Addictive Behaviors 39 (2014) 1145–1151

Contents lists available at ScienceDirect

Addictive Behaviors

Smoking, nicotine dependence and nicotine intake by socio-economic status and marital status Marjaana Pennanen a,b,c, Ulla Broms a,b,c, Tellervo Korhonen a,b, Ari Haukkala d, Timo Partonen b, Annamari Tuulio-Henriksson b,e, Tiina Laatikainen b,f,g, Kristiina Patja b,h, Jaakko Kaprio a,b,i,⁎ a

Hjelt Institute, Department of Public Health, University of Helsinki, P.O. Box 41, Mannerheimintie 172, 00014 Helsinki, Finland National Institute for Health and Welfare, P.O. Box 30, Mannerheimintie 166, FI-00271 Helsinki, Finland Health and Social Services Department, Municipality of Askola, Askolantie 30, FI-07500 Askola, Finland d Department of Social Research, University of Helsinki, P.O. Box 54, Unioninkatu 37, 00014 Helsinki, Finland e Social Insurance Institution of Finland, P.O. Box 20, Nördenskiöldinkatu 12, 00232 Helsinki, Finland f Institute of Public Health and Clinical Nutrition, University of Eastern Finland, P.O. Box 1627, Yliopistonranta 1, FI-70211 Kuopio, Finland g Hospital District of North Karelia, P.O. Box 111, Tikkamäentie 16, FI-80101 Joensuu, Finland h Pro Medico, P.O. Box 49, Mäkelänkatu 2, 00501 Helsinki, Finland i Institute for Molecular Medicine Finland (FIMM), P.O. Box 20, Tukholmankatu 8, 00290 Helsinki, Finland b c

H I G H L I G H T S • Smokers with a low socio-economic status had a greater degree of nicotine dependence • Lower levels of education and occupation associated with higher levels of cotinine • Lower socio-economic status was associated with daily smoking

a r t i c l e

i n f o

Available online 13 March 2014 Keywords: Addiction Cessation Cotinine Nicotine Smoking Socio-economic status

a b s t r a c t Introduction: Low socio-economic status (SES) is strongly related to smoking, but studies examining the association of SES with nicotine dependence (ND) are scarce. The aim of this study was to examine the associations of SES and marital status with smoking, multiple measures of ND, and cotinine as a nicotine intake biomarker. Methods: The sample comprised 1746 ever smokers, sampled from the National FINRISK 2007 Study, who had completed a tobacco specific questionnaire in addition to the standard clinical examination. The Fagerström Test for Nicotine Dependence (FTND), the Heaviness of Smoking Index (HSI), the Nicotine Dependence Syndrome Scale (NDSS), and the Hooked On Nicotine Checklist (HONC) were assessed, while plasma cotinine was measured as a biomarker of nicotine exposure in daily smokers. Univariate and multivariate associations were assessed by linear regression and multinomial logistic regression. Results: In multivariate models, lower education was associated with higher FTND and HSI, income with HSI, and occupation with HSI (men only), FTND, HONC and NDSS scores. Lower education was related to higher cotinine levels among daily smokers, although the association diminished slightly after adjusting for daily smoking amount. Living without a spouse was associated with daily smoking and higher ND. Conclusion: In this cross-sectional study low SES was linked with higher ND among current smokers, while low SES was associated with higher cotinine levels among daily smokers. Living alone was linked with higher ND. Longitudinal studies are warranted to further explore these associations. As lower SES smokers are more addicted they may need more targeted cessation services to succeed in quitting smoking. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction

⁎ Corresponding author at: Hjelt Institute, Department of Public Health, University of Helsinki, P.O. Box 41, Mannerheimintie 172, 00014 Helsinki, Finland. E-mail address: jaakko.kaprio@helsinki.fi (J. Kaprio).

http://dx.doi.org/10.1016/j.addbeh.2014.03.005 0306-4603/© 2014 Elsevier Ltd. All rights reserved.

Over the past decades, smoking prevalence in Finland is declining among men on the population level and in most socio-economic groups. Among women, however, the prevalence has plateaued, with somewhat divergent trends between socio-economic groups (Helldán, Helakorpi, Virtanen, & Uutela, 2013), possibly attributable to differences

