Nicotine & Tobacco Research, Volume 16, Number 4 (April 2014) 390–396
Original Investigation
Household Smoking Restrictions Related to Secondhand Smoke Exposure in Guangdong, China: A Population Representative Survey Xiaolin Wei MD, PhD1,2, Zhenzhen Zhang MPH1, Xiuling Song MSc3, Yanjun Xu MSc3, Wei Wu PhD3,4, Xiangqian Lao PhD1,2, Wenjun Ma MD, MPH3,4 1School of Public Health and Primary Care, Chinese University of Hong Kong, Hong Kong, China; 2Shenzhen Municipal Key Laboratory for Health Risk Analysis, CUHK Shenzhen Research Institute, Shenzhen, China; 3Guangdong Provincial Institute of Public Health, Guangzhou, China; 4Guangdong Provincial Centre for Disease Control and Prevention, Guangzhou, China
Received June 21, 2013; accepted September 6, 2013
Abstract Introduction: China has a higher household secondhand smoke exposure rate than other countries. This study aims to estimate the prevalence rate of households implementing smoking bans in Guangdong and to identify factors correlated with household smoking bans. Methods: A cross-sectional, stratified random cluster sampling survey was conducted in Guangdong in 2010. A total of 2,114 adults aged 15 and older completed the face-to-face interviews with a response rate of 70%. The survey employed an adapted and validated questionnaire from the China Global Adult Tobacco Survey. Household smoking policy was divided into 3 groups: full ban, partial ban, and no ban. A multiple logistic regression model was employed to explore factors related to a full household smoking ban. Results: The survey found 14.2% of respondents reported a full ban, 23.6% reported a partial ban, and 62.2% reported no ban of smoking at home. Current smoking status was the strongest predictor for less restrictive household smoking policies (odds ratio [OR] = 4.9, 95% CI = 2.634–8.999). Our study suggested that people with a high level of education were more likely to implement a full household smoking ban (OR = 4.4, 95% CI = 2.388–8.178). Additionally, urban residents were significantly more likely to report a full household smoking ban than rural residents (OR = 1.67, 95% CI = 1.202–2.322). Conclusions: Household smoking bans were not sufficiently established in Guangdong, China. Intensified efforts were called to promote home smoking bans, especially for those with a lower education level, with lower income, and living in rural areas.
Introduction Adverse effects of exposure to secondhand smoke (SHS) on nonsmoking adults and children have been widely reported, which include increased risk of premature death and illness such as cardiovascular disease and lung cancer in adults and children (U.S. Environmental Protection Agency, 1993; Murray, Britton, & Leonardi-Bee, 2012). SHS is classified as a Group A carcinogen by the U.S. Environmental Protection Agency (1993), thus, any exposure to SHS is harmful. China is the largest producer and consumer of cigarettes in the world (China CDC, 2010). The 2010 China Global Adult Tobacco Survey (GATS) reported that China has 301 million smokers, and 556 million nonsmokers exposed to SHS at least one day a week (China CDC, 2010). A 2002 national survey reported a 53% SHS exposure rate for nonsmokers in China (Yang, Ma,
Liu, & Zhou, 2005). SHS exposure rates were 48% among six rural counties in 2004 (Wang et al., 2009), and up to 72.4% from the 2010 GATS. SHS is responsible for nearly 100,000 deaths annually in China (China CDC, 2010). Guangdong is the most populous province in China, with a population of 104 million in 2010. In 2010, Guangzhou, the capital of Guangdong, and Jiangmen pioneered the implementation of smoking bans in public areas including hospitals, schools, restaurants, supermarkets, and government offices (China CDC, 2010). Despite these public efforts, SHS at home is often neglected. Household SHS is the primary source of exposure to tobacco smoke for nonsmokers (Merom & Rissel, 2001). The proximity, intensity, and duration of SHS exposure at home are often greater than that in public settings (Martinez-Sanchez et al., 2010). The China GATS reported that the prevalence of SHS exposure was 67% at home, higher than that in workplaces
doi:10.1093/ntr/ntt162 Advance Access publication October 15, 2013 © The Author 2013. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: [email protected].
