motives to quit smoking and reasons to relapse differ by socioeconomic status

5
Motives to quit smoking and reasons to relapse differ by socioeconomic status Charlotta Pisinger , Mette Aadahl, Ulla Toft, Torben Jørgensen Research Centre for Prevention and Health, The Capital Region of Denmark, Denmark abstract article info Available online 1 November 2010 Keywords: Smoking Smoking cessation Motives Relapse Inter99 study Socioeconomic factors Educational status Social class Smoking/prevention and control Objective. To investigate motives, strategies and experiences to quit smoking and reasons to relapse as a function of socioeconomic status. Methods. A population-based study, Inter99, Denmark. Two thousand six hundred twenty-one daily smokers with a previous quit attempt completed questionnaires at baseline. Cross-sectional baseline-data (19992001) were analysed in adjusted regression analyses. Results. Consistent ndings across the three indicators of socioeconomic status (employment, school education, higher education/vocational training): smokers with low socioeconomic status were signicantly more likely than smokers with high socioeconomic status to report that they wanted to quit because smoking was too expensive (OR: 1.85 (1.42.4), for school education) or because they had health related problems (OR: 1.75 (1.42.2)). When looking at previous quit attempts, smokers with low socioeconomic status were signicantly more likely to report that it had been a bad experience (OR: 1.41 (1.11.8)) and that they had relapsed because they were more nervous/restless/depressed (OR: 1.43 (1.11.8)). Conclusions. This study shows that smokers with low socioeconomic status have other motives to quit and other reasons to relapse than smokers with high socioeconomic status. Future tobacco prevention efforts aimed at smokers with low socioeconomic status should maybe focus on current advantages of quitting smoking, using high cost of smoking and health advantages of quitting as motivating factors and by including components of mental health as relapse prevention. © 2010 Elsevier Inc. All rights reserved. Introduction Many studies have examined smokers' motivation to quit and data strongly suggest that health concern is the primary motive for quit attempts and that health professionals should continue emphasising the negative health consequences of smoking to motivate cessation attempts (McCaul et al., 2006). However, a recent publication found that there are signicant differences in reported triggers for quit attempts as a function of socio-demographic factors. Most notably, smokers with higher socioeconomic status (SES) were more likely to report concern about future health whereas those from lower SES were more likely to cite cost and current health problems (Vangeli and West, 2008). The steep social class gradient in smoking has worsened over the past 20 years (Millward et al., 2007). It has been suggested that the most important strategy to reduce health inequalities related to social position is to promote cessation in smokers with low social position (Gruer et al., 2009). Regrettably, the English report entitled Securing good health for the whole population concluded, We do not know what messages and interventions work to get lower socio economic groups to stop smoking(Wanless, 2004). The conclusion was the same in the 2008-update of the U.S. Public Health Service Clinical Practice Guideline, which underlined that new counselling strategies should be developed and research should focus on the development of effective interventions for populations that carry a disproportionate burden from tobacco e.g., individuals with limited educational attainment/low SES (2008 Update Panel, L.a.S., 2008). Before designing interventions and counselling strategies to smokers with low SES, we need to know more about their motives to quit, strategies used at previous quit attempts, experience with previous quit attempts, and reasons to relapse. Only one previous study has examined triggers/motives to quit as a function of socio- demographic factors (Vangeli and West, 2008). The aim of this article is, in a population-based sample of daily smokers from the Inter99 study, to investigate motives to quit, strategies used at previous quit attempts, experiences with previous quit attempts and reasons to relapse as a function of socioeconomic status. Materials and methods Inter99 is a population-based intervention study initiated in March 1999 and ended in April 2006. The study design is described in detail elsewhere (Jorgensen et al., 2003; The Inter99 Steering Committee, 2008). Preventive Medicine 52 (2011) 4852 Corresponding author. Forskningscenter for Forebyggelse og Sundhed, Nordre Ringvej, Glostrup University Hospital, Bygning 84/85, DK-2600 Glostrup, Denmark. Fax: +45 3863 3283. E-mail address: [email protected] (C. Pisinger). 0091-7435/$ see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2010.10.007 Contents lists available at ScienceDirect Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed

Upload: charlotta-pisinger

Post on 31-Oct-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Preventive Medicine 52 (2011) 48–52

Contents lists available at ScienceDirect

Preventive Medicine

j ourna l homepage: www.e lsev ie r.com/ locate /ypmed

Motives to quit smoking and reasons to relapse differ by socioeconomic status

Charlotta Pisinger ⁎, Mette Aadahl, Ulla Toft, Torben JørgensenResearch Centre for Prevention and Health, The Capital Region of Denmark, Denmark

⁎ Corresponding author. Forskningscenter for ForebRingvej, Glostrup University Hospital, Bygning 84/85, DK+45 3863 3283.