1146

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151

in initiation age, number of cigarettes per day (CPD) and nicotine dependence (ND), all of which could lead to some groups having a lower cessation rate and longer duration of smoking (Hiscock, Bauld, Amos, Fidler, & Munafo, 2012). Differences by marital status have also been observed. Married or co-habiting persons tend to smoke less frequently (Goodwin, Pagura, Spiwak, Lemeshow, & Sareen, 2011). People smoking similar numbers of cigarettes may have different nicotine intake levels, depending on smoking style (depth and volume of inhalation). Cotinine as a primary metabolite of nicotine is used as a biomarker of smoking and nicotine intake (Benowitz, Hukkanen, & Jacob, 2009). Benowitz, Dains, Dempsey, Wilson, and Jacob (2011) found that weakly dependent smokers demonstrated a linear rise in cotinine with increasing consumption of cigarettes, whereas highly dependent smokers had a flat relationship, indicating that the latter smoke cigarettes more intensively in order to maximize nicotine intake (Benowitz et al., 2011). Various measures of nicotine dependence exist. The Fagerström Test for Nicotine Dependence (FTND) combines an index of cigarette consumption with the difficulty in tolerating reduced nicotine levels. The Heaviness of Smoking Index (HSI) is a short two item version of FTND (Heatherton, Kozlowski, Frecker, & Fagerström, 1991). The Nicotine Dependence Syndrome Scale (NDSS) measures multiple dimensions of nicotine dependence, including craving, withdrawal, and tolerance (Shiffman, Waters, & Hickcox, 2004), while the Hooked On Nicotine Checklist (HONC) assesses loss of autonomy over smoking (DiFranza et al., 2002). The relationship of different measures of nicotine dependence scales and cotinine with socio-economic status within the same populationbased sample may provide insights into the basis of socioeconomic differences in smoking behavior and nicotine dependence. This could significantly enhance understanding of the relationship of socioeconomic status (SES) with nicotine dependence, and consequently improve smoking cessation strategies. Our study aim was to investigate the cross-sectional association of SES and marital status with various measures of nicotine dependence and cotinine levels. We explored whether different indicators of socioeconomic status among smokers, such as education, income, occupation, and marital status were associated with cigarettes per day, four measures of dependence, and plasma cotinine levels. The hypotheses of this study were as follows: Hypothesis 1. Low levels of SES and living without a spouse are associated with daily smoking. Hypothesis 2. In daily smokers low SES and living without a spouse are associated with more cigarettes smoked per day. Hypothesis 3. Smokers with low SES and living without a spouse have a greater degree of nicotine dependence assessed by multiple measures. Hypothesis 4. In daily smokers low SES and living without a spouse are associated with higher levels of plasma cotinine. 2. Methods 2.1. Data compilation and participants The data were derived from the National FINRISK Study that monitors levels of chronic disease risk factors every five years in Finland (Vartiainen et al., 2010). In 2007, the survey was carried out in six regions: 1) Helsinki and Vantaa, 2) Turku and Loimaa, 3) North Savo, 4) North Karelia, 5) the Oulu region, and 6) Lapland. The data were gathered into two stages. The FINRISK data for the first part of the study were collected at the beginning of 2007: participants (n = 11,953) (all 6 regions) were invited to fill in an extensive baseline questionnaire (n = 7993, 67%) and to attend a locally organized health examination in which blood samples were taken (all regions but Lapland) (Peltonen et al., 2008).

After the baseline study, a self-administered questionnaire with detailed smoking and nicotine dependence (ND) items was given to individuals who had stated during the first part of the study that they had smoked at least 100 cigarettes during their lifetime (regions 1 to 3), or that they were current smokers (regions 4 to 5) (N =1992). A few months later the same individuals were invited to participate in a study of dietary and obesity-related risk factors (Peltonen et al., 2008). Completed questionnaires were returned by mail, with one reminder (Broms et al., 2012). The number of participants in the smoking sub-study was 1746 (91% response rate). Plasma cotinine was analyzed for those who identified themselves as daily smokers during the main FINRISK data collection and responded to the tobacco-specific questionnaire. 2.2. Assessment of smoking and nicotine dependence In the baseline questionnaire, the respondents were asked whether they had ever smoked. Those stating they had never smoked were categorized as never smokers. Ever smokers were defined as those who had smoked at least 100 cigarettes in their lifetime. Three questions were used to classify ever smokers as quitters, occasional smokers or daily smokers; the latter two being current smokers. Quitters reported having been either regular or occasional smokers but were not smoking currently, and had last smoked more than a month ago. Occasional smokers reported having been regular or occasional smokers, and currently smoked occasionally. Daily smokers reported regular and current daily smoking, and had smoked ‘yesterday or today’. In order to create a variable for cigarettes per day (CPD) participants were asked to indicate the average number of both manufactured and self-rolled cigarettes they smoked per day, or had smoked before quitting. Manufactured and self-rolled cigarettes were totaled. In the baseline questionnaire, the only measure of nicotine dependence was the Heaviness of Smoking Index (HSI). The HSI score is based on two items: cigarettes per day and time to first cigarette after waking up, both scored 0 to 3. These two items were totaled (range 0 to 6) (Haddock, Lando, Klesges, Talcott, & Renaud, 1999; Heatherton et al., 1991). HSI was thus available for all current smokers (daily and occasional) (n = 1609) from the baseline questionnaire. Plasma cotinine was measured by gas chromatography from fasting plasma samples collected during the baseline clinical examination (Broms et al., 2012), but only for those who responded to the tobaccospecific questionnaire. Participants were not required to abstain from smoking before sample collection, although they were asked whether they had smoked within the last hour (Broms et al., 2012). The lower limit of quantification was set at 5 μg/L, but the cut point for active smokers was set at 10 μg/L. Analyses of cotinine were restricted to daily smokers, who have stable cotinine values due to the long halflife of cotinine. Given skewness of cotinine values, log-transformed values were used in the regression models. The tobacco-specific questionnaire provided additional information about nicotine dependence on three scales. The Fagerström Test for Nicotine Dependence (FTND) consists of six items (Heatherton et al., 1991) focusing on the physical aspects of dependence and scored 0 to 10 (Haddock et al., 1999). The Hooked On Nicotine Checklist (HONC) is a 10-item scale (sum score 0 to 10) measuring loss of autonomy over tobacco, identifying the beginning of dependence and measuring its severity (DiFranza et al., 2002). The multidimensional Nicotine Dependence Syndrome Scale (NDSS) has 31 items and a total score computed by means of regression-based algorithms based on 14 items (sum score 0 to 56) (Shiffman et al., 2004); the NDSS subscales were not used in these analyses. Higher scores imply a greater degree of nicotine dependence in all four measures. They measure somewhat different aspects of dependence. Thus, the correlation among current smokers between FTND and HONC was 0.56, between FTND and NDSS 0.69, and between HONC and NDSS 0.57. Among daily smokers cotinine showed correlations of 0.49, 0.19, and 0.34 with FTND, HONC, and NDSS, respectively.