390
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
Corresponding Author: Wenjun Ma MD, MPH, Guangdong Provincial Institute of Public Health, Guangdong Provincial Centre for Disease Control and Prevention, Guangzhou, Guangdong 511430, China. Telephone/Fax: 86-20-31051602; E-mail: mwj68@ vip.tom.com
Nicotine & Tobacco Research (63%), government buildings (58%), and schools (37%) (China CDC, 2010). Furthermore, China had much higher reported rates of household SHS exposure compared with developed countries such as the United States (Burton, Baker, Mastey, & Mannino, 2010) and Australia (Merom & Rissel, 2001), and some developing countries such as Mexico and India (King, Mirza, & Babb, 2012). Home smoking bans are rules set up by household residents to restrict smoking inside the household. Studies in developed countries reported that smoking bans at home are effective in reducing exposure to SHS and cigarette consumption (Pizacani et al., 2003), promoting quitting behaviors and preventing smoking initiation (Martinez-Donate et al., 2007; Pizacani et al., 2003; Shelley, Fahs, Yerneni, Qu, & Burton, 2006; Shelley, Yerneni, Hung, Das, & Fahs, 2007). Home bans on smoking are more effective in increasing cessation than bans in the workplace (Shelley et al., 2007). However, smoking bans have not been widely adopted in developing countries (Abdullah et al., 2012; Ji et al., 2009). Specifically, little information is known regarding SHS exposure at home in rural areas of developing countries, despite rural areas generally having fewer smoking restrictions at home (Berg et al., 2006; Yao, Sung, Mao, Hu, & Max, 2012). In addition, no study has reported factors influencing smoking bans in homes of developing countries. This study aims to estimate the prevalence of home smoking bans in Guangdong, China, through a population representative survey. The second objective is to identify the correlations of smoking bans and SHS at home. This study also provides policy recommendations for tailored interventions to promote smoking bans in Chinese homes.
skills. The survey was conducted in October 2010. Standard operation procedures were employed in the survey covering the definition of each questionnaire, how to ask and explain questions, and how to record results. The provincial CDC formed a supervision team in which one supervisor worked with each city team during the survey period. The supervisor checked all questionnaires on-site. Any questionnaire with more than 5% missing variables, any logistic error, and unclear answers were returned to the research assistant to conduct a second interview with the same participant.
Methods
Smoking Status We employed the international case definition of current smokers as those who smoked at the time of the survey. Nonsmokers included those who never smoked or quitted smoking at the time of survey.
A cross-sectional household-based survey aimed to collect 3,000 face-to-face interviews of adults who were 15 years or older and who resided in Guangdong for more than 6 months before the survey. The survey employed a multistage, stratified random cluster sampling. The province was divided into five geographical regions: the east, the west, the south, the north, and the central. One prefecture in each region was randomly selected. One urban district and one rural county were randomly selected within the prefecture. In the district/county, six clusters were randomly selected. We defined the sampling cluster as a neighborhood in urban areas or a village in rural areas. Within each cluster, 50 households were randomly selected from the local household registration. At the last stage, one individual, normally aged 15 and above, was randomly invited to complete the questionnaire. Respondents whose primary place of residence was a military base or an institution were excluded. Written informed consents were collected before the interview from all participants aged 16 or above. Informed consent forms were also collected from parents or guardians of those aged 15. Ethical approval was obtained from the Ethical Committee of the Guangdong Provincial Centre for Disease Control and Prevention. A team of research assistants in each prefecture’s Centre for Disease Control (CDC) received 1-week training on how to conduct the survey, definitions of questions, and survey
The questionnaire was adapted from a validated tool used in the China GATS, including topics on smoking behavior, knowledge and attitudes toward tobacco use, and household and workplace smoking policies. Specific definitions are listed below. Sociodemographic Information We collected information on gender, age, the location of residence (urban/rural), education (primary school or less, attended secondary school, high school graduate, and college or above), household income (low, middle, and high), and occupation. We categorized household monthly income into three levels: low (less than 1,000 RMB, US$152), middle (1,000–5,000 RMB, US$152–US$758), and high (more than 5,000 RMB, >US$758) according to Guangdong’s official poverty line. Occupations were categorized as manual laborers, business or service employees, government official or clerks, professionals, and others (including retired and unemployed) according to the provincial definitions.
