Association of Sources of Worry and Stress with Smoking in Japan: Findings from the Comprehensive Survey of Living Conditions 2010

Objectives The purpose of this study is to examine which sources of worry and stress are associated with smoking behavior and if these associations are varied by gender. Methods The anonymous data of 28,588 men and 30,813 women aged 20-79 years old in Japan were analyzed. We defined the subjects who chose “smoking everyday” as “smokers.” We also assessed 20 sources of worry and stress. Multivariable odds ratios (ORs) and 95% confidence intervals (CIs) of smoking for each source of worry and stress were calculated using logistic regression analysis. Results We found significant association of financial stress with smoking behavior in both genders. The multivariable OR (95% CI) of smoking according to stress from financial stress was 1.22 (1.15-1.31) in men and 1.66 (1.53-1.80) in women. Moreover, the OR of smoking according to stress from human relationships and job affairs was significantly higher for women. Conclusion We found that some sources of worry and stress were related to smoking behavior and these associations were found to be gender-specific.


Introduction
Smoking is known as one of the coping actions for stress regardless of gender, age, and race 1) .Previous studies have suggested that psychological distress is strongly associated with a higher prevalence of smoking behavior among adults [2][3][4][5][6][7] and is also associated with the risk of lifetime smoking for adolescents 8) .Another study has suggested that reducing work stress by increasing job control, decreasing job strain, and finding a better balance between personal efforts and rewards gained from work might help smoking cessation efforts 9) .Thus, stress management is important for smoking control.
A large Japanese cross-sectional study reported a statistically significant association between being a current smoker and severe psychological distress (SPD) for both genders; however, in terms of former smokers, the study found an association Tomoyo SATO 1) , Mitsuya MAEDA 1) , Yumi KAWATA 1) , Koutatsu MARUYAMA 1,2) , with SPD in women only 10) .Another study using data from the Comprehensive Survey of Living Conditions (CSLC) showed that SPD was associated with people currently smoking in Japan, and the association was greater for women than for men 2) .These suggested that there may be gender differences in the association between smoking behaviors and stress.
Women tend to have a higher risk of anxiety and depressive disorders due to the gender-based differences that may emanate from biomedical (hormonal, anatomical, and hippocampal neurogenesis), psychosocial (personality, coping, and symptom reporting), or epidemiological population-based risk factors (cultural, social, economic, and political procedures) [11][12][13] . Sciocultural roles and adverse life events involving children, housing, or reproductive problems result in increased responsibilities for women, which may also cause a clear gender difference in the excess of depression onset among women 12) .
To promote non-smoking policies, the Japanese government's Ministry of Health, Labour and Welfare (MHLW) recommended offering smoking cessation support programs through health checkups and a health guidance system, provided mainly by public health nurses.The programs would be covered by individuals' medical insurance.Although the existence of stress is interviewed, the sources of stress that relate to smoking behavior and the advice to solve these were not considered in the programs 14) .
As the associations between smoking behaviors and stressors can differ between genders, it may be useful for effective smoking prevention programs to identify the gender-specific sources of worry and stress that could relate to smoking behaviors.However, few studies have examined the association between the sources of stress and smoking behavior using a large national dataset representative of the population in Japan.Therefore, the purpose of this study was to examine the genderspecific stress sources associated with smoking behavior among the general Japanese population.

Data source
We used the data collected in 2010 for the CSLC, a survey conducted by the MHLW.It is a self-ad-ministered questionnaire survey with the purposes of researching the basic subjects of living conditions such as health, medical care, welfare, pension, and income and of obtaining the basic data that are required for the planning and management of the Health, Labour and Welfare administration.This survey covers households and household members across the nation.Data on entire households (around 290,000 households) and household members (around 750,000 persons) in the 5,510 areas selected by stratified random sampling from sub-districts 1 and 8 of the 2005 Population Census's enumeration districts were used.We got official approval to use the secondary data from the Statistics and Information Department of the MHLW.We also used the data from the household and health questionnaires of the CSLC in 2010.

Subjects
The anonymous data of 34,483 men and 36,830 women between 20-79 years of age were obtained from the MHLW.Of these, 10,742 and 1,170 people did not answer the questions regarding smoking status and experience of stress, respectively.We excluded these non-respondents from the present study.In the end, the data from 59,401 people (28,588 men and 30,813 women) aged 20-79 years old were used for this study.

