The neighbourhood they live in—Does it matter to women's smoking habits during pregnancy?
Introduction
Smoking during pregnancy has a clear social gradient and is interpreted as the most important mediator of the association between low socioeconomic status (SES) and negative infancy outcomes reported by several researchers (Kramer et al., 2000; Spencer and Logan, 2004). The connection between smoking and low SES can, however, also be considered from a neighbourhood perspective. Several studies report an association between socioeconomic factors in the neighbourhood and smoking habits in various populations. Ohlander et al. (2006) showed an association between unemployment and men's and women's smoking habits. Chuang et al. (2005) and Duncan et al. (1999) similarly found that higher neighbourhood SES was associated with lower smoking prevalence. Shohaimi et al. (2006) used a socioeconomic index of residential segregation to predict smoking habits among men and women. An earlier review identified five studies investigating neighbourhood effects on smoking habits. In those studies, the relative risk of smoking in deprived neighbourhoods varied from 1.2 to 1.7 (Pickett and Pearl, 2001). We found only two American studies on smoking during pregnancy from a neighbourhood perspective. In a study by Finch et al. (2001), neighbourhood public assistance was significantly associated with tobacco prevalence in pregnant women. Pickett et al. (2002) investigated the risk of smoking during pregnancy based on individual and local area factors in a cohort of 878 women in California. They were able to show that living in a predominantly working-class environment significantly increased the risk of smoking during pregnancy.
There is a pattern of segregation also in Swedish neighbourhoods which has been comprehensively described by geographers such as Andersson (2002) and Molina (1997). Everyday life in deprived neighbourhoods involves high criminality, poor housing quality, deficient public amenities like school, health care and social services. Swedish registers report that several indicators of individual vulnerability, such as unemployment, low income, receipt of social allowance, poor school achievement, bad health and exposure to violence, are closely linked to the income structure of the neighbourhood (National Board on Health and Welfare, 1997, National Board on Health and Welfare, 2001b). As an example, in 1995, the unemployment rate in Swedish urban areas was three times higher in poor compared to rich neighbourhoods. Also, during recent decades there has been a substantial immigration to Sweden. Today more than 10% of the Swedish population is born in another country, 7% of whom are born outside a Nordic country (Magnusson, 2001). Many immigrants live in segregated neighbourhoods with immigrants from different nations and with different cultural patterns (Andersson, 2002; Molina, 1997). Little is known about how living in ethnically segregated neighbourhoods affects health behaviours like smoking during pregnancy. In addition, a recent report clarified that the ethnic segregation has increased and coincides progressively more with economic segregation (National Board on Health and Welfare, 2006). Living in such neighbourhoods can increase vulnerability to stress and thus increase the risk of smoking during pregnancy. However, smoking behaviour can also be considered from a social learning perspective (Bandura, 1977). In a review Conrad et al. (1992) concluded that peer influence is the most important determinant of adolescent smoking. Curry et al. (1997) showed that women whose partners are smokers are significantly more likely to smoke during pregnancy. Yet another study shows that being in the company of other smokers accounts for a substantial proportion of variance in postpartum relapse rates (Shiffman, 1982). As outlined by Logan and Spencer (1996), it can be hypothesized that women's smoking habits during pregnancy are similarly determined by social and environmental circumstances in the neighbourhood. Thus, the neighbourhood SES can be assumed to represent neighbourhood social factors and can therefore be viewed as a proxy for features of neighbourhoods that are potentially relevant for health (Diez Roux, 2004).
This study aims to determine the effect of neighbourhood economic and ethnic context on smoking habits during pregnancy over and above the effect of the woman's SES. In order to take the hierarchical nature of the data into account, i.e., individuals being nested within urban areas, a multilevel approach was employed.
Section snippets
Design and variables
This study was based on register information regarding pregnant women in the three major cities in Sweden, Stockholm, Göteborg and Malmö during the years 1992–2001 linked through each individual's unique civic registration number (National Board on Health and Welfare, 2001a; Statistics Sweden, 1998, Statistics Sweden, 2001, Statistics Sweden, 2004). Data on pregnancies came from the Swedish Medical Birth Register held by the National Board on Health and Welfare in Sweden (National Board on
Results
In all, 86.1% of women (n=109,450) did not smoke during pregnancy, while 13.8% (n=17,624) were smokers. Of the smokers, 12,103 women smoked less than 10 cigarettes a day, and 5521 smoked 10 cigarettes or more. In Table 2, we present descriptive statistics for the sample. The smokers were younger compared with non-smokers (χ2=2321.65; 4d.f.; P-value<0.001). Swedish-born women smoked to a greater extent than women born outside Sweden (χ2=149.25; 1d.f.; P-value<0.001). The non-smoking women were
Discussion
In this study, we show that smoking prevalence among pregnant women varies significantly between neighbourhoods. This variation can mainly be related to individual predictors. However, 6.4% of the variation is at the neighbourhood level. Furthermore, the odds of smoking during pregnancy are doubled in poorer areas. This finding is supported by similar findings by Pickett et al. (2002) who reported that smoking during pregnancy was more prevalent in working class contexts. The effect of
Acknowledgements
This study was made possible through a grant from the Swedish Vårdal Foundation and from MidSweden University. We are also indebted to Danuta Biterman at the Centre for Epidemiology, Swedish National Board of Health and Welfare, for creating the neighbourhood variables.
