The impact of multipolicy tobacco control legislation on smoking behavior

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 1 0 0 3 e1 0 0 6

Available online at www.sciencedirect.com

Public Health journal homepage: www.elsevier.com/puhe

Short Communication

The impact of multipolicy tobacco control legislation on smoking behavior Y.-W. Tsai a, H.-C. Chang a, S.-T. Chiou b,c,* a

Institute of Health and Welfare Policy, National Yang-Ming University, Taipei, Taiwan Health Promotion Administration, Ministry of Health and Welfare, New Taipei City, Taiwan c School of Medicine, National Yang-Ming University, Taipei, Taiwan b

article info Article history: Received 16 December 2013 Received in revised form 30 January 2015 Accepted 13 May 2015 Available online 24 June 2015

Introduction The Taiwan government enacted the Tobacco Hazards Prevention Amendment Act in 2009 (2009 Act), strengthening the existing legislations by implementing multiple tobacco control policies. The key changes included: (1) adding of graphic warnings on cigarette packages in January 2009; (2) extending of smoke-free areas in almost all enclosed workplaces and public places in January 2009; and (3) raising of tobacco health surcharges by NT $10 (US $0.30) per pack in June 2009. Before 2009, only text health warnings were shown on cigarette packets in Taiwan in 12-point font and covered 25% of the front of the packaging. After January 2009, the legislations required the warning labels to cover at least 35% of the package front and back, include one of six graphic warnings with vividly colored images, contain written warnings on the risks of smoking, and include the smoking quit-line telephone number.1 Prior to 2009, smoking was banned in public areas but allowed in indoor designated

smoking areas. The 2009 Act completely prohibits smoking in public areas with the exception of outdoor dining areas of restaurants and bars after 9 p.m. Promotional activities were held to ensure that business proprietors posted ‘No smoking’ signs at all entrances to their establishments.1 Increasing excise taxes and prices are effective in preventing people from starting to smoke, promoting cessation, and lowering the number of current smokers and cigarettes consumption.2,3 Before 2009, the average retail price of cigarettes in Taiwan was NT $55 (US $1.85)/pack. This included NT $34.16 (US $1.15)/pack for Taiwan cigarettes and NT $61.68 (US $2.08)/pack for imported cigarettes.4 On June 1st 2009, the government, as stipulated in the 2009 Act, raised the cigarette tax from NT $10 (US $0.34) to NT $20 (US $0.67) per pack. Consequently, the average cigarette retail price increased by 18% to approximately NT $65 (US $1.96) per pack.5 The new warnings and smoke-free policies have demonstrated a positive effect on intention to quit smoking among Taiwan smokers in the first three months;1 the implementation of the 2009 Act also increased quit attempt rate and annual cessation rate among recent smokers in the first year.6 To have an overall picture of the change of smoking behavior, we assessed (1) the impact of 2009 Act on smoking prevalence among Taiwan adults, and smoking intensity (average number of cigarettes smoked per day) among current smokers; and (2) factors attributable to smoking behavior. Data were retrieved from the 2007e2011 Taiwan Adult Smoking Behavior Surveys undertaken by the Health Promotion Administration aimed to monitor smoking behavior in Taiwan adults aged 18 or older. By county-based random

* Corresponding author. Health Promotion Administration, Ministry of Health and Welfare, No.2, Changqing St., Xinzhuang Dist., New Taipei City 242, Taiwan. Tel.: þ886 2 29978616; fax: þ886 2 2991 6328. E-mail address: [email protected] (S.-T. Chiou). http://dx.doi.org/10.1016/j.puhe.2015.05.013 0033-3506/© 2015 Published by Elsevier Ltd on behalf of The Royal Society for Public Health.

1004

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 1 0 0 3 e1 0 0 6

telephone interview, the surveys were administered to approximately 16,000 non-institutionalized residents (400e800 residents per county) each year from July to September. Current smokers were interviewees who had smoked more than 100 cigarettes in their lifetime, and smoked in the previous month. The non-current smokers included never-smokers (those who had never smoked or who had smoked fewer than 100 cigarettes in their lifetime) and former smokers (who had smoked over 100 cigarettes in their lifetime but had not smoked in the previous month). On exclusion of subjects with missing smoking status (0.3%), we obtained a total sample of 81,869 adults including 13,163 current smokers and 68,706 non-current smokers. Multivariate logistic regression analysis was conducted to examine the association between the 2009 Act and current smoking among all adults. We used two model specifications to analyze the correlations of smoking behavior. Model 1 included the dummy variable ‘Post 2009’ while controlling other independent variables to evaluate the overall effect after the 2009 Act. Model 2 included the four year dummy variables plus all other independent variables to evaluate the time-variant effects of the 2009 Act on smoking behavior. All statistical estimations were performed using the STATA version 11.0 software. Estimates were weighted to reflect the respondents' probability

of sampling selection and to be representative of the age-, gender-, area-, and education-specific populations, based on the 2000 Taiwanese census.

