Longitudinal Effects of Life Satisfaction on Smoking Status of Korean Adult Men

Article information

J Prev Med Public Health. 2025;58(4):388-395

Publication date (electronic) : 2025 April 21

doi : https://doi.org/10.3961/jpmph.24.664

Baksun Sung

Department of Sociology and Anthropology, Utah State University, Logan, UT, USA

Corresponding author: Baksun Sung, Department of Sociology and Anthropology, Utah State University, 0730 Old Main Hill, Logan, UT 84322-0730, USA E-mail: baksun777@gmail.com

Received 2024 November 4; Revised 2025 February 17; Accepted 2025 February 26.

Abstract

INTRODUCTION

Cigarette smoking is one of the leading causes of death worldwide [1], resulting in approximately 8 million deaths each year [2]. Smoking is associated with a range of diseases, including coronary heart disease, respiratory disease, and cancer [3]. In 2016, the prevalence of smoking in Korea was estimated at 23.6% among persons aged 15 years and over, compared with 24.5% for the Western Pacific region [4]. In this respect, Korea ranked eleventh highest among Western Pacific countries in 2016 [4]. However, among Organization for Economic Cooperation and Development nations in 2017, Korea exhibited the fifth highest smoking prevalence for men aged 15 years and older [5]. Among Korean adults aged 19 years and older, the smoking prevalence in 2017 was 37.0% for men, substantially exceeding the 5.2% reported for women [5]. Since 2008, smoking rates in Korea have consistently decreased for both men (falling from 47.7 to 37.0%) and women (declining from 7.3 to 5.2%) [5]; however, the prevalence among men is still high, representing a major health problem.

As Korea has become more advanced and wealthy in recent years, the primary concerns of its population have shifted from labor and duty to life satisfaction and well-being. Life satisfaction measures how people assess their lives as a whole rather than their current emotions and represents a key component of subjective well-being [6]. It is particularly associated with positive well-being, which is typically evaluated by measures of trait positive affect, subjective well-being, and life satisfaction [7,8]. Positive well-being can be categorized into eudaimonic and hedonic well-being, with life satisfaction falling under the latter [9]. Previous studies have reported that life satisfaction is negatively associated with poor health outcomes and unhealthy behaviors [10-13]. Although various social predictors of smoking behavior—such as socioeconomic, socio-demographic, psychological, and social-environmental factors, material well-being, and community-level social capital—have been studied [14-17], little is known about the effects of life satisfaction on smoking status among adults. Accordingly, the objective of this study was to identify any associations between various life satisfaction variables and smoking status among Korean adult men. Furthermore, non-smokers may take up smoking at any time due to various social influences, making it essential to follow individuals over time to understand the incidence of smoking. To obtain a more robust measure of the effects of life satisfaction on smoking status, longitudinal data analysis (survival analysis) was applied. Consequently, this study also aimed to examine how life satisfaction variables impact time to event (namely, smoking).

METHODS

Statistical Analysis

Descriptive statistics

Descriptive statistics (Table 1) were used to summarize the demographics and population characteristics from the first wave in 2005 (the beginning of the study period).

Table 1.

Demographics and population characteristics, KoWePS, first wave (2005, the beginning of the study period, n=4723)

Survival analysis

Survival analysis was employed to examine the longitudinal effects of life satisfaction on smoking status among Korean adult men. This analysis is used to determine the time until an event occurs [20]; here, the event is smoking a cigarette. Typical examples of survival analysis include the Kaplan-Meier estimate and the Cox proportional hazards model. Compared to the Cox model, the Kaplan-Meier estimate (a univariate approach) has limitations because it cannot control for potential confounding factors or concurrently examine multiple independent variables. For these reasons, the Cox proportional hazards model, a form of multivariate survival analysis, was applied.

Cox proportional hazards model

The Cox proportional hazards model is a widely used multivariate and semi-parametric statistical method in event history analysis or survival analysis. It is employed to assess the relationship between survival time and one or more independent variables [20]. The model is expressed through the hazard function h(t) [21], which represents the risk of the event occurring at time t [20]. It can be summarized as follows:

h(t)=h₀(t)×exp{b₁(leisure satisfaction)+b₂(social relations satisfaction)+b₃(family relations satisfaction)+b₄(health satisfaction)+b₅(job satisfaction)+b₆(family income satisfaction)+b₇(residential environment satisfaction)+b₈(overall life satisf action)+b₉(age)+b₁₀(education level)+b₁₁(annual income)+b₁₂(employment status)+b₁₃(marital status)+b₁₄(religion)+b₁₅(region)+b₁₆(alcohol use)+b₁₇(depressive symptoms)}

All statistical analyses were performed using Stata version 15.0 (StataCorp., College Station, TX, USA).

RESULTS

Table 1 presents the demographics and population characteristics of the study sample. The mean participant age was 41.99 years (standard deviation, 11.77 years).

Table 2 displays the results of the multivariate Cox proportional hazards model examining the association between life satisfaction variables and time to event (smoking) among Korean adult men aged 20-64 years, after adjusting for age, education level, annual income, marital status, employment status, religion, region, alcohol use, and depressive symptoms. As shown, the sample contributed 28 019 person-years of observation over the 11-year follow-up period. In survival analysis, person-years are calculated as the product of the duration of event-free (here, non-smoker) time and the population size. Generally, a shorter duration of event-free time results in fewer person-years.

Table 2.

