Elsevier

Social Science & Medicine

Volume 87, June 2013, Pages 132-137
Social Science & Medicine

Short report
Smoking initiation, continuation and prevalence in deprived urban areas compared to non-deprived urban areas in The Netherlands

https://doi.org/10.1016/j.socscimed.2013.03.038Get rights and content

Highlights

  • Indicators of smoking among Dutch adults were more common in deprived urban areas.

  • Area-level inequalities were larger for smoking initiation than for cessation.

  • Most of these inequalities were explained by individual-level socioeconomic factors.

  • An independent area effect was found for smoking initiation in younger generations.

  • Area-level efforts to reduce smoking inequalities may focus on smoking initiation.

Abstract

Previous studies have shown that smoking prevalence is higher in deprived areas than in affluent areas. We aimed to determine whether smoking initiation or continuation contributes most to inequalities in current smoking, and in which population subgroups these area differences were largest. Therefore, we assessed the relationship between area deprivation and current smoking, initiation and continuation in urban areas, in subgroups defined by gender, generation and educational level. Cross-sectional data of 20,603 Dutch adults (18 years and over) living in 963 urban areas in The Netherlands were obtained from the annual national health survey (2003–2009). Three interrelated smoking outcomes were used: current smoking (smokers/total population), initiation (ever-smokers/total population) and continuation (smokers/ever-smokers). Area deprivation was dichotomised; deprived urban areas (as defined by the Dutch government) and non-deprived urban areas (reference group) were distinguished. Multilevel logistic regression models controlled for individual characteristics (including education and income) and tested for interaction with gender, generation and education. After controlling for individual characteristics, odds for smoking were not significantly higher in deprived areas (current smoking: OR = 1.04 [0.92–1.18], initiation: OR = 1.05 [0.93–1.18], continuation: OR = 1.03 [0.88–1.19]). For smoking initiation, significant differences between deprived areas and other areas remained in younger generations (OR = 1.19 [1.02–1.38]) and higher educated (OR = 1.23 [1.04–1.45]) respondents. For continuation and current smoking, after controlling for individual characteristics, no associations were found in any subgroups. In conclusion, area deprivation appears to be independently related to smoking initiation in, respectively, higher educated and younger generations. These results suggest that initiatives to reduce area-level inequalities in smoking should focus on preventing smoking initiation in deprived areas.

Introduction

Even though the prevalence of smoking has been declining for several decades (Giskes et al., 2005), smoking remains the most important source of mortality and disease burden worldwide (Rehm, Taylor, & Room, 2006). A major part of the difference in mortality between lower and higher socio-economic status (SES) groups is due to differences in smoking (Jha et al., 2006). In 2010, 34% of Dutch males with a lower educational level were smokers compared to 22% of males with a higher educational level. Females showed similar differences: 33% and 18% respectively (Zeegers, Blokstra, & Zantinge, 2011).

Along with individual SES, socio-economic and physical features of the neighbourhood are also associated with tobacco use. Previous studies found residents of deprived areas to be more likely to smoke than people living in more affluent areas (Chuang, Cubbin, Ahn, & Winkleby, 2005; Diez Roux, Merkin, Hannan, Jacobs, & Kiefe, 2003; Duncan, Jones, & Moon, 1999; Ellaway & Macintyre, 2009; Giskes, van Lenthe, Turrell, Brug, & Mackenbach, 2006; Kaestle & Wiles, 2010; Kleinschmidt, Hills, & Elliott, 1995; Ohlander, Vikstrom, Lindstrom, & Sundquist, 2006; Reijneveld, 2002; Ross, 2000; Shohaimi et al., 2003; Stead, MacAskill, MacKintosh, Reece, & Eadie, 2001; Tseng, Yeatts, Millikan, & Newman, 2001; van Lenthe & Mackenbach, 2006). Most studies used current smoking as the outcome variable. Results consistently showed that smokers are more likely to live in deprived areas (Diez Roux et al., 2003; Duncan et al., 1999; Ellaway & Macintyre, 2009; Kaestle & Wiles, 2010; Kleinschmidt et al., 1995; Reijneveld, 2002; Ross, 2000; Shohaimi et al., 2003; van Lenthe & Mackenbach, 2006). Most studies showed that effects of the area socio-economic environment on smoking could not be attributed solely to residents' individual characteristics like SES (Chaix, Guilbert, & Chauvin, 2004; Chuang et al., 2005; Diez Roux et al., 2003; Duncan et al., 1999; Ellaway & Macintyre, 2009; Giskes et al., 2006; Kleinschmidt et al., 1995; Reijneveld, 2002; Ross, 2000; Shohaimi et al., 2003; van Lenthe & Mackenbach, 2006).

To effectively support policymakers in tackling smoking in deprived areas, more information is needed about why smoking rates are higher in these areas than in other, non-deprived, areas. A higher smoking prevalence in deprived areas could be attributable to residents being more likely to start smoking and/or less likely to quit smoking. Making a distinction between smoking initiation and cessation will be relevant to determining the focus of policy and interventions. Furthermore, identifying subpopulations in which differences between deprived and non-deprived areas are particularly large might help to target groups in deprived areas.

Few studies have separated the concept of smoking initiation from smoking cessation. According to Shohaimi and colleagues (Shohaimi et al., 2003), residents of more deprived areas in the United Kingdom were less likely to report having never smoked, while Tseng et al. (Tseng et al., 2001) found no association between area-level socio-economic characteristics and ever-smoking. In the latter study, continuation of smoking among ever-smokers was more common in areas with a greater number of less-educated and unemployed residents. This result is in line with a study by Giskes and colleagues (Giskes et al., 2006), which showed that residents of more deprived areas in the Dutch city of Eindhoven were less likely to quit smoking.

