Association between having a meal together with family and smoking: a cross-sectional nationwide survey

Smoking is a global public health issue and the most significant risk factor affecting health. The World Health Organization (WHO) estimates that the annual death toll from smoking is 8 million, and has rated tobacco as “one of the biggest public health threats the world has ever faced” [1]. Smoking is recognized as a risk factor for all types of cancer [2]; cardiovascular diseases such as ischemic heart disease [3, 4], arrhythmia [5, 6], and stroke [3, 7]; respiratory diseases including chronic obstructive pulmonary disease [8]; and mental disorders such as depression [9] and schizophrenia [9]. In Korea, the conventional cigarette smoking rate has decreased among adults. In 2021, the smoking rate of conventional cigarettes among adults was 19.3%, a decrease of 3% from 22.3% in 2017. However, the magnitude of this decline varies significantly by sex and has not yet reached the government’s target [10]. According to the government’s 5th National Health Plan (Health Plan 2021 ~ 2030), the smoking rate for adult men and women will reach 25.0% and 4.0% by 2030, compared to 36.7% for men and 7.5% for women in 2018. Additionally, there has been an increase in the use of e-cigarettes instead of conventional cigarettes [11]. The WHO Health Organization emphasizes that the health benefits of smoking cessation are as evident as the harmful effects of smoking [1]. Previous studies have shown that the risk of ischemic heart disease [12], acute myocardial infarction [13], and lung cancer [12] mortality are reduced in people with a long smoking cessation period. Therefore, smoking cessation is considered a worldwide major public health priority and efforts are made annually to set targets to reduce smoking rates.

Family is the most important and central experience in social relationships [14]. According to previous studies, individuals who live with their families and have a strong sense of kinship often make health-promoting decisions [15]. In particular, the odds ratio (OR) for attempting to quit smoking was higher among those who were married and cohabiting with their spouse [16]. This can be attributed to interactions, communication, and conversations, such as family members expressing dissatisfaction with second-hand smoke or concern for the family’s health, which positively influences smokers’ intentions to quit [16]. In other words, family members’ interactions and influence on health management can positively affect smokers. Furthermore, several studies have shown that dining with someone has a positive impact on mental health [17, 18]. Having meals with family members has a positive effect on family cohesion, facilitating easier control over health behaviors within the household and yielding beneficial outcomes [19, 20].

Previous studies on the association between family meals and smoking have predominantly focused on adolescents, both domestically and internationally. Cohort studies conducted by the National Heart, Lung, and Blood Institute in the United States and studies on Minnesota adolescents indicated that dining with family members increases the sense of kinship within the family and ultimately inhibits smoking behavior through the structure and flow of family dynamics [21‐23]. Moreover, based on data like the National Longitudinal Survey of Youth in the U.S., studies have analyzed the correlation between the frequency of family meals among adolescents and smoking [24]. Similar findings have been observed in Israel and Scotland [25, 26]. Studies conducted in Korea have also shown that the OR of smoking experience among groups of adolescents who do not have meals with their families is higher than those who have meals with their families, and research results indicated that a positive atmosphere during family meals reduces problematic behaviors in adolescents [27]. A previous study targeting middle-aged men revealed that higher satisfaction with family relationships was associated with lower odds of smoking [28]. Obtaining advice and support within the family enhances smoking cessation behavior, while tension within the family increases the motivation for smoking as a way to relieve stress [29].

Therefore, this study aimed to investigate the association between having meals with family members and smoking behavior in adults, and identify practical information that can be used to suggest policies and establish institutional support to enhance smoking cessation success rates.

Characteristics of the study population are presented in Table 1 and descriptive statistics for each smoking behavior of participants are shown in Supplementary Table 2. Among male participants, 3,855 (80.3%) reported having meals with their families, whereas 5,211 (83.0%) female participants reported having meals with their families. Among male participants who had meals with their families, those who currently smoke accounted for 29.6%. Among those who did not have meals with their families, the current smoking rate was 38.0%. Similarly, among female participants, the current smoking rate was 4.0% among those who had meals with their families, while the current smoking rate was 9.2% among those who did not have meals with their families. The chi-squared test revealed a statistically significant association between having a meal with family and current smoking status for both male and female participants (p < 0.0001).

Table 2 presents the results of the multiple logistic regression analysis, adjusting for all covariates and stratified by gender, to examine the association between having a meal with family and current smoking status. The odds of currently smoking among male participants were 1.27 (95% confidence interval [CI]: 1.05–1.54) if they did not have a meal with family compared to those who did. Among female participants, the odds of currently smoking were 1.90 (95% CI: 1.33–2.71) if they did not have a meal with family compared to those who did.