Results of the multivariate Cox proportional hazards model, KoWePS, 2005-2016 (total: 28 019 person-years of observation)

Based on the hazard ratios (HRs) obtained from the Cox proportional hazards model, men who reported either satisfaction (HR, 0.83; p<0.01) or moderate satisfaction (HR, 0.84; p<0.001) with their leisure activities exhibited a lower hazard of smoking compared to men who were not satisfied with these activities. Furthermore, men who reported satisfaction with social relations displayed a higher hazard of smoking compared to those who were not satisfied with these relations (HR, 1.23; p<0.05). Finally, men who reported satisfaction with family relations had a lower hazard of smoking compared to men who indicated they were not satisfied with these relations (HR, 0.84; p<0.05). Figure 1 provides a more intuitive interpretation of these results, with the horizontal axis representing time in years of follow-up and the vertical axis representing the probability of surviving (that is, remaining free of smoking). The downward trajectory of the plots indicates that, as time passes, the probability of survival (non-smoking) decreases as the event (smoking) occurs.

Figure. 1.

Survival curve for (A) leisure satisfaction, (B) social realtions, and (C) family relations.

DISCUSSION

In this study, the association between life satisfaction and smoking status among Korean adult men was examined longitudinally. Using survival analysis over an 11-year follow-up period, the findings revealed that several life satisfaction variables were associated with either increases or decreases in the hazard of smoking.

First, the risk of smoking decreased as leisure satisfaction improved. Satisfaction with leisure activities is important because such pursuits contribute to overall health and well-being. Specifically, leisure activities can reduce chronic stress over a lifetime [22,23], which may lower the risk of smoking. However, past studies have reported that leisure activities that lack cultural capital are associated with unhealthy behaviors during leisure time, such as cigarette smoking and binge drinking [24-26]. In contrast, leisure activities characterized by cultural capital (e.g., cultural events) are linked to improved health outcomes [23,27]. Therefore, increasing leisure satisfaction—particularly through culturally enriched activities—could represent an effective strategy to decrease the risk of smoking among Korean adult men.

Second, the risk of smoking increased as satisfaction with social relations improved. Although social relationships generally have a positive impact on mental and physical health [28,29] and foster healthy behaviors [30], the present results suggest an opposite effect that may be partially attributable to Korean culture and social norms. Specifically, many men in Korea try their first cigarette due to peer pressure or social influences. Smoking among this population is closely associated with military service; all Korean men must serve in the military for about 2 years, and veterans are more likely to smoke than the general population [31]. This tendency is explained by the fact that many service members smoke to improve relations with their peers [31]. The social cognitive model of peer modeling supports this observation, suggesting that individuals tend to imitate the behaviors of models they consider similar to themselves [32]. Moreover, Korean social life often revolves around drinking. It is customary for Korean adult men to drink at business dinners after work for organizational socialization [33], and they also drink to maintain relationships with business partners and friends [34]. Since alcohol consumption and cigarette smoking are socially connected [35], and because previous studies have found that drinkers are more likely to smoke [36]—a phenomenon termed “the co-occurrence of alcohol consumption and cigarette smoking”—social relations satisfaction, within the context of Korean culture, may increase the risk of smoking among men. Therefore, social campaigns that emphasize the dangers of smoking and the benefits of quitting may help reduce smoking prevalence among Korean adult men.

Third, the risk of smoking decreased as satisfaction with family relations improved. Family relations satisfaction is defined as the self-awareness of family quality (e.g., solidarity, happiness, and overall relational well-being) [37]. Systematic family patterns and standards, which reflect individuals’ perceptions of communication within the family, are closely linked to the broader culture of the family system [37]. Koreans have long been influenced by Confucian culture, which emphasizes familism and groupism, particularly through age-based hierarchical relationships among family members. If these relationships have a problem, communication challenges within the hierarchical framework can lead to psychological stress. Previous studies have reported a close association between psychological stress and smoking [38-40]. Thus, promoting open communication and granting individuals greater life autonomy may help decrease the risk of smoking among Korean adult men.

These observations should be considered in the context of several limitations. First, life satisfaction and smoking status were measured using self-reported data, potentially resulting in underestimation of these variables due to recall bias. Second, the use of longitudinal (panel) data introduces the possibility of unpredictable outcomes and data loss over time. Third, the price of a pack of cigarettes rose by 80% during the follow-up period (starting in 2015); however, the study could not account for the effects of this change because the KoWePS does not provide variables related to cigarette price changes. Fourth, the follow-up results yielded a binary outcome (smoker, failure=1; non-smoker, survival=0) based on responses to the question, “Do you smoke?”; however, due to data limitations, the non-smoker category did not differentiate between lifelong non-smokers and former smokers. It is also possible that some former smokers mistakenly responded as “smokers,” potentially leading to inaccuracies. Fifth, most Korean men have their first cigarette in their 20s, largely due to mandatory military service, which is associated with the initiation and continuation of smoking behavior. However, only 16.15% of participants in the first wave of the KoWePS were in their 20s, suggesting that changes in state variables (non-smoker [0→ smoker [1]) were primarily based on smoking relapse rather than initiation. Follow-up studies should explore the association between social relations satisfaction and smoking status among Korean men by differentiating smoking relapse from initiation or by comparing participants in their 20s with those in their 30s or older.

Despite the limitations noted above, this study provides meaningful insights from a longitudinal analysis of life satisfaction and smoking status among Korean adult men. The findings suggest that certain life satisfaction variables influence the risk of smoking and that socio-cultural factors appear intertwined with this association. Given the social and health impacts of smoking, public health authorities must develop effective strategies to reduce smoking prevalence among Korean men by considering life satisfaction and socio-cultural factors.

Article information Continued