• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nictobLink to Publisher's site
Nicotine Tob Res. Jul 2009; 11(7): 779–784.
Published online May 6, 2009. doi:  10.1093/ntr/ntp060
PMCID: PMC2699925

Employment and social “determinants” of smoking in urbanizing China: A representative survey



This study explored social determinants of smoking among a sample of male Chinese adults in Changqiao, a community representing the transition from traditional to a “mobile” urban culture in China. New commercial systems have introduced high profits but also layoffs in the absence of government security systems.


In-person interviews were conducted by trained interviewers with 123 male participants selected at random. Bivariate and multivariate analyses were computed based on the Behavioral Ecological Model (BEM). About 61% of male participants were ever-smokers and 48% were current smokers.


Current smoking was associated with involuntary unemployment (odds ratio [OR] = 6.52), the absence of home smoking restrictions (OR = 0.34), and social reinforcement such as friends’ smoking (OR = 4.02) and receiving smoking-related gifts (OR = 6.39).


Findings support the BEM. It is especially important to verify the relationship between unemployment and smoking, given the recent rise in involuntary job loss due to the transitional economy in China.


As the most populous country and growing economic power, China plays a substantial role in global health (Han, Chen, Evans, & Horton, 2008). Several studies have proven that tobacco smoking is an important threat to public health in China (Gu et al., 2009; Liu et al., 1998; Niu et al., 1998). In 2002, the national prevalence of smoking was 66% for males and 3% for females (G. Yang, Ma, Liu, & Zhou, 2005). Few cessation programs are available in China (de Vries, 2007). Most smoking correlates identified in China have been demographic (Pan, 2004) and cannot be modified for tobacco control. Thus, behavioral epidemiology is needed to inform tobacco control in China.

Recent studies have called for an ecological approach to tobacco control (Unger et al., 2003). In China, many smokers are “social smokers,” and their smoking behavior is influenced heavily by socioenvironmental factors (Pan, 2004; Pan & Hu, 2008). Social contingencies (specific situations in which reinforcement for behavior occurs) of smoking in China are dynamic due to the economic and social transitions China has experienced in the past few decades. With the evolution from a government planned to a marketplace economy, involuntary job loss (xia gang) has increased (Duckett & Hussain, 2008). National statistics have shown that between 1995 and 2002, 45 million Chinese were laid off (Giles, Park, & Zhang, 2005). Meanwhile, western practices such as cigarette smoking have increased with marketing and may be supported by social norms of collectivism (Bian, 1994). These changing conditions could be synergistic for smoking behavior in China.

The Behavioral Ecological Model (BEM) is an extension of respondent and operant principles of behavior with an emphasis on contingencies of reinforcement from proximal environmental sources, such as families, to distal environmental sources, such as international policies (Hovell, Wahlgren, & Adams, in press; Hovell, Wahlgren, & Gehrman, 2002). The BEM emphasizes cultural and social factors as important “determinants” of behavior. The present study used the BEM to determine probable correlates (determinants) of smoking in a representative sample of Chinese adults. The BEM has been applied in correlation and intervention studies in different countries (Hovell et al., in press). Studies involved acculturation (Hofstetter et al., 2004; Landrine & Klonoff, 2004), secondhand smoke exposure (Hughes, Corcos, Hofstetter, Hovell, & Irvin, 2007; Hughes et al., 2008), home smoking bans (Hughes et al., 2009; Martinez-Donate et al., 2008), substance use (Bousman et al., 2005), dietary behaviors (Song et al., 2004) and physical activity (Adams et al., 2006).

The present study is the first application of the BEM in China and the first to explore the role of unemployment. Consistent with the BEM and Chinese cultural values, we hypothesized that smoking would be more likely to occur among individuals receiving gifts related to smoking (implied antecedent and reinforcer), among those with friends who smoke (models of behavior), among those reporting the absence of reprimands for males smoking (social reinforcement), and among those reporting the absence of smoking restrictions in the home (behavioral rules and social reinforcement). We also hypothesized that stress due to job loss would increase smoking in this subgroup.