Household Smoking Policy Three choices for smoking ban policy at home were given, including full ban, partial ban, and no ban. Full ban meant that no one was allowed to smoke anywhere and anytime at home. Partial ban meant that smoking was allowed in certain areas and/or at certain times, while no ban meant that smoking was allowed anywhere and at anytime at home. Household Exposure to SHS Exposure to SHS is defined as having someone smoke in the same closed area at home daily or less frequently. This does not include open areas at home such as courtyards or open corridors. Knowledge and Attitudes Knowledge of the health risk from SHS exposure was measured by one question: Is it true or false that “breathing smoke from other people’s cigarettes can cause serious illness”? Attitudes were solicited toward smoking restrictions in public spaces including hospitals, schools, offices, public transportations, restaurants, and bars. Respondents were given the choices of “full ban smoking,” “partial ban smoking,” “no ban
391
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
Data Source and Sampling
Measures
Household smoking restrictions smoking,” or “do not care.” In addition, to measure attitudes toward smoke-free policy in indoor public places, respondents were asked their level of agreement on a 5-point Likert scale (strongly agree, agree, do not care, disagree, and strongly disagree) with the following statements: “All indoor public places should ban smoking without setting a smoking room,” and “Single office should be smoke-free.” Higher scores represented more positive attitudes toward smoke-free policy in public places. Statistical Analysis
Results Population A total of 2,114 participants successfully completed the survey, with a response rate of 70.5%. The average age of respondents was 40.2 years. Among participants, 52.4% were men and 51.3% were from rural areas. The majority of participants (58.7%) had education level lower than high school. Nearly half (46.2%) were manual workers. In all, 41% had a low level of household income, that is, less than 1,000 CNY (about US$152) per month, whereas 12% had a high level of income, that is, above 5,000 CNY (>US$758) (Table 1). In all, 505 (23.9%) participants were current smokers. Smoking rate was significantly higher in men (47.1%) than in women (2.9%). Over half of respondents (55.2%) agreed that SHS is harmful to their health. Home Smoking Policy Overall, 14.2% (301) of participants reported a full ban that smoking was not allowed anywhere and anytime in the house, whereas 23.6% of households had a partial ban and 62.2% households had no ban of smoking at home. Urban households tended to have a full smoking ban compared with rural households. Respondents who were younger, female, highly educated, professionals, and nonsmokers were more likely to report a full ban at home (p < .05). In addition, households with lower income were more relaxed on smoking at home compared with those with higher income (p < .001). Only 2.6% of
392
Enforcement of Household Smoking Policy Table 2 illustrates that household smoking policy was significantly related to exposure to SHS in homes (p < .001). Respondents living in a home with a full smoking ban reported no exposure to SHS. Of the 486 who reported a partial ban at home, 55.1% reported that someone had smoked at home. Of the 1,164 who reported no ban at home, 76.7% reported exposure to SHS at home. Of the 474 current smokers, 79% with a partial ban reported that someone including themselves had smoked at home compared with 92.8% with no ban. In other words, only 7.2% of smokers who lived in a home with no smoking ban reported not smoking at home compared with 21% of those living in a home with a partial ban. Similar results were found among nonsmokers. Of the 1,176 nonsmokers, 49% who lived in a partial ban reported that someone had smoked in the home compared with 69.1% of those living in a home with no smoking restrictions. Multivariate Associations With Household Smoking Restrictions The logistic regression model was statistically significant (model coefficient tests, p < .05) and showed good model fitness (Hosmer and Lameshow tests, p > .90). After controlling socioeconomic and demographic factors of respondents, smoking status of the respondents had the highest predictive power in whether or not a full household smoking ban was reported (OR = 4.9, 95% CI = 2.634–8.999). Respondents with an education level of college or above were 4.4 times likely to report a full smoking ban at home compared with people with only primary-level education (OR = 4.4, 95% CI = 2.388–8.178). In addition, urban residents were significantly more likely to report a full household smoking ban than rural residents (OR = 1.7, 95% CI = 1.202–2.322). Middle-income individuals were more likely to report a full smoking ban compared with low-income individuals (OR = 1.676, 95% CI = 1.134–2.478). People who understood the dangers of SHS exposure and those who agreed to ban smoking in public places were more likely to report a full smoking ban in homes (OR = 2.5, 95% CI = 1.359–4.464; OR = 1.2, 95% CI = 1.121–1.225, respectively) (Table 3).
