Background The US Food and Drug Administration has established a policy of substantially discounting the health benefits of reduced smoking in its evaluation of proposed regulations because of the cost to smokers of the supposed lost pleasure they suffer by no longer smoking. This study used data from nine countries of the former Soviet Union (fSU) to explore this association in a setting characterised by high rates of (male) smoking and smoking-related mortality.
Methods Data came from a cross-sectional population-based study undertaken in 2010/2011 in Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Russia and Ukraine. Information was collected from 18 000 respondents aged ≥18 on smoking status (never, ex-smoking and current smoking), cessation attempts and nicotine dependence. The association between these variables and self-reported happiness was examined using ordered probit regression analysis.
Results In a pooled country analysis, never smokers and ex-smokers were both significantly happier than current smokers. Smokers with higher levels of nicotine dependence were significantly less happy than those with a low level of dependence.
Conclusions This study contradicts the idea that smoking is associated with greater happiness. Moreover, of relevance for policy in the fSU countries, given the lack of public knowledge about the detrimental effects of smoking on health but widespread desire to quit reported in recent research, the finding that smoking is associated with lower levels of happiness should be incorporated in future public health efforts to help encourage smokers to quit by highlighting that smoking cessation may result in better physical and emotional health.
- nicotine dependence
- consumer surplus
Statistics from Altmetric.com
Since the seminal research of the 1950s and the subsequent reports of the Royal College of Physicians in the UK in 1962 and Surgeon General in the USA in 1964 stating that smoking causes lung cancer,1 the effects of smoking on physical health have been documented extensively. Smoking has been linked to cardiovascular, metabolic and respiratory diseases, various types of cancer and adverse perinatal outcomes2 with tobacco use having been estimated to have caused approximately 100 million deaths in the 20th century.3
There is, however, a question of how smoking affects psychological well-being. This has recently acquired much greater importance since the US Food and Drug Administration began to introduce the concept of ‘consumer surplus’ into its economic evaluations of tobacco control measures.4 In essence, this argues that, as smokers are willing to pay more than the actual price for cigarettes, as utility maximising individuals, this implies that they place a value on the assumed pleasure that they obtain from smoking. This argument is supported by reference to observations that show nicotine increases positive mood in healthy as well as in psychiatrically ill individuals,5 and that smokers frequently cite ‘enjoyment’ as a motive for smoking.6 Yet while the theoretical underpinning of the rational choice theory on which the argument of consumer surplus is based has been severely challenged, given factors such as the addictive nature of nicotine,7 the decision by the Food and Drug Administration to invoke it, while providing little empirical justification for doing so, does raise the key question of whether smoking is, indeed, associated with improved mood, especially as discounting the health benefits of reduced smoking may have potentially important consequences for reducing smoking and its effects.
In practice, much of the existing research on tobacco use and psychological health has explored its detrimental effects, with studies showing that smoking may precede depression and anxiety disorder or vice versa.8–10 The relationship between tobacco use and negative affect has been described as complex,11 with the pharmacological effect of nicotine in cigarettes possibly having a central role in this association. Specifically, imbibed nicotine binds to nicotine cholinergic receptors in the brain and can affect neurotransmitter activation.12 Through the effects of dopamine, this can produce feelings of pleasure, and alleviate stress and anxiety,12 which possibly helps explain why stress relief is often cited as one of the main reasons for smoking.6 However, neuroadaptation, receptor desensitisation and subsequent dependence in the presence of ongoing nicotine exposure mean that its absence as a result of smoking cessation can also result in withdrawal symptoms such as depressed mood and anxiety,12 especially among those with higher levels of usage. A complementary body of research has examined regret for starting smoking, evaluating responses to the question “If you had to do it over again, you would not have started smoking?” This has found that about 90% of smokers in high-income countries express regret, with the highest levels among women, older smokers, smokers who have made multiple quit attempts, individuals with higher levels of perceived addiction and those worried about subsequent damage to their health.13 High levels are also seen in middle-income countries, but these are influenced by the overall social acceptability of smoking, itself a function of tobacco control policies, in the country concerned.14