E-mail address: [email protected] (C. Pisinger).

0091-7435/$ – see front matter © 2010 Elsevier Inc. Aldoi:10.1016/j.ypmed.2010.10.007

a b s t r a c t

a r t i c l e i n f o

Available online 1 November 2010

Keywords:SmokingSmoking cessationMotivesRelapseInter99 studySocioeconomic factorsEducational statusSocial classSmoking/prevention and control

Objective. To investigate motives, strategies and experiences to quit smoking and reasons to relapse as afunction of socioeconomic status.

Methods. A population-based study, Inter99, Denmark. Two thousand six hundred twenty-one dailysmokers with a previous quit attempt completed questionnaires at baseline. Cross-sectional baseline-data(1999–2001) were analysed in adjusted regression analyses.

Results. Consistent findings across the three indicators of socioeconomic status (employment, schooleducation, higher education/vocational training): smokers with low socioeconomic status were significantlymore likely than smokers with high socioeconomic status to report that they wanted to quit because smokingwas too expensive (OR: 1.85 (1.4–2.4), for school education) or because they had health related problems(OR: 1.75 (1.4–2.2)). When looking at previous quit attempts, smokers with low socioeconomic status weresignificantly more likely to report that it had been a bad experience (OR: 1.41 (1.1–1.8)) and that they had

relapsed because they were more nervous/restless/depressed (OR: 1.43 (1.1–1.8)).

Conclusions. This study shows that smokers with low socioeconomic status have other motives to quitand other reasons to relapse than smokers with high socioeconomic status. Future tobacco prevention effortsaimed at smokers with low socioeconomic status should maybe focus on current advantages of quittingsmoking, using high cost of smoking and health advantages of quitting as motivating factors and by includingcomponents of mental health as relapse prevention.

© 2010 Elsevier Inc. All rights reserved.

Introduction

Many studies have examined smokers' motivation to quit and datastrongly suggest that health concern is the primary motive for quitattempts and that health professionals should continue emphasising thenegativehealth consequences of smoking tomotivate cessation attempts(McCaul et al., 2006). However, a recent publication found that there aresignificant differences in reported triggers for quit attempts as a functionof socio-demographic factors. Most notably, smokers with highersocioeconomic status (SES) were more likely to report concern aboutfuture healthwhereas those from lower SESweremore likely to cite costand current health problems (Vangeli and West, 2008).

The steep social class gradient in smoking has worsened over thepast 20 years (Millward et al., 2007). It has been suggested that themost important strategy to reduce health inequalities related to socialposition is to promote cessation in smokers with low social position(Gruer et al., 2009). Regrettably, the English report entitled Securinggood health for the whole population concluded, “We do not know

yggelse og Sundhed, Nordre-2600 Glostrup, Denmark. Fax:

l rights reserved.

what messages and interventions work to get lower socio economicgroups to stop smoking” (Wanless, 2004). The conclusion was thesame in the 2008-update of the U.S. Public Health Service ClinicalPractice Guideline, which underlined that new counselling strategiesshould be developed and research should focus on the development ofeffective interventions for populations that carry a disproportionateburden from tobacco e.g., individuals with limited educationalattainment/low SES (2008 Update Panel, L.a.S., 2008).

Before designing interventions and counselling strategies tosmokers with low SES, we need to know more about their motivesto quit, strategies used at previous quit attempts, experience withprevious quit attempts, and reasons to relapse. Only one previousstudy has examined triggers/motives to quit as a function of socio-demographic factors (Vangeli and West, 2008).

The aim of this article is, in a population-based sample of dailysmokers fromthe Inter99 study, to investigatemotives to quit, strategiesused at previous quit attempts, experienceswith previous quit attemptsand reasons to relapse as a function of socioeconomic status.

Materials and methods

Inter99 is a population-based intervention study initiated in March 1999and ended in April 2006. The study design is described in detail elsewhere(Jorgensen et al., 2003; The Inter99 Steering Committee, 2008).

