence.4 The TI strongly opposed TID efforts, as revealed in its internal documents of the intention “to limit the spread of ‘demonisation’ (of the TI) from the developed world to the emerging markets.”5–7 These reactions suggest that TID may be an effective tool for tobacco control. However, TID has rarely been adopted outside the United States. Even research on the public’s TID beliefs, meaning negative attitudes toward the TI, has been limited to a few western countries.4,8 However, the TI’s anti-regulation tactics in the West are being repeated in low-income and middle-income countries, on which the future of the industry will depend.7 TID-related research in these settings is thus urgently needed and may in several ways inform local TID programs. First, local data on the prevalence of TID beliefs can help gauge the potential value of TID efforts. Second, the inverse associations between TID beliefs and smoking observed in western studies may be confirmed to support local TID programs.9–13 Third, identifying any sociodemographic correlates of TID beliefs may reveal high-risk target groups. In this study, we used data from a territory-wide school-based cross-sectional survey of secondary school students in Hong Kong to investigate the prevalence of TID beliefs and their association with smoking, and sociodemographic and other smoking-related correlates. Methods Data Source A cross-sectional survey using a stratified random sample of secondary schools in all 18 districts of Hong Kong was conducted from October 2014 to April 2015. All students in the recruited schools were invited to participate. The survey was conducted in classrooms, using anonymous paper-and-pencil questionnaires in three different versions. Each version comprised version-specific questions and “core” questions that were common to all versions. Each school was given one version by random. Ethics approval was granted by the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster. A total of 41035 students from 92 schools participated in the survey, with student-level and school-level response rates of 95% and 36% respectively. This study used data collected by one questionnaire version, which was administered to 14414 students in 31 schools. The students (n = 110) with missing data for age, sex, or more than 50% of items and those (n = 90) whose responses showed multiple internal inconsistencies were excluded. This left 14214 students for analysis. With further exclusion of 1945 ever-smokers and 42 with a missing value for ever smoking status, another analytic sample comprised 12227 never-smokers. Measurements TID beliefs were assessed by two questions: (1) “Do you think the tobacco industry is respectable?” and (2) “Do you think the tobacco industry tries to get youth to smoke?” Both had response options of “definitely yes,” “probably yes,” “probably no,” and “definitely no.” Students’ responses to these questions, as study factors, had four levels. As outcome variables, the responses to whether the TI was respectable were dichotomized as no (definitely no/probably no) versus yes (definitely yes/probably yes); the responses to whether the TI tried to get youth to smoke were dichotomized as yes (definitely yes/probably yes) versus no (definitely no/probably no). Smoking susceptibility was assessed by two questions: (1) “If one of your good friends offers you a cigarette, will you smoke it?” and (2) “Do you think you will smoke cigarettes in the next 12 months?” Both had response options of “definitely not,” “probably not,” “probably will,” and “definitely will.” Those choosing “definitely not” for both questions were deemed non-susceptible to smoking, and the others susceptible to smoking. Smoking susceptibility strongly predicts future smoking in never smoking adolescents.14,15 Students also reported whether they had ever smoked (even one puff) and whether they smoked in the past 30 days. Those who smoked in the past 30 days were deemed current smokers, and the others non-current smokers.16 Students also reported their age (in years), sex, perceived family affluence (relatively poor/poor to average/average/average to rich/relatively rich), highest parental education (primary or below/secondary/post-secondary or above/do not know), and number of co-residing smokers (0/1/2 or more). In addition, the 31 schools were evenly divided into three groups that, respectively, had low (10 schools, prevalence 0.0%–1.0%), medium (10 schools, 1.5%–6.9%), and high (11 schools, 7.8%–20.5%) school-level current smoking prevalence. Analysis Descriptive analysis was used to describe students’ TID beliefs. All descriptive results were weighted by age, sex, and grade based on the target population’s characteristics provided by the Education Bureau of the Hong Kong Government. Two-level generalized linear models with a random school-level intercept were used to examine the associations of interest. With outcomes assumed to be normally distributed and log-link functions, the models produced estimates of prevalence ratios (PRs). Because the outcomes were actually binary, robust variance estimators for standard errors were used.17,18 PRs, rather than odds ratios (ORs), were estimated because the outcomes were not rare (eg, ever smoking 14.1%; considering the TI not respectable 77.6%), and ORs, if interpreted as PRs, could be misleading in such situations. We examined the associations of TID beliefs (study factors) with smoking susceptibility, ever smoking, and current smoking (outcomes), with and without adjusting for age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. The analyses for smoking susceptibility were conducted in never-smokers. TID beliefs, as study factors, were also analyzed as continuous variables to test linear trends. We also examined the associations of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence (study factors) with TID beliefs (outcomes), with mutual adjustment of the study factors. All p values were two tailed, and values <.05 were considered statistically significant. Analyses used Stata 13.1. Results Table 1 shows that students’ mean age (SD) was 15.0 (1.9) years, and 51.5% were boys. The prevalence of ever and current smoking were 14.1% and 5.5%, and 11.9% of never-smokers were susceptible to smoking. Of all students, in response to whether the TI was