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151

2.3. Socio-economic status and marital status The baseline data included three SES indicators: education, family income, and occupation. The question covering education was: “How many years of schooling and full-time education have you had (including basic levels)?”. Family annual income was assessed on nine levels in steps of ten thousand Euros, the lowest being under 10,000 Euros and the highest over 80,000 Euros. Family income was adjusted for family size using the OECD equivalence scale (the first adult in the household is weighted as 1.0, other adults as 0.7 and children under 17 as 0.5) (Harald, Salomaa, Jousilahti, Koskinen, & Vartiainen, 2007). Occupation was assessed using the open question: “What is your occupation? (If you are at present pensioned or unemployed, give your most recent occupation)”. We used Statistics Finland coding guidelines to create the occupational categories: 1) managers; 2) professionals (e.g., teachers); 3) semi-professionals (e.g. technicians); 4) clerical support workers; 5) service and sales workers; 6) skilled agricultural, forestry and fishery workers; 7) craft and related trade workers; 8) plant and machine operators and assemblers; 9) elementary occupations; 10) armed-forces occupations; 11) students; 12) homemakers and housewives; and 13) not known. We recoded them as: 1) upper white-collar (categories 1–3); 2) lower white-collar (4 & 5); 3) bluecollar (6 to 9); and 4) others (10–13). The question: “What is your marital status?” had five response options, which were recoded as: 1) living with a spouse/partner (married, cohabiting), and 2) living without a spouse (single, separated or divorced, widowed).

2.4. Analyses We used the baseline data to confirm the expected associations of SES (education, family income, and occupation) and marital status with smoking status (see Supplementary material), while cigarettes per day (CPD) and Heaviness of Smoking Index (HSI) were examined among current smokers (daily and occasional smokers). CPD and HSI were used as dependent variables, and SES and marital status as independent variables in the linear regression models.

1147

Among respondents to the tobacco-specific questionnaire, we analyzed the associations of SES and marital status with three other measures of dependence among current smokers (n = 884) by linear regression. The corresponding analysis for plasma cotinine was examined among daily smokers only (n = 557). We supplemented the above analyses with multivariate analyses that included all independent variables (SES and marital status) in order to assess their independent contribution to CPD, four measures of nicotine dependence, and cotinine. All variables except occupational category and marital status were used as continuous variables in the analyses; dummy variables were created for the two categorical variables in the regression models. All the analyses were adjusted for age and sex (if not reported otherwise). There are considerable differences in smoking status between men and women in the Finnish population. In order to see if the association of SES with smoking differs by sex, we tested the sex by SES-interactions, and report sex-specific analyses where significant interactions existed. Likewise we tested whether the SES associations with nicotine dependence measures differed between daily and occasional current smokers. The quantity smoked daily correlated positively with nicotine dependence measurements and cotinine. For FTND and HSI this is structural as both scales include an item on amount smoked. Because HONC and NDSS do not ask the amount smoked we adjusted for cigarettes per day to test whether the association of SES with these two measures of dependence (HONC, NDSS) and cotinine was independent of smoking quantity. PASW 18 statistical software was used for the data analysis. p-Values less than 0.05 were considered significant and 95% confidence intervals are presented.

3. Results 3.1. Basic characteristics Table 1 summarizes the characteristics of the FINRISK baseline survey and of the ever-smoking respondents to the tobacco-specific

Table 1 Characteristics of study participants by sex. Total baseline data

Whole sample N Age (Mean, SD) Marital status (%) Living with a spouse Living without a spouse Smoking (%) Never Former Occasionally Daily Education years (Mean, SD) Family income in Euros (%) Less than 10,000 10,001–20,000 20,001–30,000 30,001–40,000 40,001–50,000 50,001–60,000 60,001–70,000 70,001–80,000 Over 80,000 Occupation (%) Upper white-collar Lower white-collar Blue-collar Others

Tobacco-specific study participants

Men

Women

Total

Men

Women

Total

3740 51.5 (14.0)

4253 50.2 (14.1)

7993 50.8 (14.0)

977 53.5 (13.2)

769 49.1 (12.8)

1746 51.6 (13.2)

73.9 26.1

69.3 30.7

71.5 28.5

74.3 25.7

65.9 34.1

70.6 29.4

35.1 35.3 8.3 21.4 12.2 (3.9)

55.8 23.8 6.1 14.3 13.0 (4.0)

46.1 29.2 7.1 17.6 12.7 (4.0)

– 49.3 15.0 35.7 12.1 (3.8)

– 47.7 16.4 35.9 13.0 (3.7)

– 48.6 15.6 35.4 12.5 (3.8)

7.0 14.9 18.3 16.5 14.3 11.6 6.6 3.9 7.1

7.9 16.8 20.0 15.4 12.7 11.2 6.5 3.7 5.9

7.5 15.9 19.2 15.9 13.4 11.4 6.5 3.8 6.4

5.7 15.3 18.6 17.4 12.8 10.6 7.1 4.8 7.6

6.8 16.5 17.8 15.0 13.5 13.7 6.7 4.2 5.7

6.2 15.8 18.2 16.4 13.1 12.0 6.9 4.6 6.8

35.5 11.4 36.5 16.6

33.9 39.9 15.7 10.5

34.7 26.6 25.4 13.4

37.7 12.7 37.5 12.2

30.9 42.6 18.1 8.5

34.6 26.0 28.9 10.5

1148

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151

Table 2 Means (and standard deviations) of nicotine dependence measures among current (daily and occasional) smokers, and plasma cotinine among daily smokers by socio-economic status.