Table 1Baseline characteristics of 2621 daily smokers with a previous quit attempt. The Inter99study (1999–2001), Denmark.

n Mean (SD) or percentageor median (quartiles)

Age (years, mean, (SD)) 2621 45.26 (8.4)Sex=man (%) 2621 49.9Age at smoking debut (years, mean, (SD)) 2605 16.75 (4.0)Ethnicity=Danish (%) 2611 95.3Tobacco consumption(cigarettes per day, mean, (SD))

2609 17.70 (8.1)

Number of quit attempts(median, (25, 50 and 75 percentiles))

2514 2.0 (1,2,4)

Wish to quit=very much (%) 2574 19.1Motivation to quit 2444

Preparation stage (%) 297 12.2Contemplation stage (%) 933 38.2Precontemplation stage (%) 1214 49.7

Self-rated health 2607Excellent/very good (%) 747 28.7Good (%) 1491 57.2Fair/poor (%) 369 14.2

Length of school education(years, mean, (SD))

2517 10.12 (2.2)

Length of vocational training orhigher education (years, mean, (SD))

2417 3.66 (1.8)

Employed=yes (%) 2596 83.5

SD=standard deviation.

49C. Pisinger et al. / Preventive Medicine 52 (2011) 48–52

The study was performed at the Research Centre for Prevention andHealth, Glostrup University Hospital, Denmark, and was approved by TheCopenhagen County Ethical Committee (KA 98155) and the National Board ofHealth. Written informed consent was obtained from all participants. Thestudy was registered in the Clinical Trials.gov (NCT00289237). The aim of thestudy was to prevent cardiovascular disease by non-pharmacological multi-factorial intervention.

The age- (30–60 years) and sex-stratified study population (N=61,301)from 11 municipalities in the suburbs southwest of Copenhagen was drawnfrom the Civil Registration System. From this study population, a randomsample comprising 13,016 individuals was drawn for the intervention. A totalof 6784 (52.5%) accepted and were included in the intervention. From theremaining 48,285 individuals in the study population (=control group), arandom sample of 5264 persons, equally distributed on age and sex, wasdrawn.

At baseline (years 1999–2001), 3684 persons in the intervention andcontrol groups stated to be daily smokers. Of these, 2621 (71.1%) stated thatthey had a previous quit attempt and are included in this paper.

Variable description

Smoking status: ‘Do you smoke?’ Only those answering ‘yes, daily’ wereincluded as smokers.

Socioeconomic status was defined by three variables, as there is no onesingle best variable and because we assume that results found to besignificant in all three indicators of SES are more valid than if we looked atone SES-indicator only. First, we used employment status, ‘Are you employed?’(yes/no). The unemployed group included 22 housewives but no retiredpeople. Second, we used length of school education (primary, secondary, uppersecondary, commercial upper secondary school). This variable was used aswedid not have information on diploma. We used the 25, 50 and 75 percentilesto define low, medium and high educational level (9 years or less, 10 years,11 years or more) respectively, to obtain groups of comparable size. The thirdvariable describes length of vocational training or higher education, afterfinishing basic school education (e.g., unskilled worker=0 years, greenkeeper assistant=1½ year, carpenter=3½ years, teacher=4 years, medicaldoctor=6 years). As the 25 and 50 percentiles were the same, we categorisedthis variable in low, medium and high educational level by comparable groupsize (1 year or less, 2–3 years, 4 years and more). This variable is found to besignificantly associated with chronic disease (Avlund et al., 2003) and to bestrongly correlated with factors known to be differently distributed by SES.

In general, education is relevant to people regardless of age or workingcircumstances. School education may capture the transition from parents' SESto adulthood SES, as own SES is influenced by access to and performance inschool which may be partly determined by intellectual and other resources ofthe family of origin. Also, it reflects health knowledge. Higher education/vocational training are strong determinants of future employment andincome and also reflect social standing and knowledge. Employment statusmay bemore transient, is associated with income but may also reflect disease,stress and autonomy.

Motives to quit: ‘Why do you want to quit?’ Each of the five motives (to behealthier, because of health related problems, because of concern for others(family/ friends), I am dissatisfied with being dependent of cigarettes, it is tooexpensive) had to be answered (yes/no). A smoker could give none, one orseveral motives to quit.

Two questions regarded experience with previous quit attempts. ‘Howwasthe support from family and friends when you tried to quit?’ The otherquestion was, ‘Was the previous quit attempt a good or bad experience?’ Theanswers were dichotomised to good and bad.