respectable, 41.1% chose “definitely no,” and 36.5%, 19.5%, and 3.0% chose “probably no,” “probably yes,” and “definitely yes,” respectively. In response to whether the TI tried to get youth to smoke, only 15.9% chose “definitely yes,” and 40.7%, 31.5%, and 11.9% chose “probably yes,” “probably no,” and “definitely no,” respectively. Table 1. Background Characteristics (n = 14214)   n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)    n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)  aNumber and proportion unless otherwise stated. View Large Table 1. Background Characteristics (n = 14214)   n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)    n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)  aNumber and proportion unless otherwise stated. View Large Table 2 shows that, in general, considering the TI not respectable was inversely associated with smoking susceptibility, ever smoking, and current smoking in both crude and adjusted models (ps for trend < .001). Compared with choosing “definitely yes” for whether the TI was respectable, choosing “definitely no” and “probably no” was associated with adjusted PRs of 0.44 (95% confidence intervals [CI] = 0.34% to 0.55%) and 0.76 (95% CI = 0.60% to 0.97%), respectively, for smoking susceptibility. For ever smoking, the corresponding adjusted PRs were 0.51 (95% CI = 0.44% to 0.59%) and 0.65 (95% CI = 0.55% to 0.77%); for current smoking, the adjusted PRs in relation to “definitely no,” “probably no,” and “probably yes” were 0.47 (95% CI = 0.35% to 0.64%), 0.53 (95% CI = 0.39% to 0.74%), and 0.70 (95% CI = 0.55% to 0.91%), respectively. Table 2. PRs (95% CIs) of Smoking in Relation to TID Beliefs in Adolescents (Whole Sample, n = 14214; Never-Smokers, n = 12227)   Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001    Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  PRs = prevalence ratios; CIs = confidence intervals; TID = tobacco industry denormalization. aWith adjustment of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. *p < .05; **p < .01; ***p < .001. View Large Table 2. PRs (95% CIs) of Smoking in Relation to TID Beliefs in Adolescents (Whole Sample, n = 14214; Never-Smokers, n = 12227)   Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001    Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  PRs = prevalence ratios; CIs = confidence intervals; TID = tobacco industry denormalization. aWith adjustment of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. *p < .05; **p < .01; ***p < .001. View Large Table 2 also shows that, in general, believing that the TI tried to get youth to smoke was inversely associated with smoking susceptibility, ever smoking, and current smoking in both crude and adjusted models (ps for trend < .001). Compared with choosing “definitely no” for whether the TI tried to get youth to smoke, choosing “definitely yes” was associated with an adjusted PR of 0.63 (95% CI = 0.49% to 0.81%) for smoking susceptibility. For ever smoking, the adjusted PRs in relation to “definitely no,” “probably no,” and “probably yes” were 0.71 (95% CI = 0.61% to 0.82%), 0.67 (95% CI = 0.58% to 0.78%), and 0.82 (95% CI = 0.73% to 0.92%), respectively; for current smoking, the corresponding figures were 0.60 (95% CI = 0.48% to 0.74%), 0.53 (95% CI = 0.45% to 0.64%), and 0.61 (95% CI = 0.50% to 0.75%). Table 3 shows that sex, perceived family affluence, highest parental education, and school-level smoking prevalence were not associated with TID beliefs. Younger age was generally associated with TID beliefs. For example, compared with ≥17 years of age, 14, 13, and ≤12 were associated with adjusted PRs of 1.06 (95% CI = 1.03% to 1.10%), 1.08 (95% CI = 1.04% to 1.12%), and 1.12 (95% CI = 1.09% to 1.15%), respectively, for considering the TI not respectable. In addition, having no co-residing smoker, compared with having two or more, was associated with an adjusted PR of 1.08 (95% CI = 1.03% to 1.12%) for considering the TI not respectable, and 1.13 (95% CI = 1.08% to 1.18%) for believing that the TI tried to get youth to smoke. The corresponding figures for having one co-residing smoker were nonsignificant. Table 3. Adjusted PRs (95% CIs)a of TID Beliefs in Relation to Sociodemographic and Smoking-Related Characteristics (n = 14214) Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  PRs = prevalence ratios; CIs = confidence intervals. aWith mutual adjustment of all the sociodemographic and smoking-related characteristics in the table. bComprising definitely no and probably no. cComprising definitely yes and probably yes. *p < .05; **p < .01; ***p < .001. View Large Table 3. Adjusted PRs (95% CIs)a of TID Beliefs in Relation to Sociodemographic and Smoking-Related Characteristics (n = 14214) Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  PRs = prevalence ratios; CIs = confidence intervals. aWith mutual adjustment of all the sociodemographic and smoking-related characteristics in the table. bComprising definitely no and probably no. cComprising definitely yes and probably yes. *p < .05; **p < .01; ***p < .001. View Large Discussion We found that substantial proportions of Hong Kong adolescents did not hold TID beliefs. Almost one quarter of them considered the TI respectable. Moreover, more than two-fifths of them did not believe that the TI tried to get youth to smoke, and only less than one-fifth firmly believed that this was the case. Surveys in developed countries, which used different questions to assess TID beliefs, also showed awareness gaps in the TI’s malpractices among adults and adolescents.8,19 We also found that TID beliefs were inversely associated with smoking, which was consistent with the studies in the West.9–13 The above findings collectively suggest that local TID programs may have a potential to prevent adolescent smoking. In Hong Kong, after 30 years of strong tobacco control measures that have reduced the adult daily smoking prevalence to one of the lowest in the developed world (10.5% in 2015),20 TID may represent an opportunity for further progress in tobacco control. TID may be preferable to the communication of long-term health risks in adolescents because the latter are generally not their main concerns. Another advantage of TID is that it targets the industry instead of smokers. Tobacco control advocates would be