Education Low High Family income in Euros Less than 10,000 10,001–20,000 20,001–30,000 30,001–40,000 40,001–50,000 50,001–60,000 60,001–70,000 70,001–80,000 Over 80,000 Occupation Others Blue-collar Lower white-collar Upper white-collar

Baseline data

Tobacco-specific study participants

HSI (n = 1609)

Cotinine (n = 557)

FTND (n = 776)

HONC (n = 858)

NDSS (n = 863)

2.50 (1.53) 1.77 (1.50)

187.24 (92.87) 158.96 (93.48)

3.66 (2.53) 2.54 (2.50)

5.70 (2.90) 5.17 (2.96)

18.93 (8.19) 16.06 (8.25)

2.39 (1.66) 2.44 (1.52) 2.18 (1.57) 2.01 (1.51) 1.89 (1.50) 1.74 (1.55) 2.04 (1.32) 1.37 (1.42) 1.66 (1.55)

175.58 (90.65) 186.81 (97.42) 173.74 (100.42) 177.48 (94.95) 150.26 (84.50) 151.46 (97.41) 213.16 (87.61) 159.84 (65.50) 173.25 (92.92)

3.78 (2.73) 3.76 (2.70) 3.01 (2.53) 3.01 (2.46) 2.79 (2.43) 2.19 (2.56) 2.74 (2.37) 2.30 (2.42) 2.51 (2.45)

5.87 (2.97) 5.93 (3.05) 5.55 (2.73) 5.65 (2.89) 5.23 (2.98) 4.62 (2.70) 4.68 (2.70) 4.24 (3.19) 4.96 (3.14)

20.47 (10.80) 18.92 (8.59) 17.45 (7.70) 17.42 (8.41) 16.32 (7.25) 15.76 (8.38) 15.83 (6.76) 15.74 (6.76) 14.44 (8.13)

2.07 (1.59) 2.56 (1.50) 1.85 (1.43) 1.66 (1.54)

183.91 (96.51) 178.02 (97.95) 170.98 (90.61) 155.32 (89.35)

3.59 (2.87) 3.56 (2.40) 2.76 (2.37) 2.24 (2.55)

5.99 (2.96) 5.89 (2.78) 5.04 (2.90) 4.81 (2.99)

19.60 (8.48) 19.01 (8.27) 15.97 (7.46) 15.24 (8.35)

HSI = Heaviness of Smoking Index, FTND = Fagerström Test for Nicotine Dependence, HONC = Hooked On Nicotine Checklist, NDSS = Nicotine Dependence Syndrome Scale. Education low = less than 12 years, education high = 12 years or more.

questionnaire. Descriptive statistics of the four dependence scales and plasma cotinine by socio-economic status are shown in Table 2. 3.2. The association of socio-economic status and marital status with smoking behavior We replicated well-known associations of low SES and being outside a regular relationship with smoking status and amount smoked as summarized in the Supplementary material. 3.3. The association of socio-economic status and marital status with ND and cotinine Among current smokers, fewer years of education, lower income and blue-collar occupation increased the probability of higher dependence as measured by all four scales in univariate models (Tables 3 and 4). After adjusting for CPD, all associations between SES indicators and HONC and NDSS become non-significant (Table 4). In the multivariate models that adjusted for all three SES variables and marital status simultaneously, the following associations remained statistically significant: education associated with HSI and FTND, income with HSI, and occupation with HSI (men only), FTND, HONC and NDSS. Living without a spouse was associated with higher dependence on three scales (FTND, HONC and NDSS). Among the women, living without a spouse was also associated with higher HSI scores. After further adjustment for CPD there was no longer a significant association between marital status and either HONC or NDSS (Table 4). Given that occasional and daily smokers may differ in their SES relationships with nicotine dependence, we examined possible SES and smoking type (daily vs. occasional) interaction effects on nicotine dependence. None of these were statistically significant with the only exception being the association between income and smoking in the HONC (p = 0.019). Among daily smokers, fewer years of education and a lower occupational group were associated with higher levels of cotinine. Even after multivariate adjustment for all socioeconomic variables, marital status, and amount smoked, the association of fewer years of education with higher levels of cotinine remained statistically significant. Marital status was not associated with level of cotinine (Table 5). We found no substantial sex differences in the associations between socioeconomic status variables and both nicotine dependence measures and cotinine, indicated by a lack of significant interactions of sex with SES.

4. Discussion According to this cross-sectional study, lower SES based on three different measures was associated with daily smoking, as expected. In addition, among current smokers there was an association in multiple measures of lower SES with greater number of cigarettes smoked per day and higher nicotine dependence. In daily smokers low SES was also associated with higher levels of plasma cotinine. These findings not only confirm but also extend the results of earlier research showing that individuals with lower SES are more likely to be daily smokers and to smoke more, and consequently, as smokers, to be more dependent on nicotine than people in higher socio-economic groups (Hiscock et al., 2012). Our results indicate that smokers with a low SES have a greater degree of nicotine dependence as assessed by all four measures of nicotine dependence. This is in accordance with the findings from three countries showing an association between higher dependence, measured by HSI, and low education and income (Siahpush, McNeill, Borland, & Fong, 2006). Actually, even though CPD is not included in the HONC and NDSS, these scales seem to capture similar aspects of dependence as amount smoked, as adjustment with CPD diminished the significance of the SES association. There still remains some independent association, however, so dependence measures provide more information about SES differences than smoking amount by itself. A recent study using DSM-IV diagnostic criteria for nicotine dependence suggests that both quantity and frequency of use differentiate dependence status among cigarette users (Moss, Chen, & Yi, 2012). Lower education is associated with higher cotinine levels among daily smokers. Cotinine is an indicator of nicotine intake influenced by actual number of cigarettes smoked, smoking patterns, and metabolism (Perez-Stable, Herrera, Jacob, & Benowitz, 1998). We could not fully attribute socio-economic differences to the self-reported amount smoked alone, as adjustment for CPD decreased the association only slightly. Thus lower SES was associated with greater nicotine intake, or possibly different metabolism, at apparently equivalent levels of cigarettes smoked. This could imply that less educated smokers smoke differently, which may simply reflect the economic need for value for money, or else a more complex response to nicotine addiction (Bobak, Jarvis, Skodova, & Marmot, 2000). Higher proportions of self-rolled cigarettes among less educated smokers may also influence these results (those with education years b12 consumed on average 3.57 self-rolled CPD, vs. 1.42 self-rolled CPD among those with education years 12 or more). Self-rolled cigarettes have higher nicotine yields than manufactured cigarettes (Darrall & Figgins, 1998). Smokers of