Strategies used at previous quit attempts: ‘When you tried to quit, have youthen…’ (…sought assistance from a doctor or other health professional,sought assistance from alternative medicine (e.g. acupuncture, hypnosis),used will power only, started to exercise, changed diet or alcohol habits)?Answer, yes/no.

Reasons to relapse: ‘What was the reason for your relapse?’ (weight gain, Iwas nervous/restless/depressed, I missed smoking at parties, I missedsmoking everyday, my family/friends were smoking, something importanthappened (death, illness in the family, problems, unemployment). Answer,yes/no.

Motivation to quit: smokers were categorised in motivational stages usinga simplified typology of Prochaska and Di Clemente (DiClemente et al., 1991;Prochaska and DiClemente, 1983). (Preparation stage: ‘planning to quit

within onemonth.’ Contemplation stage: ‘planning to quit within the next sixmonths.’ Precontemplation stage: ‘not planning to quit.’)

Statistical analyses

All data processing was done with the SPSS 15.0 software (SPSS Inc.,Chicago, IL, USA). Categorical data were tested by Pearson's chi-square test.Continuous data and ordinal data were tested by Descriptives. Explore, thestatistics descriptive in SPSS, was used to describemedian and inter-quartiles.

Logistic regression analyses were used to look at previous quit attempts,motives to quit, strategies used at previous quit attempts and reasons torelapse by SES. Dependent variable was each of the dichotomised answers inthe above described questions. Covariates were 1) employment status(smokers with employment=reference), 2) length of school education(smokers with high education=reference) and 3) length of vocationaltraining or higher education (smokers with high education=reference). Eachvariable was first analysed unadjusted. Then, we adjusted for sex and age.Only adjusted results are shown as they are very close to unadjusted results.Furthermore, when analysing reasons to relapse, we included tobaccoconsumption as additional confounder. The results did not change fromresults adjusted for sex and age only and are not shown.

Level of significance was set to 5%. The models were controlled by theHosmer–Lemeshow goodness-of-fit-test.

Results

The daily smokers included were middle-aged, equally distributedby sex, and mostly Danish of origin (Table 1). They were aboutseventeen years at smoking debut, smoked almost a pack of cigarettesper day and had a couple of previous quit attempts. One fifth had ahigh wish to quit. They had about ten years of school education andfour years of higher education/vocational training and most wereemployed.

The most frequent motives to quit were to be healthier and costand dissatisfaction with being dependent (Fig. 1). The most frequentreasons to relapse were missed smoking at parties/on everyday andmy family/friends were smoking.

In adjusted logistic regression analyses, we saw four results beingconsistently significant across all three SES-indicators: smokers withlow SES were more likely to report that they wanted to quit becausesmoking was too expensive and because they had health relatedproblems, that previous quit attempt was a bad experience, and that

72.6

62.458.2

37.6

89.8

0

10

20

30

40

50

60

70

80

90

100

To be healthier Health relatedproblems

Concern forothers (family/

friends)

Dissatisfied withbeing dependent

It is tooexpensive

33

58.6

7474.3

49.440.8

0

10

20

30

40

50

60

70

80

90

100

Weight gain Nervous/restless/depressed

Missed smokingat parties

Missed smokingon everyday

Myfamily/friendswere smoking

Somethingimportanthappened

%

% Motives to quit

Reasons to relapse

Fig. 1. Frequencies of motives to quit and reasons to relapse in 2621 daily smokers* from a general population in the Inter99 study (1999–2001), Denmark. * Only smokers with aprevious quit attempt are included. A smoker could answer yes to none, one or several motives to quit and reasons to relapse.

50 C. Pisinger et al. / Preventive Medicine 52 (2011) 48–52

they relapsed because they were more nervous/restless/depressed,compared with smokers with high SES (Table 2). Furthermore, whenwe defined low SES by including two SES-indicators only, we saw thatlow SES smokers reported significantly worse support from familyand friends at previous quit attempts and concern for family andfriends was significantly less likely a motive to quit than for smokerswith high SES.

When looking at low SES, as defined by one SES-variable only, wefound that unemployed smokers were significantly more likely torelapse because they missed smoking everyday or because somethingimportant happened (e.g. unemployment), compared with employedsmokers. When looking at smokers with low school education, theywere significantly more likely to relapse because their family/friendswere smoking or because of weight gain, and it was significantly lesslikely that they had exercised when they tried to quit, compared withsmokers with high education. When looking at smokers with lowhigher education, it was significantly less likely that their motive toquit was that they wanted to be healthier or because they weredissatisfied with being dependent of cigarettes, compared withsmokers with high education.