Table 3 presents the subgroup analysis performed to evaluate the combined effect of having a meal together with family, age, marital status, educational level, region, occupational categories, household generation composition, and number of household members on current smoking status. Regarding male participants who did not have meals with their family, the strongest association with current smoking status was observed among the older adult population (60 + years of age: OR 1.79, 95% CI 1.22–2.62), married group (OR 1.25, 95% CI 1.00–1.57), and inoccupation group (OR 1.87, 95% CI 1.19–2.93). Regarding female participants who did not have meals with their family, an association with current smoking status was observed among middle-aged adults (30–39 years of age: OR 2.10, 95% CI 1.02–4.31; 40–49 years of age: OR 2.07, 95% CI 1.02–4.18), unmarried status (divorced or separated: OR 2.12, 95% CI 1.20–3.73; single or widowed: OR 8.93, 95% CI 1.71–46.62), and individuals in pink or blue-collar occupations (pink-collar: OR 2.42, 95% CI 1.20–4.88; blue-collar: OR 3.55, 95% CI 1.56–8.06). Additionally, for male participants, an association was observed with lower education levels (OR 1.84, 95% CI 1.14–2.95), rural region (OR 2.00, 95% CI 1.38–2.92), 1st household generation composition (OR 1.59, 95% CI 1.07–2.36), and two household members (OR 1.59, 95% CI 1.13–2.23). For female participants, the same association was shown with low education levels (OR 2.12, 95% CI 1.08–4.17]; rural region (OR 4.37, 95% CI 1.70–11.25); 1st household generation composition (OR 1.88, 95% CI 0.93–3.81); and two household members (OR 2.11, 95% CI 1.22–3.66). A linear trend was evident in ORs in accordance with these factors.

Figure 1 presents the results of the subgroup analysis stratified by gender, indicating the association between the frequency of having a meal with family and smoking status. When considering individuals who had meals with their families as the reference category, both genders showed a linear increase in the ORs for current smoking status as the frequency of shared meals decreased (1 time per day—male: OR 1.60, 95% CI 1.31–1.95; female: OR 1.74, 95% CI 1.19–2.54 / 0 times per day—male: OR 1.68, 95% CI 1.36–2.09; female: OR 2.72, 95% CI 1.77–4.19). Furthermore, when analyzing the relationship based on meal types (Table 4), the odds of current smoking were higher when male participants did not have breakfast or lunch together (not having breakfast together: OR 1.62, 95% CI 1.36–1.92; not having lunch together: OR 1.50, 95% CI 1.23–1.83), while for female participants, the odds were higher when they did not have breakfast or dinner together (not having breakfast together: OR 1.98, 95% CI 1.38–2.84; not having dinner together: OR 1.99, 95% CI 1.42–2.77).

Figure 2 presents the analysis of the association between having a meal with family and current smoking status as well as past smoking experience, types of tobacco products used, and smoking cessation attempts or plans among people who currently smoke. Generally, when not having a meal with family, people who currently smoke had higher ORs than people who previously smoked (male: OR 1.21, 95% CI 0.94–1.56; female: OR 1.91, 95% CI 1.33–2.74), and showed a strong statistical association with smoking only conventional cigarettes (male: OR 1.28, 95% CI 1.00–1.67; female: OR 2.22, 95% CI 1.42–3.46). In contract, dual smoking (male: OR 1.07, 95% CI 0.71–1.63; female: OR 1.76, 95% CI 0.86–3.62) and e-cigarette-only use (male: OR 0.97, 95% CI 0.58–1.63; female: OR 1.05, 95% CI 0.47–2.34) showed a relatively low association with family meals. Additionally, as shown in Table 5, groups that did not have meals with their families tended to have high ORs for currently having no history of smoking cessation attempts (male: OR 1.06, 95% CI 0.80–1.41) in the past year and not planning to quit smoking in the future (male: OR 1.38, 95% CI 1.02–1.86; female: OR 1.71, 95% CI 0.70–4.19).

The results showed that both genders had a lower risk of current smoking when eating with their families than when they did not. This is in line with previous studies showing that having a meal with families has a positive association with increasing health behaviors. This trend was more pronounced among female participants than male participants. According to previous studies, women who smoke are more emotionally affected than men who smoke when deciding to quit smoking, and the effect of nicotine therapy replacement is relatively low [19, 24, 27]. Considering this, it can be inferred that family meals plays a more important role in smoking cessation among female participants than male participants [34].