seen as a group acting for the public wellbeing by challenging the industry’s vested interests, rather than as they are often viewed by the public as an elitist group who unfairly impose their own standards on others.21 When fully implemented by June 2018, the number of different graphic health warnings on cigarette packets in Hong Kong will increase from six to twelve, including mainly health consequences but also one on smoking offence penalties.22 TID messages in this context should also be considered. Future tobacco control campaigns in Hong Kong may use TID messages along with other evidence-based approaches, for example, graphic health effect campaigns23; further research is needed to explore the relative effectiveness of these approaches as well as potential effect variation by target groups. However, there may be challenges in launching TID campaigns, such as other health promotion priorities, opposition from the TI, and fear of industry litigation.24,25 The local evidence provided in our study should be of value in the decision-making process. We found that younger age was associated with TID beliefs. As adolescent smoking increases with age, higher peer smoking in older adolescents may have led to more receptive attitudes toward smoking and the TI. This was in line with our finding that having no co-residing smoker was also associated with TID beliefs, although the association was weak. The association suggests that having family members who smoked may also have led adolescents to perceive the TI positively. In addition, we found no association of school-level smoking prevalence and indicators of socioeconomic status, that is, perceived family affluence and highest parental education, with TID beliefs. Our findings generally suggest that, to denormalize the TI in Hong Kong adolescents, a general population approach, for example, mass-media campaign and territory-wide school-based education, may be more appropriate than a targeted approach for individuals, schools, or communities with certain characteristics. Our study has several limitations. First, the survey’s response rate on school-level was relatively low (36%). However, the non-responses were mainly due to administrative reasons, and the recruited and not recruited schools were similar with regard to districts, mediums of instruction, sources of financial support, and single or mixed sex education (chi-square tests, ps > .05). Second, while the validated measure of smoking susceptibility comprised three questions, two of them were used in our study, and both had slightly different wordings.14,15 It was possible for these differences to affect the strength of the predictive ability of our indicator of smoking susceptibility for future smoking. Third, it was difficult to establish temporal relations with the cross-sectional design. Specifically, apart from the effects of TID beliefs, the inverse associations between TID beliefs and smoking may also have been due to smokers’ reluctance to concede that the TI was manipulative, deceitful, or disrespectable to avoid dissonance with their own smoking behaviors. Smoking susceptibility, however, seemed less likely to have such reverse causal effect. Fourth, because of the observational design, we cannot rule out residual confounding effect in the observed associations. To conclude, substantial proportions of Hong Kong adolescents did not hold TID beliefs, but those who did were less likely to smoke. These beliefs did not vary significantly by socioeconomic status or school factors. Our results suggest that TID programs that increase adolescents’ knowledge about TI malpractices may help reduce smoking uptake and that such programs may be best implemented as general population-wide approaches, rather than targeted at particular groups of adolescents, schools, or communities. Funding The survey was funded by the Food and Health Bureau of the Hong Kong Government. Declaration of Interests None declared. Acknowledgments We sincerely thank the schools and students for participation. References 1. Saloojee Y, Dagli E. Tobacco industry tactics for resisting public policy on health. Bull World Health Organ . 2000; 78( 7): 902– 910. Google Scholar PubMed  2. Brandt AM. Inventing conflicts of interest: a history of tobacco industry tactics. Am J Public Health . 2012; 102( 1): 63– 71. Google Scholar CrossRef Search ADS PubMed  3. WHO. Elaboration of guidelines for implementation of Article 12 of the Convention. In: Third Session of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; November 17–22, 2008; Durban, South Africa. 4. Malone RE, Grundy Q, Bero LA. Tobacco industry denormalisation as a tobacco control intervention: a review. Tob Control . 2012; 21( 2): 162– 170. Google Scholar CrossRef Search ADS PubMed  5. Glantz SA, Balbach ED. Tobacco War: Inside the California Battles . Berkeley, CA: University of California Press; 2000. 6. Apollonio DE, Malone RE. Turning negative into positive: public health mass media campaigns and negative advertising. Health Educ Res . 2009; 24( 3): 483– 495. Google Scholar CrossRef Search ADS PubMed  7. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry conduct in low-income and middle-income countries. Lancet . 2015; 385( 9972): 1029– 1043. Google Scholar CrossRef Search ADS PubMed  8. Moodie C, Sinclair L, Mackintosh AM, Power E, Bauld L. How tobacco companies are perceived within the United Kingdom: an online panel. Nicotine Tob Res . 2016; 18( 8): 1766– 1772. Google Scholar CrossRef Search ADS PubMed  9. Niederdeppe J, Farrelly MC, Haviland ML. Confirming “truth”: more evidence of a successful tobacco countermarketing campaign in Florida. Am J Public Health . 2004; 94( 2): 255– 257. Google Scholar CrossRef Search ADS PubMed  10. Thrasher JF, Niederdeppe JD, Jackson C, Farrelly MC. Using anti-tobacco industry messages to prevent smoking among high-risk adolescents. Health Educ Res . 2006; 21( 3): 325– 337. Google Scholar CrossRef Search ADS PubMed  11. Leatherdale ST, Sparks R, Kirsh VA. Beliefs about tobacco industry (mal)practices and youth smoking behaviour: insight for future tobacco control campaigns (Canada). Cancer Causes Control . 2006; 17( 5): 705– 711. Google Scholar CrossRef Search ADS PubMed  12. McCool J, Paynter J, Scragg R. A cross-sectional study of opinions related to the tobacco industry and their association with smoking status amongst 14–15 year old teenagers in New Zealand. N Z Med J . 