In addition, smokers with medium school education weresignificantly more likely to seek assistance to quit, and significantlyless likely to exercise or use will power only at quit attempts,compared with smokers with high education.

Discussion

In a large population-based study, we found that smokers with lowSESwere significantlymore likely than smokers with high SES to reportthat theywanted to quit because smokingwas too expensive or becausethey had health related problems. When looking at previous quitattempts, smokerswith low SESwere significantlymore likely to report

that it had been a bad experience, and that they had relapsed becausethey were more nervous/restless/depressed than smokers with highSES. These findings were consistent across three indicators of SES.Furthermore, significantly worse support from family and friends atprevious quit attempts seems an obstacle for smokers with low SES.

Vangeli andWest (2008) were the first to look at triggers to quit asa function of socio-demographic factors. They found that smokerswith lower SES were more likely to report cost and current healthproblems as triggers compared with smokers with high SES whoreported future health concerns as the most important trigger. This isin concordance with our results but we did not have information ontriggers to the most recent quit attempt. Smokers with low SES havebeen found to live more in the present and think less about future(Wardle and Steptoe, 2003), and it has been documented that regretof smoking is higher in those who perceive higher monetary costs ofsmoking (Fong et al., 2004).

The different pattern of smoking cessation according to level ofeducation suggests that health promotion actions and tobacco controlpolicies might have had a different effect among different populationsubgroups (Schiaffino et al., 2007), and a recent review concluded thatthere is considerable evidence that media campaigns to promotesmoking cessation are often less effective among socioeconomicallydisadvantaged populations (Niederdeppe et al., 2008). Till now,mediacampaigns have focused on warning of potential risk of disease e.g.,lung cancer, which have appealed to smokers with high SES. Futurecampaigns aimed at smokers with low SES should maybe focus oncurrent advantages of quitting smoking, as better economy, lesscough/cardiac symptoms/pain etc. (Pisinger et al., 2010).

Smokers with low SESmore frequently experienced bad support atprevious quit attempts and had a bad experience when they tried toquit, which is in concordance with previous studies (Reime et al.,2006; Sorensen et al., 2002). We have not been able to find previous

Table 2Previous quit attempts, reasons to quit and reasons to start smoking again by socioeconomic status. Adjusted logistic regression analyses including 2621 daily smokers with aprevious quit attempt. The Inter99 study (1999–2001), Denmark.

Employment status(reference: employed,OR=1)

School education(reference: high, 11 years or more, OR=1)

Vocational training or higher education(reference: high, 4 years or more, OR=1)

Unemployed Low9 years or less

Medium10 years

Low1 year or less

Medium2–3 years

n OR (95% CI) n OR (95% CI) OR (95% CI) n OR (95% CI) OR (95% CI)

Motives to quitTo be healthier=yes 2421 0.96 (0.7–1.4) 2354 0.92 (0.6–1.3) 0.97 (0.7–1.4) 2257 0.66 (0.5–0.9)⁎ 0.84 (0.6–1.2)Health related problems=yes 2204 2.24 (1.8–2.8)⁎⁎ 2141 1.75 (1.4–2.2)⁎⁎ 1.16 (0.9–1.5) 2059 1.38 (1.1–1.7)⁎ 0.96 (0.8–1.2)Concern for others (family/friends)=yes 2212 0.86 (0.7–1.1) 2150 0.76 (0.6–0.9)⁎ 0.82 (0.7–1.0) 2083 0.78 (0.6–1.0)⁎ 0.93 (0.7–1.2)Dissatisfied with being dependent=yes 2179 1.05 (0.8–1.3) 2122 0.81 (0.6–1.0) 1.10 (0.9–1.4) 2049 0.75 (0.6–0.9)⁎ 0.90 (0.7–1.1)It is too expensive=yes 2290 1.40 (1.1–1.8)⁎ 2216 1.85 (1.4–2.4)⁎⁎ 1.51 (1.2–1.9)⁎⁎ 2141 1.56 (1.2–2.0)⁎⁎ 1.33 (1.0–1.7)⁎