Study setting and sampling

Changqiao (population: 103,000) is located on the outskirts of urban Shanghai. Cross-sectional data were obtained through face-to-face interviews by researchers at Fudan University, using probability sampling techniques (Zheng et al., 2008). The Changqiao Residents’ Committee (Mok, 1988) facilitated the recruitment process by contacting randomly selected residents and promoting voluntary participation with a small monetary incentive, resulting in few refusals. From April to June 2006, 243 participants from the Changqiao District completed interviews. After excluding women due to low rates of smoking (e.g., zero), 123 males were included in the analysis. The San Diego State University Institutional Review Board approved the study protocol.


Individuals who smoked at least 100 cigarettes in their lifetime and currently smoked some days or every day were classified as current smokers (U.S. Department of Health and Human Services [USDHHS], 1996). Demographic variables included age, marital status, employment status, educational level, and family income. Participants reported their endorsement of the harm of smoking, and smoking restrictions in the household. They also reported the number of residents in the household and each resident's age, gender, relationship to the participant, smoking status, cessation history, tolerance of others’ smoking, and attitude toward smoking bans in public places.

Three items were used to measure the influence of social contingencies on smoking. Gift receiving was measured by the response to the question, “Have you ever received gifts that are related to smoking, such as cigarettes, lighters, and ashtrays?” The proportion of friends who smoke was used as a marker for role models for smoking and probable encouragement to smoke. We used the question, “How many of your friends smoke?” (responses coded as most and some/few due to skewness). Finally, we expected reprimands to reduce smoking level. For this measure, we used the question, “How likely will a male be reprimanded for smoking in the following places in the community?” Nine places were included: homes, workplaces, restaurants, bars and clubs, schools, public transportation, hospitals, shopping malls, and government buildings. Answers were coded on a scale from 1 to 3 (representing likely, sometimes, and never, respectively). A scale was formed by averaging scores for the nine items, with up to three missing (M = 1.804, SD 0.476, Cronbach's α = .85).

Data analyses

Data were analyzed using SPSS version 14.0. The bivariate associations between current smoking status and each theoretical correlate were estimated using chi-square tests. Variables yielding p values of less than .20 were included in sequential logistic regression analyses. To assess predictors of different proximities based on the BEM, variables were entered by blocks: sociodemographic variables, home smoking restrictions, and social contingencies. We tested the following interactions one at a time: age and friends’ smoking, age and home smoking restrictions, age and reprimands, home smoking restrictions and friends’ smoking, and home smoking restrictions and receiving gifts.


Demographic characteristics of the overall study sample (n = 243), including age, gender ratio, educational level, and marital status, resembled those of another study in the same community (Zheng et al., 2008).

Among the male participants, 61% were ever-smokers, including 48% current smokers and 13% former smokers. On average, current smokers consumed 15 cigarettes/day. Half of the current smokers (51%) had tried quitting but failed, with an average of four attempts over the lifetime. Half of the participants had no home smoking restrictions. In contrast, more than 95% reported endorsement of the negative health consequences of smoking.

Table 1 shows bivariate associations between current smoking and each independent variable. With the exception of marital status (p > .20), most factors were related to current smoking. Table 2 presents sequential multiple logistic regression models. In the final model, the odds of smoking were significantly higher among the involuntarily unemployed males, males whose friends were mostly smokers, and males who had received smoking-related gifts. The odds of smoking were marginally higher among males who had no home smoking restrictions. Age was significant in Models 1 and 2 but not in Model 3, suggesting that more proximal social factors related to age may be the mechanisms by which age is often predictive. Reprimands were not significant in any model. No first-order interactions reached significance. This final model had an adequate goodness of fit (χ2 = 5.74, df = 8, p = .677; Hosmer & Lemeshow, 1980).

Table 1.
Bivariate associations between selected characteristics and current smoking status of male participants in Changqiao, Shanghai, China, 2006 (N = 123)
Table 2.
Sequential multiple logistic regression models of current smoking in Changqiao, Shanghai, China, in 2006 (N = 117)


Findings from Changqiao have special implications for China. As an urbanizing community, Changqiao is under the influences of both traditional Chinese and modern western culture (Zheng et al., 2008), representing many areas in China that are in transition to urbanization and westernization.