There have been comparatively few studies on how smoking impacts on aspects of positive affect such as happiness.15 As positive and negative affect are not simply end points on a continuum but, rather, as Shahab and West15 have highlighted, distinct dimensions, it is possible that they may have potentially variable relations to smoking. Moreover, tobacco industry documents show that the idea of happiness has been used in the promotion and marketing of cigarettes,16 while cigarette companies when designing brands to appeal to consumers with specific psychological and psychosocial needs found that unhappiness was linked to a greater value being attached to the ‘salience of the delivery level’.17
The limited research undertaken to date on smoking and happiness has produced conflicting results. While one study at the regional level has revealed that higher levels of smoking are linked to lower levels of happiness,18 research on different study populations in various locations has produced more variable findings. Studies have shown that smoking is associated with increased happiness,19 or that there is no relationship20 or that ex-smokers are happier than current smokers21 but that this relationship is conditional on the time since quitting.15 ,22 Further, in terms of tobacco intake or its pharmacological effect, although some research has indicated that the number of cigarettes consumed among smokers is not related to happiness,21 other studies have indicated that an increase in daily smoking frequency is linked to reduced happiness.23 Smokers who attempt to quit have also been found to be unhappier than those who have never attempted to quit.21 However, the situation is complicated by the observation that heavy smokers demonstrate optimism bias and, compared to non-smokers, markedly underestimate the impact of smoking on mortality.24 ,25
The current study examined the association between smoking and happiness in nine countries in the former Soviet Union (fSU). This is a particularly apposite site to examine this relationship. Life satisfaction and happiness ratings are especially low in Eastern Europe and fSU countries such as Russia,26 ,27 while tobacco has a large impact on the disease burden in Eastern Europe,28 with smoking-related mortality being especially high in the fSU countries.29 Although there is evidence that the prevalence of smoking has stabilised in the countries in this region in recent years, it nevertheless remains very high among men,30 and understanding of the effects of tobacco use on health is still limited among the general public,31 as manifest, possibly, in low quit ratios.32 On the other hand, there is a high desire to stop smoking among individuals in these countries.32 In such circumstances, determining how smoking affects mood (happiness) may provide a motivating factor for cessation.15 Against this backdrop, the specific aims of the current study were to: (1) determine the association between smoking status and happiness; (2) examine the association between attempting to quit smoking and happiness; and (3) explore whether nicotine dependence is linked to happiness.
The data used in this study came from the Health in Times of Transition (HITT) survey. This cross-sectional survey was undertaken in nine countries: Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Moldova, Russia and Ukraine in 2010, and in Kyrgyzstan in early 2011 (due to political violence). Multistage random sampling was employed in each country to obtain household samples that were nationally representative. Random route procedures were used to select households from within primary sampling units (approximately 100–200 per country). One person aged 18 or above was randomly chosen to participate (determined by the nearest birthday) from within each household. Face-to-face interviews were conducted in the respondents’ homes by trained interviewers using a standard questionnaire. All respondents had the choice of answering in either their country's language or Russian except for in Russia and Belarus, where Russian was used.
Information was collected from 18 000 respondents in total. The sample size was 1800 respondents in six of the nine countries. Russia (n=3000) and Ukraine (n=2000) had larger sample sizes to reflect their larger and more regionally diverse populations. Georgia also had a larger sample size (n=2200). This was the result of a booster survey of 400 additional interviews that was undertaken in late 2010 to ensure that the sample was more representative. Response rates varied from 47% in Kazakhstan to 83% in Georgia. More details of the survey can be obtained from the survey's website (http://www.hitt-cis.net).