Experience with previous quit attempts#

Support from family and friends=bad 1619 1.66 (1.2–2.3)⁎⁎ 1588 1.08 (0.8–1.5) 1.20 (0.9–1.6) 1543 1.71 (1.3–2.3)⁎⁎ 0.94 (0.7–1.3)Experience of previous quit attempts=bad 1585 1.47 (1.1–2.0)⁎⁎ 1553 1.41 (1.1–1.8)⁎ 1.26 (1.0–1.6) 1507 1.53 (1.2–2.0)⁎⁎ 1.09 (0.8–1.4)

Strategies used at previous quit attemptsSought assistance from a doctor or otherhealth professional=yes

2561 1.58 (1.1–2.2)⁎ 2488 1.00 (0.6–1.5) 1.53 (1.1–2.2)⁎ 2391 0.85 (0.6–1.3) 0.76 (0.5–1.1)

Sought assistance from alternative medicine(e.g., acupuncture, hypnosis)=yes

2539 0.80 (0.6–1.1) 2468 1.23 (0.9–1.7) 1.70 (1.3–2.3)⁎⁎ 2377 0.96 (0.7–1.3) 1.22 (0.9–1.6)

Used will power only=yes 2509 1.02 (0.8–1.3) 2437 0.80 (0.6–1.1) 0.74 (0.6–1.0)⁎ 2345 0.95 (0.7–1.2) 1.02 (0.8–1.3)Started to exercise=yes 2457 1.02 (0.7–1.4) 2384 0.66 (0.5–0.9)⁎ 0.73 (0.5–1.0)⁎ 2293 0.76 (0.5–1.1) 0.82 (0.6–1.1)Changed diet or alcohol habits=yes 2452 1.37 (1.0–1.9) 2386 0.78 (0.6–1.1) 0.75 (0.6–1.0) 2298 0.78 (0.6–1.1) 0.98 (0.7–1.3)

Reasons to relapseWeight gain=yes 1825 0.89 (0.7–1.2) 1779 1.92 (1.5–2.5)⁎⁎ 1.35 (1.1–1.7)⁎ 1730 1.06 (0.8–1.4) 0.99 (0.8–1.3)Nervous/restless/depressed=yes 1809 2.70 (2.0–3.6)⁎⁎ 1764 1.43 (1.1–1.8)⁎⁎ 1.10 (0.9–1.4) 1716 1.47 (1.1–1.9)⁎ 1.03 (0.8–1.3)Missed smoking at parties=yes 1818 0.89 (0.7–1.2) 1814 1.00 (0.8–1.3) 0.90 (0.7–1.2) 1763 0.82 (0.6–1.1) 0.90 (0.7–1.2)Missed smoking everyday=yes 1941 1.56 (1.1–2.1)⁎⁎ 1888 1.18 (0.9–1.6) 1.15 (0.9–1.5) 1830 1.12 (0.8–1.5) 0.95 (0.7–1.2)My family/friends were smoking=yes 1768 1.33 (1.0–1.8) 1720 1.82 (1.4–2.4)⁎⁎ 1.05 (0.8–1.4) 1620 1.26 (1.0–1.6) 0.95 (0.8–1.2)Something important happened (death, illness inthe family, problems, unemployment)=yes

1784 1.99 (1.5–2.6)⁎⁎ 1740 0.91 (0.7–1.2) 1.11 (0.9–1.5) 1692 0.94 (0.7–1.2) 1.24 (1.0–1.6)

Adjusted for age and sex at baseline.⁎ and bold: pb0.05.⁎⁎ and bold: pb0.01.# ‘Experience with previous quit attempts,’ these questions were not included in the questionnaire for participants in the control group. There were many missing answers

regarding ‘Reasons to relapse.’

51C. Pisinger et al. / Preventive Medicine 52 (2011) 48–52

studies looking at reasons to relapse, as a function of SES. In our study,the most important reason to relapse in smokers with low SES wasthat they were more nervous/depressed/restless when they tried toquit. A recent large study reported that smokers with low SES had thehighest odds of reporting psychological distress (Dube et al., 2009).Difficult living conditions of persons with low SES (as reflected byeconomic strain, insecure employment and low-control workingcontext), a pro-smoking social context, isolation from wider socialnorms and limited opportunities for respite and recreation appear toundermine cessation (Fagan et al., 2007; Paul et al., 2010; Stead et al.,2001).