Assessment of smoking status
Regarding smoking status, participants were asked "Do you smoke?" and the options for selection were "Non-smoker," "Smoking every day," "Smoking occasionally," or "Smoking before but have not smoked for more than 1 month."We defined the subjects who chose "Smoking everyday" as "Smokers," and the subjects who chose "Non-smokers," "Smoking occasionally," or "Smoking before but have not smoked for more than 1 month" as "Non-smokers."

Assessment of worry and stress
Similar to the previous study 15) , the participants were asked "Do you have worry and stress now?", and only the persons who answered "Yes" were further asked to answer all items that matched their sources of worry and stress as follows: "Relationship with family," "Relationship with nonfamily members," "Related to love-making or sex," "Marriage," "Divorce," "Bullying, sexual harass-ment," "Lack of purpose in life," "Lack of personal time," "Income, finances, debts, etc.," "Own illness or long-term care," "Illness or long-term care of a family member," "Pregnancy or child birth," "Child rearing," "Housework," "Own studies, exams, entrance exams," "Child's education," "Own job," "Job of a family member," "House or living environment," and "Other."

Other measurements
We also assessed the following measurements via the self-administered questionnaire: education level (junior high school, high school, junior college or technical upper secondary school, or university or graduate school); marital status (single, married, divorced, or widowed); and working status (Worker, Non-worker).
Regarding working status, participants were asked the following three questions: "Did you work to obtain income in May 2010?", "Were you employed or self-employed?", and "What was your employment status?"When the participant answered "Unemployed," he or she was also asked whether he or she had the intention to work to obtain income.Participants who chose their working status as "Mainly doing job tasks," "Mainly doing housework but have a job," "Mainly doing other but have a job," "Only attending school," "Only doing housework," "Other with no job."We defined the subjects who chose "Mainly doing job tasks," "Mainly doing housework but have a job," "Mainly doing other but have a job" as "Worker" and those who chose "Only attending school," "Only doing housework," and "Other with no job" as "Non-worker."

Statistical analysis
Demographic characteristics according to stress status and sources of worry and stress stratified by gender are described, and chi-squared analysis was conducted.We calculated the odds ratios (ORs) and 95% confidence intervals (CIs) of smoking for each cause of worry and stress compared with those in the absence of any worry and stress using the multivariable logistic regression model.The confounding factors that are, age, education level, and marital and working statuses, included in the model as they were significantly associated with both stress and smoking status.All statistical analyses were performed using SAS ver.9.4 software (SAS Institute Inc., Cary, USA).All probability values for statistical tests were two-sided, and values of p < 0.05 were regarded as statistically significant.
Table 1 shows the demographic characteristics according to stress status.For both genders, the persons experiencing stress were relatively younger, and had a higher proportion of being single or divorced/widowed, workers, and heavy smokers than those with no stress.

Summary of the results
For both genders, those who had financial stress (i.e., low family budget, debts) were associated with a higher prevalence of smoking behavior.Women with worry and stress from human relationships or their jobs were associated with a higher prevalence of smoking behavior.By contrast, men with health-related stress were associated with a lower prevalence of smoking.