2011; 124( 1338): 34– 43. Google Scholar PubMed  13. Brown AK, Moodie C, Hastings G, Mackintosh AM, Hassan L, Thrasher J. The association of normative perceptions with adolescent smoking intentions. J Adolesc . 2010; 33( 5): 603– 614. Google Scholar CrossRef Search ADS PubMed  14. Choi WS, Gilpin EA, Farkas AJ, Pierce JP. Determining the probability of future smoking among adolescents. Addiction . 2001; 96( 2): 313– 323. Google Scholar CrossRef Search ADS PubMed  15. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Merritt RK. Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychol . 1996; 15( 5): 355– 361. Google Scholar CrossRef Search ADS PubMed  16. Chen J, Ho SY, Leung LT, Wang MP, Lam TH. Associations of unhappiness with sociodemographic factors and unhealthy behaviours in Chinese adolescents. Eur J Public Health . 2017; 27( 3): 518– 524. Google Scholar CrossRef Search ADS PubMed  17. Cummings P. Methods for estimating adjusted risk ratios. Stata J . 2009; 9( 2): 175– 196. 18. Lumley T, Kronmal R, Ma S. Relative risk regression in medical research: Models, contrasts, estimators, and algorithms. UW Biostatistics Working Paper Series. Working Paper 293. 2006. http://biostats.bepress.com/uwbiostat/paper293. Accessed October 2017. 19. Waller BJ, Cohen JE, Ashley MJ. Youth attitudes towards tobacco control: a preliminary assessment. Chronic Dis Can . 2004; 25( 3–4): 97– 100. Google Scholar PubMed  20. Thematic Household Survey Report No.59 . Hong Kong: Census and Statistics Department of HKSAR; 2013. http://www.digital21.gov.hk/eng/statistics/download/householdreport2016_59.pdf. Accessed October 2017. 21. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health . 1997; 87( 6): 968– 973. Google Scholar CrossRef Search ADS PubMed  22. Smoking (Public Health) (Notices) (Amendment) Order 2017 . Hong Kong: Legislative Council; 2017. http://www.legco.gov.hk/yr16-17/english/subleg/negative/2017ln066-e.pdf. Accessed October 2017. 23. White V, Tan N, Wakefield M, Hill D. Do adult focused anti-smoking campaigns have an impact on adolescents? The case of the Australian National Tobacco Campaign. Tob Control . 2003; 12( suppl 2): ii23– ii9. Google Scholar PubMed  24. Balbach ED, Glantz SA. Tobacco control advocates must demand high-quality media campaigns: the California experience. Tob Control . 1998; 7( 4): 397– 408. Google Scholar CrossRef Search ADS PubMed  25. Ibrahim JK, Glantz SA. Tobacco industry litigation strategies to oppose tobacco control media campaigns. Tob Control . 2006; 15( 1): 50– 58. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nicotine and Tobacco Research Oxford University Press

Tobacco Industry Denormalization Beliefs in Hong Kong Adolescents

Loading next page...
 
/lp/ou_press/tobacco-industry-denormalization-beliefs-in-hong-kong-adolescents-O1ggKMC52O
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1462-2203
eISSN
1469-994X
D.O.I.
10.1093/ntr/nty094
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Tobacco industry denormalization (TID) seeks to expose the industry’s misconducts. Research on TID beliefs, meaning negative attitudes toward the tobacco industry (TI), may inform TID programs, but was limited to western populations. We investigated TID beliefs and their association with smoking and sociodemographic characteristics in Hong Kong adolescents. Methods In a school-based cross-sectional survey of 14214 students (mean age 15.0 years, 51.5% boys), TID beliefs were assessed by two questions: (1) whether the TI was respectable and (2) whether the TI tried to get youth to smoke, each with four options from “definitely no” to “definitely yes.” Smoking susceptibility and behaviors were also assessed. Sociodemographic characteristics included age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. Results Of all students, 77.6% considered the TI not respectable and 56.6% believed that the TI tried to get youth to smoke. Stronger TID beliefs were inversely associated with smoking susceptibility and behaviors. For example, students considering the TI definitely not respectable (vs. definitely yes) were 56% (95% confidence intervals [CI] = 45% to 66%), 49% (95% CI = 41% to 56%), and 53% (95% CI = 36% to 65%) less likely to be susceptible to smoking (among never-smokers) and be ever- and current smokers, respectively. Of all correlates examined, only younger age and having no co-residing smoker were associated with TID beliefs. Conclusions Substantial proportions of Hong Kong adolescents did not hold TID beliefs, but those who did were less likely to smoke. Our results suggest that TID programs may help reduce adolescent smoking. Implications The TI’s misconducts and responsibility for the tobacco epidemic were not well known by Hong Kong adolescents. TID beliefs in this population were inversely associated with smoking. These findings suggest that TID programs in local adolescents may be of value. The investigation into the correlates of TID beliefs found that socioeconomic status and school-level smoking prevalence were not associated with TID beliefs. This suggests that local TID programs targeting adolescents in general, for example, mass-media campaigns, may be more appropriate than those targeting particular schools or selected groups of adolescents. Introduction Confronted by compelling scientific evidence of the harms of tobacco, the tobacco industry (TI), since the 1950s, has resisted government regulation of its products using various tactics. These tactics include buying scientific expertise to create controversy over tobacco harm, hiring lobbyists to influence policy making, and corrupting government officials.1,2 Article 12 of the World Health Organization Framework Convention on Tobacco Control supports raising public awareness of the TI’s responsibility for the tobacco epidemic and exposing its deceptive and manipulative tactics.3 Such intervention, known as tobacco industry denormalization (TID), has been the focus of several antitobacco campaigns in the United States.4 Most of the TID campaigns targeted adolescents, and some were effective in reducing adolescent smoking prevalence.4 The TI strongly opposed TID efforts, as revealed in its internal documents of the intention “to