In Changqiao, males were more likely to smoke if most of their friends were smokers. More than 50% reported that they usually smoked in gatherings of friends or smoked to start conversations in social settings (data not shown). Fewer than 10% of the men lived with another smoker, suggesting that the main pro-smoking influence came from outside the home. Policies such as bans in public settings should be implemented to limit smoking, as well as to protect nonsmokers from toxic exposure. Based on the BEM, rules and policies deliver immediate negative consequences for smoking and could reduce smoking behavior.

According to the BEM, a smoking-related gift may serve as a tangible reinforcer or a marker for social contingencies for smoking. However, gift receiving has never been studied as a predictor of smoking in China. Our findings suggested that males who received smoking-related gifts were more likely to be current smokers. In China, cigarettes and related items are frequently offered as impressive gifts and “connection builders,” regardless of the recipient's smoking status (Pan, 2004), especially in the business subculture (Cheng, 1999). Marketing campaigns that discourage the use of tobacco-related products as gifts might reduce the prevalence of such gift giving and move the culture closer to one that curtails tobacco use.

The present study found that the absence of home smoking restrictions was related to smoking. This finding is consistent with previous research in the United States (Farkas, Gilpin, Distefan, & Pierce, 1999; Gilpin, White, Farkas, & Pierce, 1999). Logistic regression findings (data not shown) suggested that home smoking restrictions were associated with having fewer friends who smoke, the spouse's low tolerance of secondhand smoke, and support for public smoking bans. In traditional Chinese culture, the family is highly valued (T. Yang, Fisher, Li, & Danaher, 2006). Targeting married women in China to reduce their husbands’ smoking by encouraging them to adopt home smoking restrictions offers promise for tobacco control. Such restrictions would reduce risks to family members otherwise exposed to passive smoke.

The present study found a high smoking prevalence among involuntarily unemployed males. This finding is consistent with findings in the United States (Falba, Teng, Sindelar, & Gallo, 2005) but has not been reported in China. With modernization of the Chinese economy, employment practices are in transition (Duckett & Hussain, 2008; Tsui, 2002). The unemployed population no longer receives government-provided income, social, or health benefits (Gao, 2008). These changes might lead to decreased quality of life and poorer health. The Chinese Urban Labor Survey estimated an unemployment rate of 14% among urban permanent residents (Giles et al., 2005). Unfortunately, few health-related studies have investigated unemployment in this new economy. We suggest that more studies of newly unemployed males in China be conducted, especially regarding depression, anxiety, and other reactions to job loss as mediators of smoking.

Several studies have shown that middle-aged Chinese males smoke more than others (Xu et al., 2007; Zheng et al., 2008); the present study found the same. However, after adjusting for other variables in Model 3, we found that the association between age and smoking was no longer significant. In this analysis, males aged 40–59 years were more likely to have a large proportion of friends who smoke (data not shown), implying the existence of a “middle-aged smoking culture” through which social contingencies might account for more smoking. In the present study, males aged 40–59 years also were more likely to have been laid off (data not shown), a finding consistent with national employment statistics (Giles et al., 2005).

Despite the high prevalence of smoking, the majority of participants endorsed the health hazards of smoking and nonsmokers’ rights to be free of exposure. Similar findings were documented by Gong et al. (1995) in the Minhang District, Shanghai. A possible explanation for this finding, based on the BEM, is that negative health consequences of smoking are delayed in time, compared with more immediate physiological and social reinforcement, which influence smoking directly.

Several demographic and individual variables have been associated with smoking in China (Chen et al., 2004; Pan, 2004; Xu et al., 2007). The present study adds to the literature by demonstrating relationships between smoking and relevant social factors, some of which may be mechanisms by which demographic factors predict smoking. Results suggest replication using a longitudinal research design among a larger sample including both males and females. To tackle the smoking epidemic in China, future interventions should target different levels of the social environment in China, such as families, friends, business, cultural norms, and policies.

Finally, the low smoking rates among women in China demonstrate that cultural factors can promote smoking and at the same time suppress smoking when applied differentially to subgroups of the population. This is a critical observation that justifies more research from which better understanding of culture and cultural processes can be forthcoming. Ultimately, such new knowledge may contribute to technologies that can alter cultures to increase the likelihood of health-promoting practices and reduce the likelihood of risk practices.