Table 3 Results of univariate and multivariate linear regression models for HSI and FTND by SES and marital status among current (daily and occasional) smokers. Adjusted for sex and age as indicated. HSI (n = 1609) from baseline FINRISK data

FTND (n = 776)

Univariate models

Multivariate models

Men adjusted for age

Univariate models

Men adjusted for age

Women adjusted for age

Multivariate models

Adjusted for age and sex

Adjusted for age and sex

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

−.11 (−.14 to −.07) −.19 (−.27 to −.10)

b.001 b.001

−.09 (−.12 to −.05) −.17 (−.25 to −.09)

b.001 b.001

−.06 (−.10 to −.02) −.13 (−.21 to −.04)

.008 .006

−.09 (−.14 to −.05) −.11 (−.19 to −.03)

b.001 .008

−.17 (−.22 to −.11) −.28 (−.42 to −.13)

b.001 .002

−.10 (−.18 to −.03) −.05 (−.20 to .10)

0 .29 (−.06 to .63) .80 (.54 to 1.05) .33 (−.03 to .69) .22 (.00 to .43)

.103 b.001 .071 .045

0 .35 (.09 to .61) .74 (.43 to 1.05) .48 (.08 to .88) .31 (.10 to .53)

.010 b.001 .015 .005

0 −.01 (−.39 to .36) .43 (.12 to .73) −.05 (−.44 to .35) .13 (−.10 to .35)

.844 .007 .772 .258

0 −.08 (−.38 to .22) .23 (−.15 to .61) .27 (−.15 to .69) .30 (.08 to .52)

0 .57 (.09 to 1.06) 1.2 (.76 to 1.66) 1.28 (.62 to 1.95) .64 (.26 to 1.01)

.025 b.001 b.001 b.001

0 .01 (−.56 to .54) .60 (.02 to 1.18) .79 (.07 to 1.52) .71 (.33 to 1.10)

.666 .213 .227 .007

p .006 .465

.962 .044 .032 b.001

SES = socio-economic status, HSI = Heaviness of Smoking Index, FTND = Fagerström Test for Nicotine Dependence, CPD = cigarettes per day, living with a spouse: 1 = yes and 2 = no, β = standardized regression coefficient, CIs = confidence intervals.

Table 4 Results of univariate and multivariate linear regression models for HONC and NDSS from the tobacco-specific questionnaire by SES and marital status among current (daily and occasional) smokers. Adjusted for sex, age and CPD as indicated. HONC (n = 858)

NDSS (n = 863)

Univariate models Adjusted for age and sex

Education Income Occupation Upper white-collar Lower white-collar Blue-collar Others Living without a spouse

Adjusted for age, sex and CPD

Multivariate models

Univariate models

Adjusted for age and sex

Adjusted for age and sex

Multivariate models Adjusted for age, sex and CPD

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151

Education Income Occupation Upper white-collar Lower white-collar Blue-collar Others Living without a spouse

Women adjusted for age

Adjusted for age and sex

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

−.12 (−.18 to −.06) −.26 (−.41 to −.10)

b.001 .012

−.03 (−.09 to .03) −.12 (−.26 to .02)

.327 .274

−.06 (−.14 to .03) −.06 (−.23 to .11)

.160 .461

−.43 (−.61 to −.25) −.90 (−1.33 to −.46)

b.001 .001

−.15 (−.31 to .01) −.49 (−.86 to .02)

.063 .066

−.22 (−.44 to .01) −.23 (−.68 to .28)

.063 .308

0 .20 (−.34 to .74) 1.12 (.62 to 1.62) 1.15 (.42 to 1.88) .46 (.05 to .88)

.517 b.001 .005 .018

0 −.11 (−.59 to .38) .37 (−.09 to .83) .70 (−.01 to 1.38) −.04 (−.42 to .33)

.446 .229 .053 .975

0 −.21 (−.84 to .41) .68 (.01 to 1.35) .88 (.06 to 1.70) .53 (.09 to .97)

.506 .046 .036 .019

0 1.04 (−.48 to 2.55) 3.44 (2.05 to 4.84) 4.23 (2.18 to 6.28) 1.48 (.30 to 2.65)

.253 b.001 b.001 .003

0 −.14 (−1.43 to 1.15) .85 (−.36 to 2.07) 2.39 (.60 to 4.18) .04 (−.95 to 1.04)

.664 .146 .011 .600

0 −.36 (−2.08 to 1.36) 2.32 (.50 to 4.14) 2.98 (.74 to 5.22) 1.49 (.28 to 2.70)

.648 .012 .009 .016

SES = socio-economic status, HONC = Hooked On Nicotine Checklist, NDSS = Nicotine Dependence Syndrome Scale, CPD = cigarettes per day, living with a spouse: 1 = yes and 2 = no, β = standardized regression coefficient, CIs = confidence intervals.