To assess happiness, respondents were asked “Taking all things together, how would you say things are these days—would you say you are?” and then presented with a single-item 10-point scale that ranged from ‘very unhappy’ (scored 1) to ‘very happy’ (scored 10). This single-item measure has been widely employed in subjective well-being research and has been found to be valid and reliable.33 Smoking status was determined using two questions: ‘Do you smoke at least one cigarette per day (1 papirossi, 1 pipe, cigar etc)?’ (with those who answered yes being categorised as current smokers), and ‘Have you ever smoked?’ with answer options ‘Yes, I smoked but stopped’ (ex-smokers) and ‘No, I don't smoke and I have never smoked’ (never smokers). Attempting to stop smoking (cessation attempts) was assessed by asking smokers, ‘How many times have you tried to give up smoking?’ with response options ‘never’, ‘once’, ‘twice’ and ‘more than twice’. Nicotine dependence was assessed using the Fagerström Test for Nicotine Dependence,34 which is comprised of 6 questions, the answers to which when summed form a score ranging from 0–10 with cut points of 0–2 (very low), 3–4 (low), 5 (medium), 6–7 (high) and 8–10 (very high).35
Eight factors were controlled for in the analyses as various demographic and socioeconomic factors have been reported to be linked to happiness across countries26 and may also affect smoking. Age in years was divided into five categories: 18–29, 30–39, 40–49, 50–59 and 60 and above. Three marital status categories were created: married/cohabiting, never married and divorced/widowed. Educational attainment was also divided into three categories: high (complete/incomplete higher education), middle (complete secondary/secondary special education) and low (incomplete secondary education or lower). Wealth was assessed using a household asset index that consisted of 10 items (fridge, TV, computer, etc). Principal component analysis was used to generate wealth tertiles (high, average and low). Self-rated health was categorised as either very good/good/fair or poor/very poor. The residential location of respondents was classified as either urban or rural. In addition, as recent research has suggested that alcohol consumption is related to lower happiness in these countries,36 and that heavy drinking and smoking are associated in this region,37 we also controlled for heavy episodic drinking. Following the definition employed in several recent studies in these countries, this was defined as consumption of at least one of the following on one occasion: ≥2 L of beer, ≥750 g of wine, or ≥200 g of strong spirits.38 ,39
The prevalence of the smoking status categories and mean (SD) happiness scores by baseline characteristics were calculated (table 1). Next, multivariable ordered probit regression analysis was used to examine the relationship between the different smoking status categories and happiness in each individual country, and in a pooled analysis of all countries while controlling for age, sex, marital status, education, wealth, self-rated health, location, heavy episodic drinking and individual country effects (table 2). Multivariable ordered probit regression analysis was also used to test the association between happiness and (1) smoking cessation attempts; (2) nicotine dependence among current smokers (table 3). These models were also adjusted for age, sex, marital status, education, wealth, self-rated health, location and heavy episodic drinking. For this analysis, as the restriction only to those who currently smoke resulted in a small sample size in some countries, countries’ data were combined in order to increase the statistical power. Pooled country analyses were all adjusted for country. All regression analyses were performed using ordered probit regression analysis, which is theoretically more efficient than ordinary least squares (OLS) for analysing ordered outcomes40 and has been extensively used in happiness research. The β-coefficients and SEs are reported in the tables. In order to facilitate understanding of these estimates, we also report the results (OR and 95% CIs) obtained from ordinal logistic regression analyses, which are presented in an online appendix (see online supplementary tables A1 and A2). The statistical analysis was carried out with Stata V.12.1 (Stata Corp LP, College Station, Texas, USA). The level of statistical significance was set at p<0.05.
The characteristics of the study sample are illustrated in table 1. The prevalence of current smoking ranged from 19.9% in Moldova to 31.4% in Russia. The corresponding figures among males and females were 48.5% and 8.5%, respectively. Heavy episodic drinkers were much more likely to be current smokers than those who did not engage in this behaviour (64.4% vs 20.8%). The prevalence of ex-smokers was lowest in Azerbaijan (2.6%) and highest in Belarus (15.7%). The overall mean (SD) happiness score was 6.4 (2.0).