limit the spread of ‘demonisation’ (of the TI) from the developed world to the emerging markets.”5–7 These reactions suggest that TID may be an effective tool for tobacco control. However, TID has rarely been adopted outside the United States. Even research on the public’s TID beliefs, meaning negative attitudes toward the TI, has been limited to a few western countries.4,8 However, the TI’s anti-regulation tactics in the West are being repeated in low-income and middle-income countries, on which the future of the industry will depend.7 TID-related research in these settings is thus urgently needed and may in several ways inform local TID programs. First, local data on the prevalence of TID beliefs can help gauge the potential value of TID efforts. Second, the inverse associations between TID beliefs and smoking observed in western studies may be confirmed to support local TID programs.9–13 Third, identifying any sociodemographic correlates of TID beliefs may reveal high-risk target groups. In this study, we used data from a territory-wide school-based cross-sectional survey of secondary school students in Hong Kong to investigate the prevalence of TID beliefs and their association with smoking, and sociodemographic and other smoking-related correlates. Methods Data Source A cross-sectional survey using a stratified random sample of secondary schools in all 18 districts of Hong Kong was conducted from October 2014 to April 2015. All students in the recruited schools were invited to participate. The survey was conducted in classrooms, using anonymous paper-and-pencil questionnaires in three different versions. Each version comprised version-specific questions and “core” questions that were common to all versions. Each school was given one version by random. Ethics approval was granted by the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster. A total of 41035 students from 92 schools participated in the survey, with student-level and school-level response rates of 95% and 36% respectively. This study used data collected by one questionnaire version, which was administered to 14414 students in 31 schools. The students (n = 110) with missing data for age, sex, or more than 50% of items and those (n = 90) whose responses showed multiple internal inconsistencies were excluded. This left 14214 students for analysis. With further exclusion of 1945 ever-smokers and 42 with a missing value for ever smoking status, another analytic sample comprised 12227 never-smokers. Measurements TID beliefs were assessed by two questions: (1) “Do you think the tobacco industry is respectable?” and (2) “Do you think the tobacco industry tries to get youth to smoke?” Both had response options of “definitely yes,” “probably yes,” “probably no,” and “definitely no.” Students’ responses to these questions, as study factors, had four levels. As outcome variables, the responses to whether the TI was respectable were dichotomized as no (definitely no/probably no) versus yes (definitely yes/probably yes); the responses to whether the TI tried to get youth to smoke were dichotomized as yes (definitely yes/probably yes) versus no (definitely no/probably no). Smoking susceptibility was assessed by two questions: (1) “If one of your good friends offers you a cigarette, will you smoke it?” and (2) “Do you think you will smoke cigarettes in the next 12 months?” Both had response options of “definitely not,” “probably not,” “probably will,” and “definitely will.” Those choosing “definitely not” for both questions were deemed non-susceptible to smoking, and the others susceptible to smoking. Smoking susceptibility strongly predicts future smoking in never smoking adolescents.14,15 Students also reported whether they had ever smoked (even one puff) and whether they smoked in the past 30 days. Those who smoked in the past 30 days were deemed current smokers, and the others non-current smokers.16 Students also reported their age (in years), sex, perceived family affluence (relatively poor/poor to average/average/average to rich/relatively rich), highest parental education (primary or below/secondary/post-secondary or above/do not know), and number of co-residing smokers (0/1/2 or more). In addition, the 31 schools were evenly divided into three groups that, respectively, had low (10 schools, prevalence 0.0%–1.0%), medium (10 schools, 1.5%–6.9%), and high (11 schools, 7.8%–20.5%) school-level current smoking prevalence. Analysis Descriptive analysis was used to describe students’ TID beliefs. All descriptive results were weighted by age, sex, and grade based on the target population’s characteristics provided by the Education Bureau of the Hong Kong Government. Two-level generalized linear models with a random school-level intercept were used to examine the associations of interest. With outcomes assumed to be normally distributed and log-link functions, the models produced estimates of prevalence ratios (PRs). Because the outcomes were actually binary, robust variance estimators for standard errors were used.17,18 PRs, rather than odds ratios (ORs), were estimated because the outcomes were not rare (eg, ever smoking 14.1%; considering the TI not respectable 77.6%), and ORs, if interpreted as PRs, could be misleading in such situations. We examined the associations of TID beliefs (study factors) with smoking susceptibility, ever smoking, and current smoking (outcomes), with and without adjusting for age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. The analyses for smoking susceptibility were conducted in never-smokers. TID beliefs, as study factors, were also analyzed as continuous variables to test linear trends. We also examined the associations of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence (study factors) with TID beliefs (outcomes), with mutual adjustment of the study factors. All p values were two tailed, and values <.05 were considered statistically significant. Analyses used Stata 13.1. Results Table 1 shows that students’ mean age (SD) was 15.0 (1.9) years, and 51.5% were boys. The prevalence of ever and current smoking were 14.1% and 5.5%, and 11.9% of never-smokers were susceptible to smoking. Of all students, in response to whether the TI was respectable, 41.1% chose “definitely no,” and 36.5%, 19.5%, and 3.0% chose “probably no,” “probably yes,” and “definitely yes,” respectively. In response to whether the TI tried to get youth to smoke, only 15.9% chose “definitely yes,” and 40.7%, 31.5%, and 11.9% chose “probably yes,” “probably no,” and “definitely no,” respectively. Table 1. Background Characteristics (n = 14214)   n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)    n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)  aNumber and proportion unless otherwise stated. View Large Table 1. Background Characteristics (n = 14214)   n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)    n (%)a  Age (years)     ≤12  1614 (11.4)   13  2051 (14.4)   14  2308 (16.2)   15  2156 (15.2)   16  2310 (16.3)   ≥17  3772 (26.5)  Mean age in years (SD)  15.0 (1.9)  Sex     Girls  6889 (48.5)   Boys  7321 (51.5)  Perceived family affluence     Relatively poor  941 (6.6)   Poor to average  3600 (25.4)   Average  8078 (57.0)   Average to rich  1349 (9.5)   Relatively rich  209 (1.5)  Highest parental education     Primary or below  945 (6.7)   Secondary  8163 (57.5)   Post-secondary or above  2836 (20.0)   Do not know  2265 (15.9)  Number of co-residing smokers     0  8839 (62.6)   1  3929 (27.8)   2 or more  1364 (9.7)  The tobacco industry is respectable     Definitely no  5829 (41.1)   Probably no  5176 (36.5)   Probably yes  2764 (19.5)   Definitely yes  421 (3.0)  The tobacco industry tries to get youth to smoke     Definitely no  1683 (11.9)   Probably no  4471 (31.5)   Probably yes  5769 (40.7)   Definitely yes  2260 (15.9)  Smoking susceptibility (in never-smokers)     No  10722 (88.2)   Yes  1441 (11.9)  Smoking status     Never smoking  12163 (85.9)   Ever smoking  2003 (14.1)   Non-current smoking  13385 (94.5)   Current smoking  785 (5.5)  aNumber and proportion unless otherwise stated. View Large Table 2 shows that, in general, considering the TI not respectable was inversely associated with smoking susceptibility, ever smoking, and current smoking in both crude and adjusted models (ps for trend < .001). Compared with choosing “definitely yes” for whether the TI was respectable, choosing “definitely no” and “probably no” was associated with adjusted PRs of 0.44 (95% confidence intervals [CI] = 0.34% to 0.55%) and 0.76 (95% CI = 0.60% to 0.97%), respectively, for smoking susceptibility. For ever smoking, the corresponding adjusted PRs were 0.51 (95% CI = 0.44% to 0.59%) and 0.65 (95% CI = 0.55% to 0.77%); for current smoking, the adjusted PRs in relation to “definitely no,” “probably no,” and “probably yes” were 0.47 (95% CI = 0.35% to 0.64%), 0.53 (95% CI = 0.39% to 0.74%), and 0.70 (95% CI = 0.55% to 0.91%), respectively. Table 2. PRs (95% CIs) of Smoking in Relation to TID Beliefs in Adolescents (Whole Sample, n = 14214; Never-Smokers, n = 12227)   Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001    Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  PRs = prevalence ratios; CIs = confidence intervals; TID = tobacco industry denormalization. aWith adjustment of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. *p < .05; **p < .01; ***p < .001. View Large Table 2. PRs (95% CIs) of Smoking in Relation to TID Beliefs in Adolescents (Whole Sample, n = 14214; Never-Smokers, n = 12227)   Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001    Smoking susceptibility (in never-smokers)  Ever smoking  Current smoking  TID beliefs  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  Crude PR (95% CI)  Adjusted PR (95% CI)a  The tobacco industry is respectable   Definitely yes  1  1  1  1  1  1   Probably yes  1.02 (0.79–1.31)  1.05 (0.82–1.35)  0.86 (0.69–1.05)  0.86 (0.70–1.05)  0.67 (0.51–0.87)**  0.70 (0.55–0.91)**   Probably no  0.72 (0.56–0.93)*  0.76 (0.60–0.97)*  0.60 (0.49–0.74)***  0.65 (0.55–0.77)***  0.40 (0.31–0.52)***  0.53 (0.39–0.74)***   Definitely no  0.43 (0.33–0.56)***  0.44 (0.34–0.55)***  0.42 (0.35–0.50)***  0.51 (0.44–0.59)***  0.27 (0.21–0.33)***  0.47 (0.35–0.64)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  The tobacco industry tries to get youth to smoke   Definitely no  1  1  1  1  1  1   Probably no  0.99 (0.82–1.20)  1.03 (0.84–1.26)  0.85 (0.73–0.99)*  0.82 (0.73–0.92)**  0.57 (0.46–0.69)***  0.61 (0.50–0.75)***   Probably yes  0.84 (0.70–1.01)  0.87 (0.72–1.06)  0.65 (0.52–0.80)***  0.67 (0.58–0.78)***  0.45 (0.32–0.62)***  0.53 (0.45–0.64)***   Definitely yes  0.63 (0.49–0.81)***  0.63 (0.49–0.81)***  0.62 (0.52–0.75)***  0.71 (0.61–0.82)***  0.50 (0.36–0.69)***  0.60 (0.48–0.74)***  p for trend  <.001  <.001  <.001  <.001  <.001  <.001  PRs = prevalence ratios; CIs = confidence intervals; TID = tobacco industry denormalization. aWith adjustment of age, sex, perceived family affluence, highest parental education, numbers of co-residing smokers, and school-level smoking prevalence. *p < .05; **p < .01; ***p < .001. View Large Table 2 also shows that, in general, believing that the TI tried to get youth to smoke was inversely associated with smoking susceptibility, ever smoking, and current smoking in both crude and adjusted models (ps for trend < .001). Compared with choosing “definitely no” for whether the TI tried to get youth to smoke, choosing “definitely yes” was associated with an adjusted PR of 0.63 (95% CI = 0.49% to 0.81%) for smoking susceptibility. For ever smoking, the adjusted PRs in relation to “definitely no,” “probably no,” and “probably yes” were 0.71 (95% CI = 0.61% to 0.82%), 0.67 (95% CI = 0.58% to 0.78%), and 0.82 (95% CI = 0.73% to 0.92%), respectively; for current smoking, the corresponding figures were 0.60 (95% CI = 0.48% to 0.74%), 0.53 (95% CI = 0.45% to 0.64%), and 0.61 (95% CI = 0.50% to 0.75%). Table 3 shows that sex, perceived family affluence, highest parental education, and school-level smoking prevalence were not associated with TID beliefs. Younger age was generally associated with TID beliefs. For example, compared with ≥17 years of age, 14, 13, and ≤12 were associated with adjusted PRs of 1.06 (95% CI = 1.03% to 1.10%), 1.08 (95% CI = 1.04% to 1.12%), and 1.12 (95% CI = 1.09% to 1.15%), respectively, for considering the TI not respectable. In addition, having no co-residing smoker, compared with having two or more, was associated with an adjusted PR of 1.08 (95% CI = 1.03% to 1.12%) for considering the TI not respectable, and 1.13 (95% CI = 1.08% to 1.18%) for believing that the TI