General statistics During the period 2007 to 2011, current smoking prevalence dropped from 22.3% to 18.17% (P < 0.0001); there was no significant change in smoking intensity among current smokers (dropped from 19.3 cigarettes per day to 19.0 cigarettes, P ¼ 0.4554). Over the five-year period, males had much higher current smoking prevalence rate than females (35.4% vs 4.3%, P < 0.0001). On average, male smokers smoked 19.8 cigarettes/ day, significantly higher than that of females at 12.4 cigarettes/day (P < 0.0001). Current smoking prevalence rates (over 20%, P < 0.0001) were higher in age groups between 25 and 54 years. Adults with junior high school and high school education level had the highest current smoking prevalence rates (31.7% and 24.6%, P < 0.0001). For employment, the employed subjects had higher smoking rates (26.9% vs 12.3%, P < 0.0001) than the unemployed. For marital status, single adults had higher current smoking prevalence rate (23.6% vs 19.0%, P < 0.0001) than the married and the divorced/widowed/ separated. The above results are not shown in the table.

Table 1 e Logistic regressions on smoking among Taiwanese adults aged 18 years or above, 2007e2011. Model 1 OR Period Pre-act (2007e2008) (Ref) Post-act (2009e2011) Year 2007 (Ref) 2008 2009 2010 2011 Gender Female (Ref) Male Age group in years 18e24 (Ref) 25e34 35e44 45e54 55e64 65 Education degree Undergraduate or above (Ref) High school Junior high school Elementary or below Employment Unemployed (Ref) Employed Marital status Single (Ref) Married Divorced/widowed/separated

Model 2 P-value

OR

e 0.000

e e

e e e e e

e 0.96 0.87 0.83 0.78

e 11.58e15.79

e 0.000

e 13.51

e 2.54 2.38 1.42 0.91 0.60

e 2.22e2.91 2.01e2.82 1.17e1.73 0.76e1.08 0.48e0.73

e 0.000 0.000 0.001 0.274 0.000

e 2.48 4.10 3.17

e 2.32e2.65 3.64e4.62 2.76e3.64

e 1.45 e 0.94 1.77

e 0.84 e e e e e e 13.52

95% CI e 0.79e0.90 e e e e e

95% CI e e

P-value e e

e 0.90e1.02 0.80e0.94 0.75e0.91 0.71e0.85

e 0.164 0.001 0.000 0.000

e 11.58e15.76

e 0.000

e 2.54 2.38 1.42 0.91 0.60

e 2.21e2.91 2.01e2.82 1.17e1.73 0.76e1.09 0.48e0.74

e 0.000 0.000 0.001 0.280 0.000

e 0.000 0.000 0.000

e 2.48 4.10 3.17

e 2.32e2.65 3.64e4.62 2.76e3.64

e 0.000 0.000 0.000

e 1.32e1.59

e 0.000

e 1.45

e 1.33e1.59

e 0.000

e 0.85e1.04 1.52e2.07

e 0.223 0.000

e 0.94 1.76

e 0.84e1.04 1.51e2.06

e 0.208 0.000

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 1 0 0 3 e1 0 0 6

Multivariate logistic regression: post Act smoking prevalence and factors contributed to the change Table 1 shows the multivariate logistic regression results of current smoking prevalence among all adults. Model 1 found that smoking prevalence decreased in the post-Act when compared with the pre-Act period (OR ¼ 0.84, 95% CI ¼ 0.79e0.90). Smoking prevalence was higher among adults with lower education levels than those with an undergraduate degree or higher (high school: OR ¼ 2.48, 95% CI ¼ 2.32e2.65; junior high school: OR 4.10, 95% CI ¼ 3.64e4.62; elementary school or below: OR ¼ 3.17, 95% CI ¼ 2.76e3.64). In Model 2, three of the four year dummy variables were significant. Compared with the reference year 2007, the 2008 mass media campaign on smoke-free policy did not reduce smoking prevalence significantly (OR ¼ 0.96, 95% CI ¼ 0.90e1.02). In the post-Act period, the odds of smoking reduced significantly in year 2009 (OR ¼ 0.87, 95% CI ¼ 0.80e0.94), 2010 (OR ¼ 0.83, 95% CI ¼ 0.75e0.91) and 2011 (OR ¼ 0.78, 95% CI ¼ 0.71e0.85). The marginal decreasing on odds was significant only between the year 2008 and 2009 (OR ¼ 0.90, 95% CI ¼ 0.84e0.98). These results suggest that the 2009 Act was effective in reducing smoking prevalence and durable in the following years. Smoking prevalence was higher in the 25e34, 35e44, and 45e54 year age groups when compared to the youngest age group (18e24 years old), and no significant difference was observed between the 55e64 year age group and the youngest group in both models. Employment and divorced/widowed/ separated status were significantly associated with higher smoking prevalence, and the over 65s had significantly lower smoking prevalence compared to the youngest age group. The results from both models were consistent and robust.