Discussion The study found that 14.2% of the survey respondents in Guangdong, China, had adopted a full smoking ban at home in 2010. This rate was lower than the 26% reported in Shanghai urban residents (Ji et al., 2009), and the 66% reported among
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
All analyses were performed using SPSS 18.0. We used p = .05 as the level for statistical significance. Standard descriptive statistics were used to summarize data collected for demographics, smoking-related characteristics, and household smoking policies. The chi-square test was used to examine the relationship between household smoking ban and other variables regarding demographic information and smoking status. Independent-sample t tests were used to compare continuous variables. We used multiple logistic regression analyses to estimate predictors of smoking policies at home. The dependent variable was whether or not a full smoking ban was adopted at home, whereas the independent variables included educational level, household income, gender, smoking status, occupation, knowledge of SHS risk, and attitudes toward smoking bans in public places. Variables were included in the model if they were significant at a p value of .05 or less in the bivariate analysis. The adjusted odds ratios (ORs) with confidence intervals were reported.
current smokers reported a complete household smoking ban at home compared with 17.9% of nonsmokers doing so. Household smoking policy was related to people’s knowledge regarding the consequences of SHS exposure and attitudes toward smoking restrictions in public places. In all, 20.3% of respondents who agreed that SHS exposure was harmful reported a full smoking ban at home, whereas only 6.8% of those who disagreed had a full smoking ban at home. Conversely, respondents who reported the presence of a full ban at home supported implementing smoking restrictions in public places more than those who reported a partial ban or no ban at home.
Nicotine & Tobacco Research Table 1. Distributions of Different Smoking Policy in Household by Demographics and Smoking-Related Characteristics Smoking policies in household Characteristics
Partial ban (%)
No ban (%)
Total
301 (14.2)
498 (23.6)
1,315 (62.2)
2,114
134 (12.1) 167 (16.6)
246 (22.2) 252 (25)
728 (65.7) 587 (58.3)
1,108 (52.4) 1,006 (47.6)
216 (21) 85 (7.8)
304 (29.5) 194 (17.9)
509 (49.5) 806 (74.3)
1,029 (48.7) 1,085 (51.3)
30 (5.6) 55 (7.8) 95 (18.6) 121 (33.7)
72 (13.4) 158 (22.4) 131 (25.5) 137 (38.2)
434 (81) 492 (69.8) 288 (56) 101 (28.1)
536 (25.4) 705 (33.3) 514 (24.3) 359 (17)
75 (7.7) 39 (14.9) 41 (28.1) 74 (30.1) 72 (14.9)
181 (18.5) 68 (26.1) 53 (36.3) 68 (27.5) 128 (26.5)
721 (73.8) 154 (59) 52 (35.6) 105 (42.5) 283 (58.6)
977 (46.2) 261 (12.3) 146 (6.9) 247 (11.7) 483 (22.8)
54 (6.2) 200 (19.9)
114 (13.2) 300 (29.9)
698 (80.6) 504 (50.2)
866 (41) 1,004 (47.5)
47 (19.3)
84 (34.4)
113 (46.3)
244 (11.5)
13 (2.6) 288 (17.9)
100 (19.8) 398 (24.7)
392 (77.6) 923 (57.4)
505 (23.9) 1,608 (76.1)
237 (20.3) 64 (6.8) 38.69 22.95
350 (30) 148 (15.7) 39.28 21.32
579 (49.7) 733 (77.6) 40.93 18.39
1,166 (55.2) 945 (44.8) 40.2
Note. SHS = secondhand smoke. Conversion rate is USD1.00 = RMB 6.59, www.oanda.com, January 1, 2011. aContinuous variables. *p ≤ .05. **p ≤ .001.