National Heart, Lung, and Blood Institute (grant HL066307); Fudan University; the Center for Behavioral Epidemiology and Community Health.

Declaration of Interests

None declared.

Supplementary Material

[Article Summary]


The authors thank Jodi Kudas and Norma Kelley for their coordinative and technical support and Tyler and Besa Smith for help editing this article.


  • Adams MC, Hovell MF, Irvin V, Sallis JF, Coleman KJ, Liles S. Promotion stair use by modeling: An experimental application of the Behavioral Ecological Model. American Journal of Health Promotion. 2006;21:101–109. [PubMed]
  • Bian Y. Guanxi and the allocation of urban jobs in China. China Quarterly. 1994;140:971–1000.
  • Bousman CA, Blumberg EL, Shilington AM, Hovell MF, Ji M, Lehman S, et al. Predictors of substance use among homeless youth in San Diego. Addictive Behavior. 2005;30:1100–1110. [PubMed]
  • Cheng T. Teenage smoking in China. Journal of Adolescence. 1999;22:607–620. [PubMed]
  • Chen X, Li X, Stanton B, Fang X, Lin D, Cole M, et al. Cigarette smoking among rural-to-urban migrants in Beijing, China. Preventive Medicine. 2004;39:666–673. [PubMed]
  • de Vries H. Comment on “Modifiable family and school environmental factors associated with smoking status among adolescents in Guangzhou, China. Preventive Medicine. 2007;45:119–120. [PubMed]
  • Duckett J, Hussain A. Tackling unemployment in China: State capacity and governance issues. Pacific Review. 2008;21:211–229.
  • Falba T, Teng H, Sindelar J, Gallo W. The effect of involuntary job loss on smoking intensity and relapse. Addiction. 2005;100:1330–1339. [PMC free article] [PubMed]
  • Farkas A, Gilpin E, Distefan J, Pierce J. The effects of household and workplace smoking restrictions on quitting behaviors. Tobacco Control. 1999;8:261–265. [PMC free article] [PubMed]
  • Gao Q. The Chinese social benefit system in transition: Reform and impacts on income inequality. Annals of the New York Academy of Sciences. 2008;1136:342–347. [PubMed]
  • Giles J, Park A, Zhang J. What is China's true unemployment rate? China Economic Review. 2005;16:149–170.
  • Gilpin E, White M, Farkas A, Pierce J. Home smoking restrictions: Which smokers have them and how they are associated with smoking behavior. Nicotine & Tobacco Research. 1999;1:153–162. [PubMed]
  • Gong Y, Koplan J, Feng W, Chen C, Zheng P, Harris J. Cigarette smoking in China: Prevalence, characteristics, and attitudes in Minhang District. Journal of the American Medical Association. 1995;274:1232–1234. [PubMed]
  • Gu D, Kelly TN, Wu X, Chen J, Samet JM, Huang J, et al. Mortality attributable to smoking in China. New England Journal of Medicine. 2009;360:150–159. [PubMed]
  • Han Q, Chen L, Evans T, Horton R. China and global health. Lancet. 2008;372:1439–1441. [PubMed]
  • Hofstetter CR, Hovell MF, Lee J, Zakarian J, Park H, Paik H, et al. Tobacco use and acculturation among Californians of Korean descent: A behavioral epidemiological analysis. Nicotine & Tobacco Research. 2004;6:481–489. [PubMed]
  • Hosmer DW, Lemeshow S. Goodness of fit tests for the multiple logistic regression model. Communications in Statistics—Theory and Methods. 1980;9:1043–1069.
  • Hovell MF, Wahlgren DR, Adams M. The logical and empirical basis for the behavioral ecological model. In: DiClemente RJ, Crosby RA, Kegler M, editors. Emerging theories and models in health promotion research and practice: Strategies for enhancing public health. 2nd ed. San Francisco: Jossey-Bass; in press.
  • Hovell MF, Wahlgren DR, Gehrman CA. The behavioral-ecological model: Integrating public health and behavioral science. In: DiClemente RJ, Crosby RA, Kegler M, editors. Emerging theories in health promotion practice and research: strategies for improving public health. San Francisco: Jossey-Bass; 2002. pp. 347–385.