Few studies have looked into possible differences between population subgroups. Previous studies have found larger area-level differences in smoking among individuals in low SES groups (Ohlander et al., 2006; van Lenthe & Mackenbach, 2006). Öhlander et al. found that the association between area deprivation and smoking was stronger in younger adults (Ohlander et al., 2006). Results were ambiguous when stratified by gender (Cohen, Sonderman, Mumma, Signorello, & Blot, 2011; Ellaway & Macintyre, 2009; Ohlander et al., 2006; Ross, 2000; Shohaimi et al., 2003). Öhlander et al. (Ohlander et al., 2006) and Cohen and colleagues (Cohen et al., 2011) suggested that differences are clearest among women, while Ross (Ross, 2000) found stronger associations in men.

The aim of this study was to determine the relationship between area deprivation and current smoking, smoking initiation (i.e. ever-smoking) and smoking continuation (i.e. not quitting) in urban areas. We also aimed to identify subpopulations (defined by age, gender and education) in which area-level differences in smoking are most profound. We used an extensive national survey that was large enough to distinguish between different smoking outcomes and between various subpopulations.

Section snippets

Study population

Variables were measured at individual and area levels. Individual level data of 20,603 Dutch adults age 18 and over were derived from the 2003–2009 national Integrated Survey on Living Conditions (POLS). The POLS is an extended interview questionnaire conducted annually among a population-based sample of the Dutch population. According the Medical Ethics Committee of the Academic Medical Centre Amsterdam, an official ethical approval is not required. The manuscript is entirely based on

Results

Residents of deprived areas differed from residents of non-deprived urban areas on several characteristics. Table 1 shows that deprived areas had a larger proportion of residents from non-Western ethic minorities and that educational level and household income were lower.

In Table 2, differences between deprived and non-deprived areas are also presented for smoking outcome variables. In the total population as well as in all the separate subgroups, there were more current smokers in deprived

Key findings

Smoking initiation, continuation and current smoking were more prevalent in deprived areas. Associations between area deprivation and current smoking and smoking continuation were accounted for by individual characteristics. Significant associations were found between area deprivation and smoking initiation for younger generations and more highly educated respondents. Therefore, the higher prevalence of smoking in deprived areas seems mainly attributable to smoking initiation rather than to

Acknowledgements

We gratefully acknowledge Statistics Netherlands for preparing micro-level data of the POLS survey, and providing access to these data.

References (39)

  • B. Chaix et al.

    A multilevel analysis of tobacco use and tobacco consumption levels in France: are there any combination risk groups?

    The European Journal of Public Health

    (2004)
  • T. Chandola et al.

    Socio-demographic predictors of quitting smoking: how important are household factors?

    Addiction

    (2004)
  • Y.C. Chuang et al.

    Effects of neighbourhood socioeconomic status and convenience store concentration on individual level smoking

    Journal of Epidemiology and Community Health

    (2005)
  • S.S. Cohen et al.

    Individual and neighborhood-level socioeconomic characteristics in relation to smoking prevalence among black and white adults in the southeastern United States: a cross-sectional study

    BMC Public Health

    (2011)
  • H. de Vries et al.

    Parents' and friends' smoking status as predictors of smoking onset: findings from six European countries

    Health Education Research

    (2003)
  • A.V. Diez Roux et al.

    Area characteristics, individual-level socioeconomic indicators, and smoking in young adults: the coronary artery disease risk development in young adults study

    American Journal of Epidemiology

    (2003)
  • A. Ellaway et al.

    Are perceived neighbourhood problems associated with the likelihood of smoking?

    Journal of Epidemiology and Community Health

    (2009)
  • B. Federico et al.

    Educational inequalities in smoking cessation trends in Italy, 1982–2002

    Tobacco Control

    (2009)
  • E. Fernandez et al.

    Social class, education, and smoking cessation: long-term follow-up of patients treated at a smoking cessation unit

    Nicotine & Tobacco Research

    (2006)
  • Cited by (9)

    • Educational inequalities in exposure to tobacco promotion at the point of sale among adolescents in four Dutch cities

      2022, Health and Place
      Citation Excerpt :

      Using a location-based indicator of socioeconomic status, instead of an individual-level indicator, could lead to larger differences, because location also determines population and tobacco outlet density. Moreover, Scotland may have larger neighbourhood-level differences than the Dutch cities in income levels, tobacco retail outlets density (Macdonald et al., 2018) and smoking prevalence (Laird et al., 2017; Kuipers et al., 1982). Sensitivity analyses using 20 and 30 m as minimum exposure distance showed weaker associations than our default 10 m.

    • Individual and contextual determinants of perceived peer smoking prevalence among adolescents in six European cities

      2016, Preventive Medicine
      Citation Excerpt :

      Perceived access and cigarette offers may differ between groups of education. Third, perceived smoking may be influenced by exposure to smoking outside the school context, for example in social/sports clubs, the neighbourhood, via internet (including social media), or in television and movies (Charlesworth and Glantz, 2005; Kuipers et al., 2013; Strasburger et al., 2010; Unger and Rohrbach, 2002). The exposure to such images may be relatively high among students of low education groups.

    View all citing articles on Scopus
    View full text