Odds of relapse were higher in low SES smokers than in high SESsmokers, when family and friends were smoking, but this was onlysignificant for school educational levels. Relapse due to weight gainwas twice as likely in persons with low school education, comparedwith those with high education. The reason for this may be thatsmokers with low school education have the lowest knowledge ofprevention of weight gain.

The strengths of this study are the large size, the random sample ina general population and the inclusion of smokers in all stages ofmotivation to quit and all educational levels. This should give a highgeneralizability. Furthermore, we used three different indicators ofSES. Results found to be significant in all three indicators of SES areprobably more valid than if we had looked at one SES-indicator only.

The major weakness of this study is the low participation rate, ageneral problem in population-based studies. However, only 6% of thenon-participants in the intervention group stated that smoking wasthe reason why they did not participate. We cannot rule out thepossibility that self-report and recall bias may have affected results.Data were collected between 1999 and 2001 and attitudes to smoking

and motives to quit may have changed over time. Furthermore, theanswer categories were constructed by researchers, and importantmotives/reasons may have been missed. Missing data are always areason of concern.

Measuring SES is very complex and there is no single best indicator(Galobardes et al., 2006a,b). We used three different variables todefine SES from different angles. All three variables are correlated.Measures of income and/or occupation would have been valuablesupplements.

Conclusion

This study shows that smokers with low SES have other motives toquit and other reasons to relapse than smokers with high SES. Tillnow, we have focused on warning of risk of disease but futureantismoking-campaigns aimed at smokers with low SES shouldmaybe instead focus on current advantages of quitting smoking,using high price of cigarettes and health advantages of quitting asmotivating factors. Also, future tobacco prevention efforts aimed atsmokers with low SES may benefit by more intensive support and byincluding components of mental health. Smokers with low SES havethe highest distress and this study found that the most importantreason to relapse was that they were nervous/depressed/restlesswhen they tried to quit and that they received bad support at quitattempts.

Funding

This work was supported by Tryg Foundation, Helse Foundation,Danish Medical Research Council, The Danish Centre for Evaluation

52 C. Pisinger et al. / Preventive Medicine 52 (2011) 48–52

and Health Technology Assessment, Novo Nordisk, CopenhagenCounty, Danish Heart Foundation, The Danish Pharmaceutical Asso-ciation, Augustinus Foundation, Becket Foundation and Ib HenriksensFoundation. The researchers are all independent of the founders.

Contributors

CP and TJ contributed to the design of the study, collection andassembly of the data, analysis and interpretation of data and drafting ofthe article. UT and MA and took part in the interpretation of data anddrafting of the article. All authors approved the final manuscript. CP istheguarantor. Theguarantor accepts full responsibility for the conduct ofthe study, had access to the data and controlled the decision to publish.

Conflict of interest statementThere was no conflict of interest.

Acknowledgments

We thank the whole Inter99-staff and all persons participating inthe study.

The Steering Committee of the Inter99 study includes thefollowing: D.M.Sci. Torben Jorgensen (principal investigator), D.M.Sci. Knut Borch-Johnsen (principal investigator on the diabetes part)and Ph.D. MPH Charlotta Pisinger.

References

2008 Update Panel, Liaisons, and Staff, 2008. A clinical practice guideline for treatingtobacco use and dependence: 2008 update. A U.S. Public Health Service report. Am.J. Prev. Med. 35, 158–176.

Avlund, K., Holstein, B.E., Osler, M., Damsgaard, M.T., Holm-Pedersen, P., Rasmussen, N.K.,2003. Social position and health in old age: the relevance of different indicators ofsocial position. Scand. J. Public Health 31, 126–136.

DiClemente, C.C., Prochaska, J.O., Fairhurst, S.K., Velicer, W.F., Velasquez, M.M., Rossi, J.S.,1991. The process of smoking cessation: an analysis of precontemplation, contem-plation, and preparation stages of change. J. Consult. Clin. Psychol. 59, 295–304.

Dube, S.R., Caraballo, R.S., Dhingra, S.S., Pearson, W.S., McClave, A.K., Strine, T.W., Berry,J.T., Mokdad, A.H., 2009. The relationship between smoking status and seriouspsychological distress: findings from the 2007 Behavioral Risk Factor SurveillanceSystem. Int. J. Public Health 54 (Suppl 1), 68–74.

Fagan, P., Moolchan, E.T., Lawrence, D., Fernander, A., Ponder, P.K., 2007. Identifyinghealth disparities across the tobacco continuum. Addiction 102 (Suppl 2), 5–29.