1149

1150

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151

Table 5 Results of univariate and multivariate linear regression models (beta & 95% confidence intervals) for level of cotinine (N10 μg/L) among daily smokers by SES and marital status. Level of cotinine (n = 557) Univariate models

Multivariate models

Adjusted for age and sex

Education Income Occupation Upper white-collar Lower white-collar Blue-collar Others Living without a spouse

Adjusted for age, sex and CPD

Adjusted for age, sex and CPD

β (CI 95%)

p

β (CI 95%)

p

β (CI 95%)

p

−.03 (−.05 to −.01) −.03 (−.07 to .01)

.003 .162

−.02 (−.04 to −.01) −.01 (−.05 to .03)

.016 .709

−.03 (−.05 to −.01) .01 (−.04 to .05)

.024 .734

0 .19 (.03 to .35) .15 (.01 to .29) .21 (.00 to .41) .04 (−.07 to .16)

.024 .043 .050 .486

0 .15 (.01 to .30) .06 (−.07 to .20) .15 (−.05 to .34) −.04 (−.15 to .07)

.047 .353 .140 .459

0 .08 (−.10 to .25) −.05 (−.23 to .13) .10 (−.13 to .31) −.05 (−.16 to .07)

.385 .607 .417 .405

SES = socio-economic status, CPD = cigarettes per day, living with a spouse: 1 = yes and 2 = no, β = standardized regression coefficient, CIs = confidence intervals.

self-rolled cigarettes tend to smoke more intensively (more puff, more smoke per cigarette) and for longer as well as are less likely using a filter compared to smokers of manufactured cigarettes (Laugesen, Epton, Frampton, Glover, & Lea, 2009). There may also be other factors affecting smoking patterns among people with low SES, warranting further study. Interestingly, daily smoking (vs. never smoking) was associated with lower SES levels, yet occasional smoking showed weaker relationships (Supplementary material). However, due to limited statistical power, we could not demonstrate significant SES by smoking status (daily vs. occasional) interaction for dependence measures. A Swedish study (Lindström & Ostergren, 2001) found that daily as opposed to occasional smoking was related to lower SES and less social participation. A Finnish study reported occasional smokers having more education, healthier lifestyles, and more favorable mental health than regular smokers (Korhonen, Broms, Levalahti, Koskenvuo, & Kaprio, 2009). According to a review, occasional smokers have higher SES, quit rates, and lower nicotine dependence than daily smokers (Coggins, Murrelle, Carchman, & Heidbreder, 2009). Socio-economic differences among smokers may also be attributed to social disadvantages associated with lower SES. For instance, individuals in lower positions are more likely to share social environments with smokers: smoking is more frequent both inside and outside their homes this being associated with initiation, continued smoking, and relapse after quitting (Hiscock et al., 2012). Smoking may make it easier to socialize with others and increase the sense of belonging (Katainen, 2011). Furthermore, less educated people may be less responsive to health promotion and have less information about the health consequences of smoking and less access to cessation services (BenjaminGarner et al., 2002; Hiscock et al., 2012). We found that smokers living without a spouse have higher nicotine dependence as measured on four scales, although this association disappeared following adjustment of HONC and NDSS for CPD. Supporting previous findings (Lindström, 2010), our study indicates that living without a spouse increases daily smoking rates among both sexes. Among male smokers living alone was also associated with a lower probability of quitting, as found previously (Broms, Silventoinen, Lahelma, Koskenvuo, & Kaprio, 2004). A longitudinal study (Giordano & Lindström, 2011) reported a positive association between cessation and support mechanisms (via marriage and employment) as well as social capital (trust and social participation), but no links with income or baseline occupation. Cessation may not reflect increased financial security, but may foster active participation and social integration (Giordano & Lindström, 2011). The measures we used may reveal different aspects of nicotine dependence, as suggested by the variability in associations with socioeconomic variables among smokers and their sensitivity to adjustment for amount smoked. HSI and FTND measure cigarette consumption in terms of the difficulty in tolerating reduced nicotine levels (Heatherton

et al., 1991) and predict smoking cessation (Kozlowski, Porter, Orleans, Pope, & Heatherton, 1994). HONC measures dependence in terms of loss of autonomy over smoking behavior (Wellman et al., 2006), which can occur even at very low levels of nicotine exposure (DiFranza et al., 2002). NDSS measures multiple aspects of dependence such as craving, withdrawal, and tolerance (Shiffman et al., 2004), but here the scale is used as a unitary scale. Nicotine dependence is a great barrier to smoking cessation since smoking prevents and reduces nicotine withdrawal symptoms (Silagy, Lancaster, Stead, Mant, & Fowler, 2004). It seems that stronger dependence among people with low SES contributes to socio-economic differences in successful quitting to some extent. 4.1. Limitations and strengths Some limitations must be considered when interpreting these results. The study was cross-sectional, so evidence associating SES and marital status with nicotine dependence is restricted. The National FINRISK Study is based on extensive, recently collected populationbased data, and thus accurately captures current smoking prevalence in Finland. The sub-sample of those responding to the tobacco-specific questionnaire is similarly representative in that prevalence and SES distribution were similar to the rates in the whole dataset. A further strength is that we were able to apply multiple dependence phenotypes, which is rare in population-based studies (Broms et al., 2012). We also had access to cotinine measurements. Cotinine levels accurately reflect nicotine intake, with a correlation of 0.7 in experimental conditions, whereas the correlation between cotinine and amount smoked was around 0.5 (Benowitz et al., 2012). 5. Conclusion Our results suggest that individuals in lower socioeconomic groups may be more nicotine-dependent because they smoke more cigarettes and tend to be daily smokers more often than individuals with higher status. Furthermore, even with equivalent amounts of cigarettes smoked, lower SES groups had higher cotinine levels, suggesting that these individuals are smoking their cigarettes more intensely or completely. In contrast to the extensive literature on smoking status and SES, only a limited body of research has examined the association of SES and nicotine dependence among smokers. As lower SES smokers are more addicted they may need more targeted cessation services to succeed in quitting smoking. One important initiative would be the provision of proactive and effective nationwide smoking cessation services, including affordable behavioral support and low-cost pharmacological treatment for disadvantaged smokers (Hiscock, Judge, & Bauld, 2011; Murray, Bauld, Hackshaw, & McNeill, 2009).