References (35)
- et al.
Assessment of community-level influences on individuals’ attitudes about cigarette smoking, alcohol use, and consumption of dietary fat
American Journal of Preventive Medicine
(1993) - et al.
Reasons for quitting: intrinsic and extrinsic motivation for smoking cessation in a population-based sample of smokers
Addictive Behaviours
(1997) Estimating neighborhood health effects: the challenges of causal inference in a complex world
Social Science & Medicine
(2004)- et al.
Smoking and deprivation: are there neighbourhood effects?
Social Science & Medicine
(1999) - et al.
Substance use during pregnancy in the state of California, USA
Social Science & Medicine
(2001) - et al.
The working-class context of pregnancy smoking
Health Place
(2002) - et al.
Maternal and child health and neighborhood context: the selection and construction of area-level variables
Health Place
(2006) - et al.
Decomposing the total variation in a nested random effects model of neighborhood, household, and individual components when the dependent variable is dichotomous: implications for adolescent marijuana use
Drug and Alcohol Dependence
(2005) Residential segregation and ethnic hierarchies
Social Learning Theory
(1977)
Neighbourhoods
Effects of neighbourhood socioeconomic status and convenience store concentration on individual level smoking
Journal of Epidemiology and Community Health
Why children start smoking cigarettes: predictors of onset
British Journal of Addiction
Multilevel Modelling of Health Statistics
Socio-economic disparities in pregnancy outcome: why do the poor fare so poorly?
Paediatric and Perinatal Epidemiology
Smoking and other health related behaviour in the social and environmental context
Archives of Disease in Childhood
Immigration and segregation
Cited by (20)
Smoking during pregnancy was up to 70% more common in the most deprived municipalities - A multilevel analysis of all singleton births during 2005-2010 in Finland
2014, Preventive MedicineCitation Excerpt :Smoking during pregnancy is still quite prevalent; an estimated 5–19% of pregnant women in European Union member states smoked in 2010 (EURO-PERISTAT Project, 2010). Maternal low education and socioeconomic status (SES) are significant correlates of smoking during pregnancy (Cnattingius et al., 1992; Räisänen et al., 2013; Sellstrom et al., 2008). In addition, residential area deprivation has been shown to contribute to poor health, health behavior (Pickett and Pearl, 2001) and adverse perinatal outcomes (Blumenshine et al., 2010; Pickett and Pearl, 2001).
Racial segregation and maternal smoking during pregnancy: Amultilevel analysis using the racial segregation interaction index
2014, Social Science and MedicineCitation Excerpt :In addition, socioeconomically disadvantaged women-women with a low household income (Hunt & Whitman, 2011; Martin et al., 2008; Wakschlag et al., 2003), low educational attainment (Kahn et al., 2002; Martin et al., 2008; Orr et al., 2005; Wakschlag et al., 2003), or living in poverty (Yu, Park, & Schwalberg, 2002) - are also more prone to smoke during pregnancy (Pickett et al., 2008). Although there has been a scholarly push to move beyond the individual level and to consider contextual factors in understanding maternal smoking during pregnancy (Pickett, Wakschlag, Rathouz, Leventhal, & Abrams, 2002; Sellstrom, Arnoldsson, Bremberg, & Hjern, 2008), studies investigating the contextual level influences on individual maternal smoking during pregnancy behavior have been scant up to date (Gage, Everett, & Bullock, 2007). In addition, those few previous studies investigating the contextual level influences on individual maternal smoking during pregnancy behavior have focused on only one or two racial and ethnic groups (e.g., non-Hispanic white and non-Hispanic black) (Bell, Zimmerman, Mayer, Almgren, & Huebner, 2007; Pickett et al., 2002) and limited geographic locations (e.g., a specific state or metropolitan areas) (Pickett et al., 2002).
Contribution of local area deprivation to cultural-linguistic inequalities in foetal growth restriction: Trends over time in a canadian metropolitan centre
2013, Health and PlaceCitation Excerpt :Indeed, in this study, inequalities in SGA birth related to material deprivation decreased over time for Francophones, yet Anglophone SGA birth prevalence increased in deprived as well as, unexpectedly, advantaged neighbourhoods (preferentially over intermediate quintiles). As smoking is a strong risk factor for SGA birth (Salihu and Wilson, 2007), and neighbourhoods have potential to influence smoking behaviour (Sellstrom et al., 2008; Vinikoor-Imler et al., 2011), differentials in the trajectory of smoking behaviour between Anglophones and Francophones may have shaped our results, reflecting a greater tendency towards smoking among Anglophones in materially advantaged areas. This hypothesis is difficult to confirm, however, as existing data on smoking are not disaggregated by deprivation level, though there may have been a greater decline in smoking among Francophones than Anglophones in recent decades (Institut national de santé publique du Québec, 2010).
Persisting inequalities in birth outcomes related to neighbourhood deprivation
2020, Journal of Epidemiology and Community HealthMultilevel Modelling for Public Health and Health Services Research: Health in Context
2020, Multilevel Modelling for Public Health and Health Services Research: Health in ContextIndividual and regional characteristics associated with maternal smoking during pregnancy in Japan: Healthy parents and children 21
2020, International Journal of Environmental Research and Public Health