Stressor types and smoking behavior
Multifarious types of stressors are known to give rise to severe psychological distress (SPD), which may affect smoking behavior.Our study is the first to provide evidence for the association between the prevalence of smoking behavior and different sources of worry and stress, rather than just stress itself, using a large, nationally representative dataset in Japan.Our results show that associations between the sources of worry and stress with smoking behavior differed by gender, which suggests that gender-specific considerations will be needed when developing effective smoking control policies; these should be specific to each source of worry and stress.
For both genders, we found a significant association between financial problems and smoking behavior.Several studies have shown a significant association between smoking behavior and socioeconomic indicators that include an individual's economic situation, such as education, occupational status, household income, housing tenure, economic difficulties, and economic satisfaction [16][17] . I Japan, the National Health and Nutrition Survey 2010 conducted by the MHLW revealed that the association between household income and smoking behavior was more significant in low-income families 18) .The present study is consistent with previous studies and the government data.When we further analyzed the data by gender, we found that smoking behavior was associated with the stress related to financial problems, regardless of gender.A previous study in the US revealed that when smokers did quit, their financial stress decreased 19) ; therefore, a crucial aspect of promoting the cessation of smoking may involve the consideration of monetary expense.
Our findings also suggested that smoking behavior was associated with job stress in women.In Japan, the employment rate of women is increasing, and the proportion of women among the total working population was 43.4% in 2016 20) .However, culturally, the burden of domestic tasks, like housework, care of the elderly, and childcare duties at home, tends to fall mainly on women.The time given for childcare and housework (in families with a child less than 6 years old) is 454 hours by women and 83 hours by men 21) .In addition, according to the Caregiving Household Service Survey in 2000, most caregivers in the household are women (72.2% women, 19.5% men) 22) .Married women with children often reported having poor health due to the strain of occupying multiple roles (i.e., being a wife, mother, worker, and homemaker concurrently) 23) .
Another study reported that employees who were taking care of Japanese elderly relatives were significantly associated with an increased risk of depression 24) .The study also investigated the association between work-related stress, caregiver role, and depressive symptoms using K10, the 10question self-rating scale developed by Kessler in the US to screen for severe psychological distress (SPD).The proportion of respondents with a high K10 score was significantly higher in women than in men.Women cope with multiple tasks, including housework, care for the elderly, and childcare duties at home, in addition to jobs outside the home; therefore, women are subjected to increased stress from various sources 24) .Combining these results with previous studies may give some insight into the potential significance of our findings of the gender-specific sources of stress associated with smoking behavior.Furthermore, our study showed that worry and stress emanating from interpersonal relationships were associated with a higher prevalence of smoking behavior in women.Additionally, women were more likely to suffer from multiple sources of stress in the present study.By contrast, men were not subjected to certain stressors pertaining to interpersonal relationships.As mentioned previously, the traditional social role of men in Japan is to work 20) , and as such men often prioritize their efforts towards their jobs.On the other hand, women are traditionally expected to multitask and cope with not only their jobs but also the housework, care of the elderly, and childcare duties.Our data suggested that given these differences in societal expectations and increased caretaking demands, women were more likely to suffer stress brought on by their various interpersonal relationships.

Strengths
In this study, we used the anonymous data derived from the CSLC, covering approximately 290 thousand households and household members (approximately 750 thousand persons) who were randomly sampled.The strength of the present study is that it is the first to use a large national dataset of 20-79 years-old adults, representative of the population in Japan, to provide results on the association of the prevalence of smoking behavior with not only the experience of stress, but also the different sources of stress.This study also provides insights into gender-specific differences in sensibility towards worry and stress.

Limitations
Our study has several limitations.First, because the study design was cross-sectional, we could not determine any causal relationship between smoking behavior and stressors.Second, people responded to the questionnaire according to their personal view of their own levels of stress rather than an objective measure.Individuals might have had different definitions of what constituted "a lot of stress," and their responses to a stressful event might also have varied.Third, while income level would likely be related to both smoking behavior and stressors, we could not consider income level as a variable because the data were unavailable.Fourth, we categorized "smoking occasionally" or "smoking before but have not smoked for more than 1 month" as "non-smokers" since the frequency of smoking was not asked in this survey and this might result in either an underestimate or an overestimate of the true association.Fifth, we excluded non-respondents from the present study.However, people with severe SPD or heavy smokers were considered to be more likely to decline to participate in the survey, which might have led to a selection bias.Finally, the prevalence of experiencing overall stress was higher among women than men in our present study.It could be possible that the gender difference in the prevalence of stress has influenced on our present results.However, many of previous studies have consistently revealed that women reported being more distressed than men 25) .There is also growing evidence suggested that women and men are stressed by different types of situations [26][27] . Nnetheless, further sensitivity study may need to be conducted by using population of no gender difference in prevalence of stress.

Conclusion
The various sources of stress were associated with smoking behavior in a gender-specific manner.Both genders suffered from stressors related to financial issues (i.e., low family budget, debts), which correlated with smoking behaviors.Women experiencing stress associated with human relationships and work-related issues had a significantly higher OR for smoking, and men with health-related and educational stress had a significantly lower OR for smoking.To create effective tobacco control policies, it is important to be aware of the gender-specific sources of stress that may be behind an individual's smoking behavior.

Table 1
Demographic characteristics according to Stress status of 59,401 people (28,588 men and 30,813 women) *Chi-square test.

Table 2
Sources of worries and stress by sex *Chi-square test.

Table 3
Multivariable-adjusted odds ratio (OR) according to types of stress