1005

warning labels might have played a greater role in the delivery of health information. Further study of the 2009 Act on smoking prevalence and intensity in interaction with education levels may reveal underlying effects of the 2009 Act on health inequality caused by smoking. This study has several limitations. First, the cross-sectional nature of this study does not allow the analysis of cause effect relationships. Second, we did not analyze how the smokers responded to the 2009 Act because of their economic status (the Taiwan Adult Smoking Behavior Surveys studied overall household income without information on family size). As the unadjusted family income information is highly correlated to family size, the individual economic capability could not be accurately revealed in this study. Finally, the sample drawn from telephone interview survey might not be representative of the entire population. Nevertheless, because the Taiwanese Adult Smoking Behavior Surveys were conducted using a consistent sampling scheme, the systematic difference over time should have reflected the changes in smoking behavior in response to the 2009 Act. This study assessed the impact of tobacco control legislation involving multiple tobacco control measures on smoking prevalence among all adults and smoking intensity among current smokers. The 2009 Act reduced smoking prevalence among Taiwanese adults sustainably, but did not reduce smoking intensity among current smokers. Strengthening of current policies or the implementation of different initiatives is necessary.

Author statements Ethical approval Review of Institutional Review Board for this study was not required according to the regulations.

Effects of the 2009 Act and sustainability Funding Increasing excise taxes and prices of tobacco products have been proved effective in reducing smoking prevalence and consumption.2,3 However, the effectiveness of warning messages, images or texts,7 and smoke-free policies on the reduction of smoking prevalence remained inconclusive.8 Also, there is little research on how multipolicy tobacco control legislation affects smoking behavior. Our study used a large national interview survey database to examine the three-year impact of the 2009 Act on smoking behavior. We showed that the 2009 Act significantly decreased smoking prevalence rates and the effect sustained in the following years. The finding that Taiwan adults with lower education levels were more likely to be current smokers is an issue of concern in reducing health inequality. Although a previous study has showed that smokers with lower education levels in Taiwan were less aware of smoke-related legislation,9 cigarette tax and price increase of the new act were possibly influential to them because low education level tends to be highly correlated with low income, and low income smokers are more responsive to price change. Also, given that smokers with lower education levels are generally less aware of the smoking health consequences and literacy limited, graphic

None declared.

Competing interests The authors declare that they have no competing interests.

references

1. Chang FC, Chung CH, Yu PT, Chao KY. The impact of graphic cigarette warning labels and smoke-free law on health awareness and thoughts of quitting in Taiwan. Health Educ Res 2011;26:179e91. 2. Chaloupka FJ, Straif K, Leon ME. Effectiveness of tax and price policies in tobacco control. Tob Control 2011;20:235e8. 3. Wilson LM, Avila Tang E, Chander G, Hutton HE, Odelola OA, Elf JL, et al. Impact of tobacco control interventions on smoking initiation, cessation, and prevalence: a systematic review. J Environ Public Health 2012;2012:961724. 4. Chen SH, Lee JM, Liu HH, Wang HC, Ye CY. The cross-effects of cigarette and betel nut consumption in Taiwan: have tax increases made a difference? Health Policy Plan 2011;26:266e73.

1006

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 1 0 0 3 e1 0 0 6

5. Lee JM, Chen MG, Hwang TC, Yeh CY. Effect of cigarette taxes on the consumption of cigarettes, alcohol, tea and coffee in Taiwan. Public Health 2010;124:429e36. 6. Chang FC, Sung HY, Zhu SH, Chiou ST. Impact of the 2009 Taiwan tobacco hazards prevention act on smoking cessation. Addiction 2014;109:140e6. 7. Partos TR, Borland R, Yong HH, Thrasher J, Hammond D. Cigarette packet warning labels can prevent relapse: findings from the international tobacco control 4-country policy evaluation cohort study. Tob Control 2013;22:e43e50.

8. Bajoga U, Lewis S, McNeill A, Szatkowski L. Does the introduction of comprehensive smoke-free legislation lead to a decrease in population smoking prevalence? Addiction 2011;106:1346e54. 9. Chang FC, Chung CH, Chuang YC, Hu TW, Yu PT, Chao KY, et al. Effect of media campaigns and smoke-free ordinance on public awareness and secondhand smoke exposure in Taiwan. J Health Commun 2011;16:343e58.