In the pooled country analysis, not smoking was associated with greater happiness. Specifically, never smokers (β=0.0579; OR=1.12, CI=1.04 to 1.20 (see online supplementary table A1)) and ex-smokers (β=0.0973) (OR=1.17, CI=1.06 to 1.29 (see online supplementary table A1)) were both significantly happier than current smokers (table 2). In the individual countries, in every country except Kyrgyzstan, ex-smokers were happier than current smokers with this difference being statistically significant in Azerbaijan and Ukraine, while never smokers were happier in every country with the difference reaching statistical significance in Kyrgyzstan and Ukraine (table 2). For the control variables, in the pooled country analysis, being older, male, never married or divorced/widowed, having lower levels of education, wealth, poor self-rated health, living in an urban area and engaging in heavy episodic drinking were all associated with a significantly lower level of happiness.
There was no difference in terms of happiness between those who had never attempted to quit smoking and those who had attempted to quit one or more times (table 3). However, for nicotine dependence, those with higher levels of dependence (a score of 6 or above) were significantly less happy than those smokers in the reference category with a low level of dependence (0–2), that is, for scores of 6–7, β=−0.1172 (OR 0.81, CI 0.69 to 0.95 (online supplementary table A2)) and for 8–10, β=−0.2038 (OR 0.70, CI 0.55 to 0.89 (online supplementary table A2)).
This study examined the relationship between smoking status and happiness in nine fSU countries. In a pooled country analysis that controlled for a variety of potential confounders, we found that ex-smokers and never smokers were significantly happier than current smokers. Within the individual countries, ex-smokers were happier than current smokers in all countries except Kyrgyzstan, with this difference being statistically significant in Azerbaijan and Ukraine, while never smokers were happier than current smokers in every country with the effect being especially strong in Kyrgyzstan and Ukraine. Smokers with higher nicotine dependence scores were significantly less happy when compared with smokers who reported very low levels of nicotine dependence. Attempting to quit smoking was not associated with happiness among current smokers.
The finding that in the fSU countries ex-smokers were happier than current smokers accords with results from a recent study in Hong Kong, but unlike that study, we found that never smokers were also significantly happier than current smokers.21 Our results also support the findings from an earlier study, which showed that the mean happiness scores of smokers are lower than non-smokers across the course of the day.41 This latter study suggested that what has been termed the ‘nicotine deprivation reversal model’ might explain the poorer mood of smokers, that is, that the pharmacological effects of nicotine dependency are manifest in a worse mood state in-between cigarettes when this chemical is absent, and that the repeated experience of nicotine absence across the course of the day results in smokers’ mood state being poorer overall.42 ,43
Alternatively, individual benefits that have been linked to stopping smoking, such as better health, improved self-confidence and saving money44 might also explain why ex-smokers feel happier. In addition, it is possible that stress might also have been important for this relationship. For example, there is some evidence that while quitters experience a decrease in the number of stressors experienced over time the opposite is true for smokers,45 while other research has pointed to smokers being more reactive to stress, that is, ‘easily upset, nervous, sensitive, and prone to worry and guilt’.46 Given that happiness and stress seem to be inversely related47 and that an upsurge in societal stress has been noted in the post-Soviet transition period,48 it is possible that smokers in these countries not only experience more stress but are unhappier as a result of it.
Our results also indicated that among smokers, higher levels of nicotine dependence are associated with lower happiness. This seemingly contradicts the finding that the number of cigarettes smoked per day is not related to happiness,21 but mirrors the result from a study where an increase in the frequency of daily smoking was linked to lower happiness.23 It also accorded with the finding from a study where nicotine dependence was negatively associated with ‘life enjoyment’.15 It is possible that this association might stem from the stronger pharmacological effects experienced by heavy smokers. Specifically, it has been claimed that the amount of distress experienced by smokers during periods of abstinence is related to the volume of smoking, with heavier smokers and those who are more dependent on nicotine having larger mood diminutions.42 It can also be hypothesised that those who are most dependent on nicotine might be more vulnerable to the negative effects of smoking more generally (in terms of health and cost, etc), and so may be more prone to experience negative mood states. However, regardless of the specific mechanism(s) underpinning this finding, it does seem to accord with other recent research from the fSU countries, which has linked a high level of nicotine dependence with higher levels of psychological distress49 and suggests that smokers with a high level of nicotine dependence might be especially vulnerable to poorer psychological health in the countries in this region. Given the move by tobacco corporations into alternative nicotine delivery devices, such as electronic cigarettes, this potential risk must be considered along with the growing evidence of the harms associated with nicotine.50
Although previous research has produced differing results concerning the relationship between smoking cessation attempts and happiness,15 ,21 the finding from the current study that attempting to quit smoking on one or more occasions was not related to happiness does, at first sight, seem to contradict the results from several earlier studies that have highlighted how unsuccessful quitting is associated with both psychological distress and depression.51 ,52 It is possible, however, that our result might stem from several factors. Some research suggests, for example, that people who quit and then relapse may change their beliefs about the impact of smoking so that it is seen as providing more psychological benefits and as being less dangerous for health,53 which might also protect against the negative emotional consequences of failure. It can also be speculated that, as previous research has shown, there are high levels of tobacco use30 and large gaps in knowledge about the impact of smoking on health31 in these countries, so failure to quit might not have the same impact as where smoking is less prevalent and its detrimental effects are better understood. Finally, we had no information on when or why quit attempts were made. Such information might have been important when examining this association.