tried to get youth to smoke. The corresponding figures for having one co-residing smoker were nonsignificant. Table 3. Adjusted PRs (95% CIs)a of TID Beliefs in Relation to Sociodemographic and Smoking-Related Characteristics (n = 14214) Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  PRs = prevalence ratios; CIs = confidence intervals. aWith mutual adjustment of all the sociodemographic and smoking-related characteristics in the table. bComprising definitely no and probably no. cComprising definitely yes and probably yes. *p < .05; **p < .01; ***p < .001. View Large Table 3. Adjusted PRs (95% CIs)a of TID Beliefs in Relation to Sociodemographic and Smoking-Related Characteristics (n = 14214) Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  Characteristics  The tobacco industry is respectable  The tobacco industry tries to get youth to smoke  Nob vs. Yesc (reference)  Yesc vs. Nob (reference)  Age (years)       ≥17  1  1   16  1.01 (0.98–1.04)  1.01 (0.95–1.08)   15  1.02 (0.99–1.06)  1.06 (0.96–1.16)   14  1.06 (1.03–1.10)***  1.12 (1.00–1.26)*   13  1.08 (1.04–1.12)***  1.18 (1.05–1.33)**   ≤12  1.12 (1.09–1.15)***  1.21 (1.05–1.40)**  Sex       Girls  1  1   Boys  0.99 (0.97–1.01)  1.03 (1.00–1.06)  Perceived family affluence       Relatively poor  1  1   Poor to average  1.02 (0.98–1.07)  0.95 (0.89–1.01)   Average  1.01 (0.98–1.05)  0.96 (0.91–1.01)   Average to rich  1.03 (0.98–1.08)  0.98 (0.91–1.06)   Relatively rich  0.93 (0.82–1.05)  0.95 (0.84–1.07)  Highest parental education       Primary or below  1  1   Secondary  1.00 (0.96–1.04)  0.99 (0.92–1.06)   Post-secondary or above  0.99 (0.94–1.05)  0.97 (0.90–1.04)   Do not know  0.97 (0.92–1.02)  1.00 (0.92–1.09)  Number of co-residing smokers       2 or more  1  1   1  1.05 (1.00–1.09)  1.05 (1.00–1.11)   0  1.08 (1.03–1.12)***  1.13 (1.08–1.18)***  School-level smoking prevalence       High  1  1   Medium  1.02 (0.98–1.06)  1.06 (0.95–1.18)   Low  1.02 (0.99–1.07)  1.00 (0.90–1.10)  PRs = prevalence ratios; CIs = confidence intervals. aWith mutual adjustment of all the sociodemographic and smoking-related characteristics in the table. bComprising definitely no and probably no. cComprising definitely yes and probably yes. *p < .05; **p < .01; ***p < .001. View Large Discussion We found that substantial proportions of Hong Kong adolescents did not hold TID beliefs. Almost one quarter of them considered the TI respectable. Moreover, more than two-fifths of them did not believe that the TI tried to get youth to smoke, and only less than one-fifth firmly believed that this was the case. Surveys in developed countries, which used different questions to assess TID beliefs, also showed awareness gaps in the TI’s malpractices among adults and adolescents.8,19 We also found that TID beliefs were inversely associated with smoking, which was consistent with the studies in the West.9–13 The above findings collectively suggest that local TID programs may have a potential to prevent adolescent smoking. In Hong Kong, after 30 years of strong tobacco control measures that have reduced the adult daily smoking prevalence to one of the lowest in the developed world (10.5% in 2015),20 TID may represent an opportunity for further progress in tobacco control. TID may be preferable to the communication of long-term health risks in adolescents because the latter are generally not their main concerns. Another advantage of TID is that it targets the industry instead of smokers. Tobacco control advocates would be seen as a group acting for the public wellbeing by challenging the industry’s vested interests, rather than as they are often viewed by the public as an elitist group who unfairly impose their own standards on others.21 When fully implemented by June 2018, the number of different graphic health warnings on cigarette packets in Hong Kong will increase from six to twelve, including mainly health consequences but also one on smoking offence penalties.22 TID messages in this context should also be considered. Future tobacco control campaigns in Hong Kong may use TID messages along with other evidence-based approaches, for example, graphic health effect campaigns23; further research is needed to explore the relative effectiveness of these approaches as well as potential effect variation by target groups. However, there may be challenges in launching TID campaigns, such as other health promotion priorities, opposition from the TI, and fear of industry litigation.24,25 The local evidence provided in our study should be of value in the decision-making process. We found that younger age was associated with TID beliefs. As adolescent smoking increases with age, higher peer smoking in older adolescents may have led to more receptive attitudes toward smoking and the TI. This was in line with our finding that having no co-residing smoker was also associated with TID beliefs, although the association was weak. The association suggests that having family members who smoked may also have led adolescents to perceive the TI positively. In addition, we found no association of school-level smoking prevalence and indicators of socioeconomic status, that is, perceived family affluence and highest parental education, with TID beliefs. Our findings generally suggest that, to denormalize the TI in Hong Kong adolescents, a general population approach, for example, mass-media campaign and territory-wide school-based education, may be more appropriate than a targeted approach for individuals, schools, or communities with certain characteristics. Our study has several limitations. First, the survey’s response rate on school-level was relatively low (36%). However, the non-responses were mainly due to administrative reasons, and the recruited and not recruited schools were similar with regard to districts, mediums of instruction, sources of financial support, and single or mixed sex education (chi-square tests, ps > .05). Second, while the validated measure of smoking susceptibility comprised three questions, two of them were used in our study, and both had slightly different wordings.14,15 It was possible for these differences to affect the strength of the predictive ability of our indicator of smoking susceptibility for future smoking. Third, it was difficult to establish temporal relations with the cross-sectional