Table 2. Number and Percentage of Respondents Reporting Exposure to Household SHS by Household Smoking Policy Household smoking policy Variables Home exposure to SHS Yes No Total (n = 1,650) Home exposure to SHS (smokers) Yes No Total (n = 474) Home exposure to SHS (nonsmokers) Yes No Total (n = 1,176)
Partial ban (%)
No ban (%)
268 (55.1) 218 (44.9) 486
893 (76.7) 271 (23.3) 1,164
79 (79) 21 (21) 100
347 (92.8) 27 (7.2) 374
189 (49) 197 (51) 386
546 (69.1) 244 (30.9) 790
p value
Original Investigation
Household Smoking Restrictions Related to Secondhand Smoke Exposure in Guangdong, China: A Population Representative Survey Xiaolin Wei MD, PhD1,2, Zhenzhen Zhang MPH1, Xiuling Song MSc3, Yanjun Xu MSc3, Wei Wu PhD3,4, Xiangqian Lao PhD1,2, Wenjun Ma MD, MPH3,4 1School of Public Health and Primary Care, Chinese University of Hong Kong, Hong Kong, China; 2Shenzhen Municipal Key Laboratory for Health Risk Analysis, CUHK Shenzhen Research Institute, Shenzhen, China; 3Guangdong Provincial Institute of Public Health, Guangzhou, China; 4Guangdong Provincial Centre for Disease Control and Prevention, Guangzhou, China
Received June 21, 2013; accepted September 6, 2013
Abstract Introduction: China has a higher household secondhand smoke exposure rate than other countries. This study aims to estimate the prevalence rate of households implementing smoking bans in Guangdong and to identify factors correlated with household smoking bans. Methods: A cross-sectional, stratified random cluster sampling survey was conducted in Guangdong in 2010. A total of 2,114 adults aged 15 and older completed the face-to-face interviews with a response rate of 70%. The survey employed an adapted and validated questionnaire from the China Global Adult Tobacco Survey. Household smoking policy was divided into 3 groups: full ban, partial ban, and no ban. A multiple logistic regression model was employed to explore factors related to a full household smoking ban. Results: The survey found 14.2% of respondents reported a full ban, 23.6% reported a partial ban, and 62.2% reported no ban of smoking at home. Current smoking status was the strongest predictor for less restrictive household smoking policies (odds ratio [OR] = 4.9, 95% CI = 2.634–8.999). Our study suggested that people with a high level of education were more likely to implement a full household smoking ban (OR = 4.4, 95% CI = 2.388–8.178). Additionally, urban residents were significantly more likely to report a full household smoking ban than rural residents (OR = 1.67, 95% CI = 1.202–2.322). Conclusions: Household smoking bans were not sufficiently established in Guangdong, China. Intensified efforts were called to promote home smoking bans, especially for those with a lower education level, with lower income, and living in rural areas.
Introduction Adverse effects of exposure to secondhand smoke (SHS) on nonsmoking adults and children have been widely reported, which include increased risk of premature death and illness such as cardiovascular disease and lung cancer in adults and children (U.S. Environmental Protection Agency, 1993; Murray, Britton, & Leonardi-Bee, 2012). SHS is classified as a Group A carcinogen by the U.S. Environmental Protection Agency (1993), thus, any exposure to SHS is harmful. China is the largest producer and consumer of cigarettes in the world (China CDC, 2010). The 2010 China Global Adult Tobacco Survey (GATS) reported that China has 301 million smokers, and 556 million nonsmokers exposed to SHS at least one day a week (China CDC, 2010). A 2002 national survey reported a 53% SHS exposure rate for nonsmokers in China (Yang, Ma,
Liu, & Zhou, 2005). SHS exposure rates were 48% among six rural counties in 2004 (Wang et al., 2009), and up to 72.4% from the 2010 GATS. SHS is responsible for nearly 100,000 deaths annually in China (China CDC, 2010). Guangdong is the most populous province in China, with a population of 104 million in 2010. In 2010, Guangzhou, the capital of Guangdong, and Jiangmen pioneered the implementation of smoking bans in public areas including hospitals, schools, restaurants, supermarkets, and government offices (China CDC, 2010). Despite these public efforts, SHS at home is often neglected. Household SHS is the primary source of exposure to tobacco smoke for nonsmokers (Merom & Rissel, 2001). The proximity, intensity, and duration of SHS exposure at home are often greater than that in public settings (Martinez-Sanchez et al., 2010). The China GATS reported that the prevalence of SHS exposure was 67% at home, higher than that in workplaces
doi:10.1093/ntr/ntt162 Advance Access publication October 15, 2013 © The Author 2013. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: [email protected].