It is particularly noteworthy that the risk of smoking was lower in the group that often had meals with their families. Both men and women were less likely to smoke when they ate at least one meal a day with their families than the group who ate alone, and both men and women were less likely to smoke in the group that ate more than two meals a day with their families; both were statistically significant while controlling for other covariates or confounding variables such as age and socioeconomic status. This supports previous studies showing that the more frequent family meals are, the lower the likelihood of smoking among male and female adolescents [24]. Additionally, another study on middle-aged men found that high family relationship satisfaction lowered the risk of smoking and suggested that family advice may have strengthened smoking cessation behavior [29]. The present study’s findings support those of previous studies and show an association between family meals and smoking in adults, both men and women, and between adolescents and middle-aged men [24, 29]. Furthermore, even when the three meals (breakfast, lunch, and dinner) were analyzed separately, the ORs for current smoking status, regardless of the three meals with family, were lower. However, it is assumed that the reason lunch was not statistically significant is that the number of samples analyzed was insufficient due to participants often not spending time with their families in the age group with an active social life.

According to the results of the independent subgroup analysis, the age groups with a relatively higher association between having a meal together with family and smoking were male participants in their 60 s and older and female participants in their 30 s and 40 s. According to previous studies, it can be assumed that older adult men’s family ties play an important role in health care [35, 36]. Additionally, the fact that women in their 30 s and 40 s constitute an age group that focuses on pregnancy, childbirth, and childcare may have increased the relationship between family ties and smoking [37‐40]. In addition, smoking cessation programs using family support could be more active in rural than metropolitan or urban areas, and the need to be actively implemented in two-person households, which are simple households, was emphasized.

Finally, the analysis conducted by dividing the relationship between having a meal with family and smoking by cigarette type revealed remarkable results. The group that smoked only conventional cigarettes was the most affected by family meals. People who used both conventional cigarettes and e-cigarettes were relatively less affected, and the group that used only e-cigarettes had the lowest association. Because e-cigarettes smell relatively less than conventional cigarettes, e-cigarette users are presumed to have fewer opportunities to receive health advice from families while eating meals together [41, 42]. Further, the subgroup analysis results of currently smoking people’s attempts or plans to quit smoking showed that women who had a meal with family are likely to have attempted to quit smoking within the one year. Moreover, they were more likely to plan to quit smoking the following month. This is in line with a previous study [34], and it is suggested that currently smoking women who have meals with their family have a relative intention to quit smoking due to emotional effects such as family pressure to induce smoking cessation.

This study has several limitations. First, this study was cross-sectional, which means that the temporal relationship is unclear, and reverse causality may be possible. Regarding family meals, it is unknown when the habit of eating with the family started and whether it preceded smoking initiation. Therefore, caution is required when interpreting these results and further prospective cohort studies are required to clarify these findings. Second, the KNHANES uses self-report surveys, which introduce potential limitations in the reliability and accuracy of health-related, socioeconomic, and smoking statuses. This can result in recall bias, particularly regarding the underestimation of smoking prevalence. Third, despite attempts to include as many independent variables related to family meals and smoking as possible, potentially uncontrolled confounding variables may still exist. Fourth, it is important to note that the presence or absence of family meals does not necessarily reflect the depth of family relationships or the frequency of face-to-face interactions. It is possible for families to have a close bond even without regularly eating meals together, or conversely, not to have a strong bond despite sharing meals together. Finally, this study did not assess the quality of family meals. According to a previous study, even when families eat together, there can be significant differences in the proportion of mealtime spent in conversation depending on family members’ participation [25]. In this study, only the presence, frequency, and types of family meals were analyzed, while, due to limitations in the KNHANES data, data on the number of family members participating in meals or the extent of conversation with families during meals were not collected.

Despite these limitations, this study has several strengths. First, this study utilized data from the KNHANES, a nationally representative survey that reflects the health behaviors and characteristics of South Koreans. Second, the inclusion of recent data from 2019 to 2021 is significant, as it encompasses not only current smoking status but also factors such as e-cigarette use, past smoking history, and smoking cessation attempts and plans. Third, although previous studies examining the association between family meals and smoking have often focused only on conventional cigarette smoking or adolescents, this study included all adults and e-cigarette smoking.

This study found that having a meal with family members may have a positive effect on adult smoking control and smoking cessation intention. Considering tobacco addiction and the continued release of new e-cigarettes, the government needs to actively utilize social support for smoking cessation activities, such as encouraging having meals with family members. Additionally, educational programs that promote family dialogue and remind people who smoke of their bonds with their families are important. Thus, it is necessary to develop programs and actively promote smoking cessation clinics at public health centers and educate family members about the importance of family meals. By combining these efforts, the health of people who smoke can be improved and a healthy culture of smoking cessation can be created through family and social support.