design. Specifically, apart from the effects of TID beliefs, the inverse associations between TID beliefs and smoking may also have been due to smokers’ reluctance to concede that the TI was manipulative, deceitful, or disrespectable to avoid dissonance with their own smoking behaviors. Smoking susceptibility, however, seemed less likely to have such reverse causal effect. Fourth, because of the observational design, we cannot rule out residual confounding effect in the observed associations. To conclude, substantial proportions of Hong Kong adolescents did not hold TID beliefs, but those who did were less likely to smoke. These beliefs did not vary significantly by socioeconomic status or school factors. Our results suggest that TID programs that increase adolescents’ knowledge about TI malpractices may help reduce smoking uptake and that such programs may be best implemented as general population-wide approaches, rather than targeted at particular groups of adolescents, schools, or communities. Funding The survey was funded by the Food and Health Bureau of the Hong Kong Government. Declaration of Interests None declared. Acknowledgments We sincerely thank the schools and students for participation. References 1. Saloojee Y, Dagli E. Tobacco industry tactics for resisting public policy on health. Bull World Health Organ . 2000; 78( 7): 902– 910. Google Scholar PubMed  2. Brandt AM. Inventing conflicts of interest: a history of tobacco industry tactics. Am J Public Health . 2012; 102( 1): 63– 71. Google Scholar CrossRef Search ADS PubMed  3. WHO. Elaboration of guidelines for implementation of Article 12 of the Convention. In: Third Session of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; November 17–22, 2008; Durban, South Africa. 4. Malone RE, Grundy Q, Bero LA. Tobacco industry denormalisation as a tobacco control intervention: a review. Tob Control . 2012; 21( 2): 162– 170. Google Scholar CrossRef Search ADS PubMed  5. Glantz SA, Balbach ED. Tobacco War: Inside the California Battles . Berkeley, CA: University of California Press; 2000. 6. Apollonio DE, Malone RE. Turning negative into positive: public health mass media campaigns and negative advertising. Health Educ Res . 2009; 24( 3): 483– 495. Google Scholar CrossRef Search ADS PubMed  7. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry conduct in low-income and middle-income countries. Lancet . 2015; 385( 9972): 1029– 1043. Google Scholar CrossRef Search ADS PubMed  8. Moodie C, Sinclair L, Mackintosh AM, Power E, Bauld L. How tobacco companies are perceived within the United Kingdom: an online panel. Nicotine Tob Res . 2016; 18( 8): 1766– 1772. Google Scholar CrossRef Search ADS PubMed  9. Niederdeppe J, Farrelly MC, Haviland ML. Confirming “truth”: more evidence of a successful tobacco countermarketing campaign in Florida. Am J Public Health . 2004; 94( 2): 255– 257. Google Scholar CrossRef Search ADS PubMed  10. Thrasher JF, Niederdeppe JD, Jackson C, Farrelly MC. Using anti-tobacco industry messages to prevent smoking among high-risk adolescents. Health Educ Res . 2006; 21( 3): 325– 337. Google Scholar CrossRef Search ADS PubMed  11. Leatherdale ST, Sparks R, Kirsh VA. Beliefs about tobacco industry (mal)practices and youth smoking behaviour: insight for future tobacco control campaigns (Canada). Cancer Causes Control . 2006; 17( 5): 705– 711. Google Scholar CrossRef Search ADS PubMed  12. McCool J, Paynter J, Scragg R. A cross-sectional study of opinions related to the tobacco industry and their association with smoking status amongst 14–15 year old teenagers in New Zealand. N Z Med J . 2011; 124( 1338): 34– 43. Google Scholar PubMed  13. Brown AK, Moodie C, Hastings G, Mackintosh AM, Hassan L, Thrasher J. The association of normative perceptions with adolescent smoking intentions. J Adolesc . 2010; 33( 5): 603– 614. Google Scholar CrossRef Search ADS PubMed  14. Choi WS, Gilpin EA, Farkas AJ, Pierce JP. Determining the probability of future smoking among adolescents. Addiction . 2001; 96( 2): 313– 323. Google Scholar CrossRef Search ADS PubMed  15. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Merritt RK. Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychol . 1996; 15( 5): 355– 361. Google Scholar CrossRef Search ADS PubMed  16. Chen J, Ho SY, Leung LT, Wang MP, Lam TH. Associations of unhappiness with sociodemographic factors and unhealthy behaviours in Chinese adolescents. Eur J Public Health . 2017; 27( 3): 518– 524. Google Scholar CrossRef Search ADS PubMed  17. Cummings P. Methods for estimating adjusted risk ratios. Stata J . 2009; 9( 2): 175– 196. 18. Lumley T, Kronmal R, Ma S. Relative risk regression in medical research: Models, contrasts, estimators, and algorithms. UW Biostatistics Working Paper Series. Working Paper 293. 2006. http://biostats.bepress.com/uwbiostat/paper293. Accessed October 2017. 19. Waller BJ, Cohen JE, Ashley MJ. Youth attitudes towards tobacco control: a preliminary assessment. Chronic Dis Can . 2004; 25( 3–4): 97– 100. Google Scholar PubMed  20. Thematic Household Survey Report No.59 . Hong Kong: Census and Statistics Department of HKSAR; 2013. http://www.digital21.gov.hk/eng/statistics/download/householdreport2016_59.pdf. Accessed October 2017. 21. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health . 1997; 87( 6): 968– 973. Google Scholar CrossRef Search ADS PubMed  22. Smoking (Public Health) (Notices) (Amendment) Order 2017 . Hong Kong: Legislative Council; 2017. http://www.legco.gov.hk/yr16-17/english/subleg/negative/2017ln066-e.pdf. Accessed October 2017. 23. White V, Tan N, Wakefield M, Hill D. Do adult focused anti-smoking campaigns have an impact on adolescents? The case of the Australian National Tobacco Campaign. Tob Control . 2003; 12( suppl 2): ii23– ii9. Google Scholar PubMed  24. Balbach ED, Glantz SA. Tobacco control advocates must demand high-quality media campaigns: the California experience. Tob Control . 1998; 7( 4): 397– 408. Google Scholar CrossRef Search ADS PubMed  25. Ibrahim JK, Glantz SA. Tobacco industry litigation strategies to oppose tobacco control media campaigns. Tob Control . 2006; 15( 1): 50– 58. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nicotine and Tobacco ResearchOxford University Press

Published: May 15, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off