390
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
Corresponding Author: Wenjun Ma MD, MPH, Guangdong Provincial Institute of Public Health, Guangdong Provincial Centre for Disease Control and Prevention, Guangzhou, Guangdong 511430, China. Telephone/Fax: 86-20-31051602; E-mail: mwj68@ vip.tom.com
Nicotine & Tobacco Research (63%), government buildings (58%), and schools (37%) (China CDC, 2010). Furthermore, China had much higher reported rates of household SHS exposure compared with developed countries such as the United States (Burton, Baker, Mastey, & Mannino, 2010) and Australia (Merom & Rissel, 2001), and some developing countries such as Mexico and India (King, Mirza, & Babb, 2012). Home smoking bans are rules set up by household residents to restrict smoking inside the household. Studies in developed countries reported that smoking bans at home are effective in reducing exposure to SHS and cigarette consumption (Pizacani et al., 2003), promoting quitting behaviors and preventing smoking initiation (Martinez-Donate et al., 2007; Pizacani et al., 2003; Shelley, Fahs, Yerneni, Qu, & Burton, 2006; Shelley, Yerneni, Hung, Das, & Fahs, 2007). Home bans on smoking are more effective in increasing cessation than bans in the workplace (Shelley et al., 2007). However, smoking bans have not been widely adopted in developing countries (Abdullah et al., 2012; Ji et al., 2009). Specifically, little information is known regarding SHS exposure at home in rural areas of developing countries, despite rural areas generally having fewer smoking restrictions at home (Berg et al., 2006; Yao, Sung, Mao, Hu, & Max, 2012). In addition, no study has reported factors influencing smoking bans in homes of developing countries. This study aims to estimate the prevalence of home smoking bans in Guangdong, China, through a population representative survey. The second objective is to identify the correlations of smoking bans and SHS at home. This study also provides policy recommendations for tailored interventions to promote smoking bans in Chinese homes.
skills. The survey was conducted in October 2010. Standard operation procedures were employed in the survey covering the definition of each questionnaire, how to ask and explain questions, and how to record results. The provincial CDC formed a supervision team in which one supervisor worked with each city team during the survey period. The supervisor checked all questionnaires on-site. Any questionnaire with more than 5% missing variables, any logistic error, and unclear answers were returned to the research assistant to conduct a second interview with the same participant.
Methods
Smoking Status We employed the international case definition of current smokers as those who smoked at the time of the survey. Nonsmokers included those who never smoked or quitted smoking at the time of survey.
A cross-sectional household-based survey aimed to collect 3,000 face-to-face interviews of adults who were 15 years or older and who resided in Guangdong for more than 6 months before the survey. The survey employed a multistage, stratified random cluster sampling. The province was divided into five geographical regions: the east, the west, the south, the north, and the central. One prefecture in each region was randomly selected. One urban district and one rural county were randomly selected within the prefecture. In the district/county, six clusters were randomly selected. We defined the sampling cluster as a neighborhood in urban areas or a village in rural areas. Within each cluster, 50 households were randomly selected from the local household registration. At the last stage, one individual, normally aged 15 and above, was randomly invited to complete the questionnaire. Respondents whose primary place of residence was a military base or an institution were excluded. Written informed consents were collected before the interview from all participants aged 16 or above. Informed consent forms were also collected from parents or guardians of those aged 15. Ethical approval was obtained from the Ethical Committee of the Guangdong Provincial Centre for Disease Control and Prevention. A team of research assistants in each prefecture’s Centre for Disease Control (CDC) received 1-week training on how to conduct the survey, definitions of questions, and survey
The questionnaire was adapted from a validated tool used in the China GATS, including topics on smoking behavior, knowledge and attitudes toward tobacco use, and household and workplace smoking policies. Specific definitions are listed below. Sociodemographic Information We collected information on gender, age, the location of residence (urban/rural), education (primary school or less, attended secondary school, high school graduate, and college or above), household income (low, middle, and high), and occupation. We categorized household monthly income into three levels: low (less than 1,000 RMB, US$152), middle (1,000–5,000 RMB, US$152–US$758), and high (more than 5,000 RMB, >US$758) according to Guangdong’s official poverty line. Occupations were categorized as manual laborers, business or service employees, government official or clerks, professionals, and others (including retired and unemployed) according to the provincial definitions.