M. Pennanen et al. / Addictive Behaviors 39 (2014) 1145–1151 Role of funding sources This study was supported by the Cancer Society of Finland and the Juho Vainio Foundation. No funders had any direct input on the study design, collection, analysis, or interpretation of data, writing the manuscript, and the decision to submit the manuscript for publication. Contributors Authors Marjaana Pennanen, Ulla Broms and Jaakko Kaprio designed the study. Marjaana Pennanen wrote the first draft of the manuscript and conducted the statistical analysis. All authors contributed to and have approved the final manuscript. Conflict of interest Jaakko Kaprio and Tellervo Korhonen have consulted for Pfizer on nicotine dependence in 2011 and 2012. The other authors have no conflicts to declare.

Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.addbeh.2014.03.005. References Benjamin-Garner, R., Oakes, J. M., Meischke, H., Meshack, A., Stone, E. J., Zapka, J., et al. (2002). Sociodemographic differences in exposure to health information. Ethnicity & Disease, 12(1), 124–134. Benowitz, N. L., Dains, K. M., Dempsey, D., Wilson, M., & Jacob, P. (2011). Racial differences in the relationship between number of cigarettes smoked and nicotine and carcinogen exposure. Nicotine & Tobacco Research, 13(9), 772–783. http://dx.doi.org/10.1093/ntr/ ntr072. Benowitz, N. L., Hukkanen, J., & Jacob, P., III (2009). Nicotine chemistry, metabolism, kinetics and biomarkers. Handbook of Experimental Pharmacology, 192, 29–60. http://dx.doi.org/ 10.1007/978-3-540-69248-5_2. Benowitz, N. L., Renner, C. C., Lanier, A. P., Tyndale, R. F., Hatsukami, D. K., Lindgren, B., et al. (2012, Jun). Exposure to nicotine and carcinogens among Southwestern Alaskan Native cigarette smokers and smokeless tobacco users. Cancer epidemiology, biomarkers & prevention: A publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology, 21(6), 934–942. http://dx. doi.org/10.1158/1055-9965.EPI-11-1178 (Epub 2012 Apr 6). Bobak, M., Jarvis, M. J., Skodova, Z., & Marmot, M. (2000). Smoke intake among smokers is higher in lower socioeconomic groups. Tobacco Control, 9(3), 310–312. http://dx.doi.org/10.1136/tc.9.3.310. Broms, U., Pennanen, M., Patja, K., Ollila, H., Korhonen, T., Haukkala, A., et al. (2012, Jan 25). Diurnal evening type is associated with current smoking and nicotine dependence in population based FINRISK 2007 Study. Journal of Addiction Research & Therapy(S2:002). http://dx.doi.org/10.4172/2155-6105 (S2-002). Broms, U., Silventoinen, K., Lahelma, E., Koskenvuo, M., & Kaprio, J. (2004, Jun). Smoking cessation by socioeconomic status and marital status: The contribution of smoking behavior and family background. Nicotine & Tobacco Research, 6(3), 447–455. http://dx.doi.org/10.1080/14622200410001696637. Coggins, C. R., Murrelle, E. L., Carchman, R. A., & Heidbreder, C. (2009). Light and intermittent cigarette smokers: A review (1989–2009). Psychopharmacology, 207(3), 343–363. http://dx.doi.org/10.1007/s00213-009-1675-4. Darrall, K. G., & Figgins, J. A. (1998). Roll-your-own smoke yields: Theoretical and practical aspects. Tobacco Control, 7(2), 168–175. http://dx.doi.org/10.1136/tc.7.2.168. DiFranza, J. R., Savageau, J. A., Rigotti, N. A., Fletcher, K., Ockene, J. K., McNeill, A.D., et al. (2002). Development of symptoms of tobacco dependence in youths: 30 month follow up data from the DANDY study. Tobacco Control, 11, 228–235. http://dx.doi.org/ 10.1136/tc.11.3.228. Giordano, G. N., & Lindström, M. (2011). The impact of social capital on changes in smoking behavior: A longitudinal cohort study. European Journal of Public Health, 21(3), 347–354. http://dx.doi.org/10.1093/eurpub/ckq048. Goodwin, R. D., Pagura, J., Spiwak, R., Lemeshow, A.R., & Sareen, J. (2011). Predictors of persistent nicotine dependence among adults in the United States. Drug and Alcohol Dependence, 118(2–3), 127–133. http://dx.doi.org/10.1016/j.drugalcdep.2011.03.010. Haddock, C. K., Lando, H., Klesges, R. C., Talcott, G. W., & Renaud, E. A. (1999). A study of the psychometric and predictive properties of the Fagerström Test for Nicotine Dependence in a population of young smokers. Nicotine & Tobacco Research, 1(1), 59–66. http://dx.doi.org/10.1080/14622299050011161.