Fong, G.T., Hammond, D., Laux, F.L., Zanna, M.P., Cummings, K.M., Borland, R., Ross, H.,2004. The near-universal experience of regret among smokers in four countries:

findings from the International Tobacco Control Policy Evaluation Survey. NicotineTob. Res. 6 (Suppl 3), S341–S351.

Galobardes, B., Shaw, M., Lawlor, D.A., Lynch, J.W., Davey, S.G., 2006a. Indicators ofsocioeconomic position (part 1). J. Epidemiol. Community Health 60, 7–12.

Galobardes, B., Shaw, M., Lawlor, D.A., Lynch, J.W., Davey, S.G., 2006b. Indicators ofsocioeconomic position (part 2). J. Epidemiol. Community Health 60, 95–101.

Gruer, L., Hart, C.L., Gordon, D.S., Watt, G.C., 2009. Effect of tobacco smoking onsurvival of men and women by social position: a 28 year cohort study. BMJ338, b480.

Jorgensen, T., Borch-Johnsen, K., Thomsen, T.F., Ibsen, H., Glumer, C., Pisinger, C., 2003. Arandomized non-pharmacological intervention study for prevention of ischaemicheart disease: baseline results Inter99. Eur. J. Cardiovasc. Prev. Rehabil. 10,377–386.

McCaul, K.D., Hockemeyer, J.R., Johnson, R.J., Zetocha, K., Quinlan, K., Glasgow, R.E.,2006. Motivation to quit using cigarettes: a review. Addict. Behav. 31, 42–56Report.

Millward, M., Wohlgemuth, C., Taske, N., McLean, C., Naidoo, B., Taylor, L., Warm, D.,2007. Smoking and public health: a compendium of smoking behaviour initiativesthat address socially disadvantaged populations. Evidence review. NationalInstitute for Health and Clinical Excellence.

Niederdeppe, J., Kuang, X., Crock, B., Skelton, A., 2008. Media campaigns to promotesmoking cessation among socioeconomically disadvantaged populations: what dowe know, what do we need to learn, and what should we do now? Soc. Sci. Med. 67,1343–1355.

Paul, C.L., Ross, S., Bryant, J., Hill, W., Bonevski, B., Keevy, N., 2010. The social context ofsmoking: a qualitative study comparing smokers of high versus low socioeconomicposition. BMC Public Health 10, 211.

Pisinger, C., Aadahl, M., Toft, U., Jørgensen, T. The association between active andpassive smoking and frequent pain in a general population, 2010. European Journalof Pain. Ref Type: Accepted for publication.

Prochaska, J.O., DiClemente, C.C., 1983. Stages and processes of self-change ofsmoking: toward an integrative model of change. J. Consult. Clin. Psychol. 51,390–395.

Reime, B., Ratner, P.A., Seidenstucker, S., Janssen, P.A., Novak, P., 2006. Motives forsmoking cessation are associated with stage of readiness to quit smoking andsociodemographics among German industrial employees. Am. J. Health Promot. 20,259–266.

Schiaffino, A., Fernandez, E., Kunst, A., Borrell, C., Garcia, M., Borras, J.M., Mackenbach, J.P.,2007. Time trends and educational differences in the incidence of quitting smoking inSpain (1965–2000). Prev. Med. 45 (2–3), 226–232.

Sorensen, G., Emmons, K., Stoddard, A.M., Linnan, L., Avrunin, J., 2002. Do socialinfluences contribute to occupational differences in quitting smoking and attitudestoward quitting? Am. J. Health Promot. 16, 135–141.

Stead, M., MacAskill, S., MacKintosh, A.M., Reece, J., Eadie, D., 2001. “It's as if you'relocked in”: qualitative explanations for area effects on smoking in disadvantagedcommunities. Health Place 7 (4), 333–343.

The Inter99 Steering Committee, 2008. Homepage of the Inter99 study. www.Inter99.dk. Ref Type: Computer Program.

Vangeli, E., West, R., 2008. Sociodemographic differences in triggers to quit smoking:findings from a national survey. Tob. Control 17, 410–415 Report.

Wanless, D., 2004. Securing good health for the whole population: final report. In:Treasury, H.M. (Ed.), Crown.

Wardle, J., Steptoe, A., 2003. Socioeconomic differences in attitudes and beliefs abouthealthy lifestyles. J. Epidemiol. Community Health 57, 440–443.