Household Smoking Policy Three choices for smoking ban policy at home were given, including full ban, partial ban, and no ban. Full ban meant that no one was allowed to smoke anywhere and anytime at home. Partial ban meant that smoking was allowed in certain areas and/or at certain times, while no ban meant that smoking was allowed anywhere and at anytime at home. Household Exposure to SHS Exposure to SHS is defined as having someone smoke in the same closed area at home daily or less frequently. This does not include open areas at home such as courtyards or open corridors. Knowledge and Attitudes Knowledge of the health risk from SHS exposure was measured by one question: Is it true or false that “breathing smoke from other people’s cigarettes can cause serious illness”? Attitudes were solicited toward smoking restrictions in public spaces including hospitals, schools, offices, public transportations, restaurants, and bars. Respondents were given the choices of “full ban smoking,” “partial ban smoking,” “no ban
391
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
Data Source and Sampling
Measures
Household smoking restrictions smoking,” or “do not care.” In addition, to measure attitudes toward smoke-free policy in indoor public places, respondents were asked their level of agreement on a 5-point Likert scale (strongly agree, agree, do not care, disagree, and strongly disagree) with the following statements: “All indoor public places should ban smoking without setting a smoking room,” and “Single office should be smoke-free.” Higher scores represented more positive attitudes toward smoke-free policy in public places. Statistical Analysis
Results Population A total of 2,114 participants successfully completed the survey, with a response rate of 70.5%. The average age of respondents was 40.2 years. Among participants, 52.4% were men and 51.3% were from rural areas. The majority of participants (58.7%) had education level lower than high school. Nearly half (46.2%) were manual workers. In all, 41% had a low level of household income, that is, less than 1,000 CNY (about US$152) per month, whereas 12% had a high level of income, that is, above 5,000 CNY (>US$758) (Table 1). In all, 505 (23.9%) participants were current smokers. Smoking rate was significantly higher in men (47.1%) than in women (2.9%). Over half of respondents (55.2%) agreed that SHS is harmful to their health. Home Smoking Policy Overall, 14.2% (301) of participants reported a full ban that smoking was not allowed anywhere and anytime in the house, whereas 23.6% of households had a partial ban and 62.2% households had no ban of smoking at home. Urban households tended to have a full smoking ban compared with rural households. Respondents who were younger, female, highly educated, professionals, and nonsmokers were more likely to report a full ban at home (p < .05). In addition, households with lower income were more relaxed on smoking at home compared with those with higher income (p < .001). Only 2.6% of
392
Enforcement of Household Smoking Policy Table 2 illustrates that household smoking policy was significantly related to exposure to SHS in homes (p < .001). Respondents living in a home with a full smoking ban reported no exposure to SHS. Of the 486 who reported a partial ban at home, 55.1% reported that someone had smoked at home. Of the 1,164 who reported no ban at home, 76.7% reported exposure to SHS at home. Of the 474 current smokers, 79% with a partial ban reported that someone including themselves had smoked at home compared with 92.8% with no ban. In other words, only 7.2% of smokers who lived in a home with no smoking ban reported not smoking at home compared with 21% of those living in a home with a partial ban. Similar results were found among nonsmokers. Of the 1,176 nonsmokers, 49% who lived in a partial ban reported that someone had smoked in the home compared with 69.1% of those living in a home with no smoking restrictions. Multivariate Associations With Household Smoking Restrictions The logistic regression model was statistically significant (model coefficient tests, p < .05) and showed good model fitness (Hosmer and Lameshow tests, p > .90). After controlling socioeconomic and demographic factors of respondents, smoking status of the respondents had the highest predictive power in whether or not a full household smoking ban was reported (OR = 4.9, 95% CI = 2.634–8.999). Respondents with an education level of college or above were 4.4 times likely to report a full smoking ban at home compared with people with only primary-level education (OR = 4.4, 95% CI = 2.388–8.178). In addition, urban residents were significantly more likely to report a full household smoking ban than rural residents (OR = 1.7, 95% CI = 1.202–2.322). Middle-income individuals were more likely to report a full smoking ban compared with low-income individuals (OR = 1.676, 95% CI = 1.134–2.478). People who understood the dangers of SHS exposure and those who agreed to ban smoking in public places were more likely to report a full smoking ban in homes (OR = 2.5, 95% CI = 1.359–4.464; OR = 1.2, 95% CI = 1.121–1.225, respectively) (Table 3).