1151

Harald, K., Salomaa, V., Jousilahti, P., Koskinen, S., & Vartiainen, E. (2007). Nonparticipation and mortality in different socioeconomic groups: The FINRISK population surveys in 1972–92. Journal of Epidemiology and Community Health, 61(5), 449–454. http://dx.doi.org/10.1136/jech.2006.049908. Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerström, K. O. (1991). The Fagerström Test for Nicotine Dependence: A revision of the Fagerström Tolerance Questionnaire. British Journal of Addiction, 86(9), 1119–1127. Helldán, A., Helakorpi, S., Virtanen, S., & Uutela, A. (2013). Health behaviour and health among the Finnish adult population, Spring 2012. (Report 15/2013, 192 pages). 1798-0070Helsinki: Publications of National Institute for Health and Welfare978952-245-930-5 ((print), 978-952-245-931-2 (pdf), (print), 1798-0089 (pdf)). Hiscock, R., Bauld, L., Amos, A., Fidler, J. A., & Munafo, M. (2012). Socioeconomic status and smoking: A review. Annals of the New York Academy of Sciences, 1248, 107–123. http://dx.doi.org/10.1111/j.1749-6632.2011.06202.x. Hiscock, R., Judge, K., & Bauld, L. (2011). Social inequalities in quitting smoking: What factors mediate the relationship between socioeconomic position and smoking cessation? Journal of Public Health, 33(1), 39–47. http://dx.doi.org/10.1093/pubmed/ fdq097. Epub 2010 Dec 22. Katainen, A. (2011). Smoking and workers' autonomy: A qualitative study on smoking practices in manual work. Health (London, England), 16(2), 134–150. http://dx.doi. org/10.1177/1363459311403944. Korhonen, T., Broms, U., Levalahti, E., Koskenvuo, M., & Kaprio, J. (2009). Characteristics and health consequences of intermittent smoking: Long-term follow-up among Finnish adult twins. Nicotine & Tobacco Research, 11(2), 148–155. http://dx.doi.org/ 10.1093/ntr/ntn023. Kozlowski, L. T., Porter, C. Q., Orleans, C. T., Pope, M.A., & Heatherton, T. (1994). Predicting smoking cessation with self-reported measures of nicotine dependence: FTQ, FTND, and HSI. Drug and Alcohol Dependence, 34(3), 211–216. Laugesen, M., Epton, M., Frampton, C. M., Glover, M., & Lea, R. A. (2009). Hand-rolled cigarette smoking patterns compared with factory-made cigarette smoking in New Zealand men. BMC Public Health, 9, 194. http://dx.doi.org/10.1186/1471-2458-9-194. Lindström, M. (2010). Social capital, economic conditions, marital status and daily smoking: A population-based study. Public Health, 124(2), 71–77. http://dx.doi.org/ 10.1016/j.puhe.2010.01.003. Lindström, M., & Ostergren, P. O. (2001). Intermittent and daily smokers: Two different socioeconomic patterns, and diverging influence of social participation. Tobacco Control, 10(3), 258–266. http://dx.doi.org/10.1136/tc.10.3.258. Moss, H. B., Chen, C. M., & Yi, H. Y. (2012). Measures of substance consumption among substance users, DSM-IV abusers, and those with DSM-IV dependence disorders in a nationally representative sample. Journal of Studies on Alcohol and Drugs, 73(5), 820–828. Murray, R. L., Bauld, L., Hackshaw, L. E., & McNeill, A. (2009). Improving access to smoking cessation services for disadvantaged groups: A systematic review. Journal of Public Health, 31(2), 258–277. http://dx.doi.org/10.1093/pubmed/fdp008. Epub 2009 Feb 10. Peltonen, M., Harald, K., Männistö, S., Saarikoski, L., Peltomäki, P., Lund, L., et al. (2008). The National FINRISK 2007 Study. (No. B34/2008, 72 pages). 0359-3576. Helsinki: Publications of the National Public Health Institute. 978-951-740-911-7 ((print), 978-951-740-905-6 (pdf)). Perez-Stable, E. J., Herrera, B., Jacob, P., III, & Benowitz, N. L. (1998). Nicotine metabolism and intake in black and white smokers. JAMA: The Journal of the American Medical Association, 280(2), 152–156. Shiffman, S., Waters, A., & Hickcox, M. (2004). The Nicotine Dependence Syndrome Scale: A multidimensional measure of nicotine dependence. Nicotine & Tobacco Research, 6(2), 327–348. http://dx.doi.org/10.1080/1462220042000202481. Siahpush, M., McNeill, A., Borland, R., & Fong, G. (2006). Socioeconomic variation in nicotine dependence, self-efficacy, and intention to quit across four countries: Findings from the international tobacco control (ITC) four country survey. Tobacco Control, 15, iii71–iii75. http://dx.doi.org/10.1136/tc.2004.008763. Silagy, C., Lancaster, T., Stead, L., Mant, D., & Fowler, G. (2004). Nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews(3). http://dx.doi. org/10.1002/14651858.CD000146.pub2 (Art. No.: CD000146). Vartiainen, E., Laatikainen, T., Peltonen, M., Juolevi, A., Männistö, S., Sundvall, J., et al. (2010). Thirty-five-year trends in cardiovascular risk factors in Finland. International Epidemiological Association, 39(2), 504–518. http://dx.doi.org/10.1093/ ije/dyp330 (Epub 2009 Dec 3). Wellman, R. J., Savageau, J. A., Godiwala, S., Savageau, N., Friedman, K., Hazelton, J., et al. (2006). A comparison of the Hooked On Nicotine Checklist and the Fagerström Test for Nicotine Dependence in adult smokers. Nicotine & Tobacco Research, 8(4), 575–580. http://dx.doi.org/10.1080/14622200600789965.