This study had several limitations. The data on smoking were self-reported and there was no validation of this with biomarkers such as serum cotinine. However, previous research has indicated that the validity of self-reported smoking data is high54 and that data collected by interviewers have a higher sensitivity and specificity than those obtained from self-administered questionnaires.55 Second, doubts have been raised about whether the phenomenon of happiness is understood and expressed in the same way across different cultures and if current research methodology is adequate given such differences.56 Other research has suggested, however, that the concept of happiness is comparable across nations,57 with empirical research supporting this idea by showing that results can be generalised across countries.26 Finally, as this study was cross-sectional, we had no way of knowing the temporal nature of the relationship between smoking and happiness. Prospective research is needed in these countries to determine how smoking and happiness are associated across time including measuring the more immediate mood-altering effects of tobacco smoking as well as its long-term effects.
In conclusion, this study has shown that ex-smokers and never smokers are significantly happier than current smokers and that those smokers with higher levels of nicotine dependence were more likely to have low levels of happiness. These findings have implications for the countries concerned and more generally. First, given that recent research has highlighted the need for large-scale public health campaigns about the negative effects of smoking in the countries in this region,31 the results of this study suggest that such campaigns should also include a component about how stopping smoking can be beneficial in terms of mood. Further, this information should be disseminated by health professionals in their attempts to encourage smokers to quit as it may serve as an additional element that helps translate the desire to quit, which has recently been recorded in the countries,32 into reality. Second, even if smoking may offer some short-term pleasure to individuals, it does not seem to bring them happiness, thereby suggesting that the approach taken by the Food and Drug Administration that places a value on their perceived pleasure should be reconsidered.
What this paper adds
The US Food and Drug Administration has established a policy of substantially discounting the health benefits of reduced smoking in its evaluation of proposed regulations because of the cost to smokers of the supposed lost pleasure they suffer by no longer smoking.
Little is known, however, about the association between smoking and positive affect despite the fact that pleasure is often cited as a reason for smoking.
This study examined this relation in nine countries of the former Soviet Union where there are high rates of (male) smoking and smoking-related mortality.
Results showed that ex-smokers and never smokers were significantly happier than current smokers, while among smokers, higher levels of nicotine dependence were associated with lower levels of happiness.
Future public health interventions to reduce smoking should highlight that smoking cessation may result in better physical and emotional health.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online appendix
Contributors AS conceived the study idea and wrote the main text of the study. AK analysed the data and commented on the manuscript for intellectual content. BR and MM supervised the HITT study and commented on the manuscript for intellectual content. ML and YG commented on the manuscript for intellectual content.
Funding This work was supported by the European Union's 7th Framework Programme, project HEALTH-F2-2009-223344. The European Commission cannot accept any responsibility for any information provided or views expressed. AK's work was supported by a Miguel Servet contract by CIBERSAM.
Competing interests None.
Ethics approval Ethics committee of the London School of Hygiene and Tropical Medicine and was carried out in accordance with the Helsinki Declaration.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Tabulations of all the HITT variables are freely available on the HITT study website at: http://www.hitt-cis.net.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.