Discussion The study found that 14.2% of the survey respondents in Guangdong, China, had adopted a full smoking ban at home in 2010. This rate was lower than the 26% reported in Shanghai urban residents (Ji et al., 2009), and the 66% reported among
Downloaded from http://ntr.oxfordjournals.org/ at University of North Dakota on August 2, 2015
All analyses were performed using SPSS 18.0. We used p = .05 as the level for statistical significance. Standard descriptive statistics were used to summarize data collected for demographics, smoking-related characteristics, and household smoking policies. The chi-square test was used to examine the relationship between household smoking ban and other variables regarding demographic information and smoking status. Independent-sample t tests were used to compare continuous variables. We used multiple logistic regression analyses to estimate predictors of smoking policies at home. The dependent variable was whether or not a full smoking ban was adopted at home, whereas the independent variables included educational level, household income, gender, smoking status, occupation, knowledge of SHS risk, and attitudes toward smoking bans in public places. Variables were included in the model if they were significant at a p value of .05 or less in the bivariate analysis. The adjusted odds ratios (ORs) with confidence intervals were reported.
current smokers reported a complete household smoking ban at home compared with 17.9% of nonsmokers doing so. Household smoking policy was related to people’s knowledge regarding the consequences of SHS exposure and attitudes toward smoking restrictions in public places. In all, 20.3% of respondents who agreed that SHS exposure was harmful reported a full smoking ban at home, whereas only 6.8% of those who disagreed had a full smoking ban at home. Conversely, respondents who reported the presence of a full ban at home supported implementing smoking restrictions in public places more than those who reported a partial ban or no ban at home.
Nicotine & Tobacco Research Table 1. Distributions of Different Smoking Policy in Household by Demographics and Smoking-Related Characteristics Smoking policies in household Characteristics
Partial ban (%)
No ban (%)
Total
301 (14.2)
498 (23.6)
1,315 (62.2)
2,114
134 (12.1) 167 (16.6)
246 (22.2) 252 (25)
728 (65.7) 587 (58.3)
1,108 (52.4) 1,006 (47.6)
216 (21) 85 (7.8)
304 (29.5) 194 (17.9)
509 (49.5) 806 (74.3)
1,029 (48.7) 1,085 (51.3)
30 (5.6) 55 (7.8) 95 (18.6) 121 (33.7)
72 (13.4) 158 (22.4) 131 (25.5) 137 (38.2)
434 (81) 492 (69.8) 288 (56) 101 (28.1)
536 (25.4) 705 (33.3) 514 (24.3) 359 (17)
75 (7.7) 39 (14.9) 41 (28.1) 74 (30.1) 72 (14.9)
181 (18.5) 68 (26.1) 53 (36.3) 68 (27.5) 128 (26.5)
721 (73.8) 154 (59) 52 (35.6) 105 (42.5) 283 (58.6)
977 (46.2) 261 (12.3) 146 (6.9) 247 (11.7) 483 (22.8)
54 (6.2) 200 (19.9)
114 (13.2) 300 (29.9)
698 (80.6) 504 (50.2)
866 (41) 1,004 (47.5)
47 (19.3)
84 (34.4)
113 (46.3)
244 (11.5)
13 (2.6) 288 (17.9)
100 (19.8) 398 (24.7)
392 (77.6) 923 (57.4)
505 (23.9) 1,608 (76.1)
237 (20.3) 64 (6.8) 38.69 22.95
350 (30) 148 (15.7) 39.28 21.32
579 (49.7) 733 (77.6) 40.93 18.39
1,166 (55.2) 945 (44.8) 40.2
Note. SHS = secondhand smoke. Conversion rate is USD1.00 = RMB 6.59, www.oanda.com, January 1, 2011. aContinuous variables. *p ≤ .05. **p ≤ .001.
Table 2. Number and Percentage of Respondents Reporting Exposure to Household SHS by Household Smoking Policy Household smoking policy Variables Home exposure to SHS Yes No Total (n = 1,650) Home exposure to SHS (smokers) Yes No Total (n = 474) Home exposure to SHS (nonsmokers) Yes No Total (n = 1,176)
Partial ban (%)
No ban (%)
268 (55.1) 218 (44.9) 486
893 (76.7) 271 (23.3) 1,164
79 (79) 21 (21) 100
347 (92.8) 27 (7.2) 374
189 (49) 197 (51) 386
546 (69.1) 244 (30.9) 790
p value