  • Hughes SC, Corcos IA, Hofstetter CR, Hovell HF, Irvin VL. Environmental tobacco smoke exposure among Korean American nonsmokers in California. Nicotine & Tobacco Research. 2007;10:663–670. [PubMed]
  • Hughes SC, Corcos IA, Hofstetter CR, Hovell MF, Irvin VL, Park HR, et al. Home smoking restrictions among Koreans in Seoul. Asia Pacific Journal of Public Health. 2009;21:63–70. [PubMed]
  • Hughes SC, Corcos IA, Hofstetter CR, Hovell MF, Seo DC, Irvin VL, et al. Secondhand smoke exposure among non-smoking adults in Seoul, Korea. Asia Pacific Journal of Cancer Prevention. 2008;9:247–252. [PubMed]
  • Landrine H, Klonoff EA. Cultural and ethnic-minority health behavior: An operant theory of acculturation. Journal of Behavioral Medicine. 2004;27:527–553. [PubMed]
  • Liu B, Peto R, Chen Z, Boreham J, Wu Y, Li J. Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths. British Medical Journal. 1998;317:1411–1422. [PMC free article] [PubMed]
  • Martinez-Donate AP, Hovell MF, Hofstetter CR, Gonzalez-Perez GJ, Kotay A, Adams MA. Crossing borders: The impact of the California Tobacco Control Program on both sides of the US-Mexico Border. American Journal of Public Health. 2008;98:258–267. [PMC free article] [PubMed]
  • Mok BH. Grassroots organising in China: The residents’ committee as a linking mechanism between the bureaucracy and the community. Community Development Journal. 1988;23:164–169.
  • Niu S, Yang G, Chen Z, Wang J, Wang G, He X, et al. Emerging tobacco hazards in China: 2. Early mortality results from a prospective study. British Medical Journal. 1998;317:1423–1424. [PMC free article] [PubMed]
  • Pan Z. Socioeconomic predictors of smoking and smoking frequency in urban China: Evidence of smoking as a social function. Health Promotion International. 2004;19:309–315. [PubMed]
  • Pan Z, Hu D. Hierarchical linear modeling of smoking prevalence and frequency in China between 1991 and 2004. Health Policy and Planning. 2008;23:118–124. [PubMed]
  • Song YJ, Hofstetter CR, Hovell MF, Paik HY, Park HR, Lee J, et al. Acculturation and health risk behaviors among Californians of Korean descent. Preventive Medicine. 2004;39:147–156. [PubMed]
  • Tsui M. Managing transition: Unemployment and job hunting in urban China. Pacific Affairs. 2002;75:515–534.
  • Unger JB, Cruz T, Shakib S, Mock J, Shields A, Baezconde-Garbanati L, et al. Exploring the cultural context of tobacco use: A transdisciplinary framework. Nicotine & Tobacco Research. 2003;5(Suppl. 1):S101–S117. [PubMed]
  • U.S. Department of Health and Human Services. National Center for Health Statistics. NHANES 3 reference manuals and reports (CD-ROM) Hyattsville, MD: Center for Disease Control and Prevention; 1996.
  • Xu F, Yin X, Zhang M, Ware S, Leslie E, Owen N. Cigarette smoking is negatively associated with family average income among urban and rural men in regional Mainland China. International Journal of Mental Health Addition. 2007;5:17–23.
  • Yang T, Fisher K, Li F, Danaher B. Attitudes to smoking cessation and triggers to relapse among Chinese male smokers. BMC Public Health. 2006;6:65. doi: 10.1186/1471-2458-6-65. [PMC free article] [PubMed]
  • Yang G, Ma J, Liu N, Zhou L. Smoking and passive smoking in Chinese. Chinese Journal of Epidemiology. 2005;26:77–83. [PubMed]
  • Zheng P, Fu Y, Lu Y, Ji M, Hovell M, Fu H. Community smoking behavior in Changqiao, Shanghai. Asian-Pacific Journal of Public Health. 2008;20:94–101. [PubMed]

Articles from Nicotine & Tobacco Research are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...