The Study On Related Influencing Factors Of Smoking Abstinence Self-Efficacy

BackgroundTobacco use is among the leading preventable causes of death worldwide, and nearly half of the smokers die from tobacco use. As the world’s largest cigarette consumer, producer and victim, China is facing a public health crisis with current smoking rates. Meanwhile, as a large agricultural country, China’s rural population accounts for more than half of the total population, and the smoking rate of male residents in rural areas is generally higher than that of urban residents. Besides, the level of economic development in rural area and the scientific cultural quality of the rural population is low. In terms of current tobacco control strategies and measures in China, the work of tobacco control in rural population is lack of pertinence. Therefore, it is of great significance to carry out tobacco control activities and studies in rural areas in China.Although various strategies are available for conducting tobacco control in the population, promoting smokers to quit is one of the main tactics. Evidence showed that numerous factors are responsible for successful smoking cessation, and smoking abstinence self-efficacy (SASE) is one of these important factors, higher score indicates the higher level of self-efficacy. Ma’s study of the Korean American population showed that young smokers (<45 years) had higher perceived self-efficacy in quitting smoking compared with old smokers (≥45 years); other potential predictors were also reported to be associated with SASE, such as nicotine dependence, peer pressure, and social factors.So far, fewer studies related to SASE were conducted in China, and what motioned above were just presented on a superficial level. Therefore, in our study, we use a cross-sectional survey with face-to-face interview to investigate the smoking status of male smokers, who live in several rural areas in China, and to further explore the related influencing factors of SASE, so as to provide scientific theoretical basis and guidance for tobacco control interventions in the future.Objectives1. To detect the association between age and smoking abstinence self-efficacy.2. To detect the potential predictors of smoking abstinence self-efficacy (SASE) and to examine the relative importance of these predictors for SASE.MethodsThis study was conducted in April and May,2013. Subjects investigated were chosen from the population residing in 17 villages of three counties (Ping Yin, Liang Shan and Ju Nan) in Shandong, China. Male population aged 18 and above who ever quitted smoking were face-to-face interviewed and the questionnaire was designed based on Global Adult Tobacco Survey (GATS) Core Questionnaire with Optional Questions. In this cross-sectional study, current smokers were selected as subjects. According to the international division standard of elders before 1994 and the WHO division standard of age, we divided the age of research subjects into the young(<45), middle-aged(45-65) and elderly(≥65); we divided occupation into farmer and others according to the present work status of rural population in China; we divided education into three levels:low (below primary school), middle(primary and junior high school) and high(high school or above); we divided marital status into married, unmarried and others.Smoking abstinence self-efficacy was determined by the Smoking Abstinence Self-Efficacy Scale, and potential confounders were adjusted by multiple linear regression analysis for quantitative data, then calculate the adjusted mean of the dependent variable after its effect was adjusted by the mean of covariates. Fractional polynomials was used to detect the relationship between age and SASE with adjusting for potential confounders of marital status, occupation, and education. The total score of SASE (SASET) and its three context-specific situations scores, i.e. positive/social situation (SASEP), negative/affective situation (SASEN), and habit/addictive situation (SASEH) were regarded as independent variables; nicotine dependence (FTND),trait coping style, including positive trait coping style (TCSQP) and negative trait coping style (TCSQN), general self-efficacy (GSEST) and the level of anxiety (SASST) were regarded as potential predictors; Partial correlation analysis and dominance analysis were performed to determine the primary potential predictors of SASE and to examine their relative importance for SASE with adjusting for potential confounders of age group, marital status, occupation, and education.Main results1. After adjusting marital status, occupation, and education by using multivariate linear regression analysis, the average scores of SASE were different in age groups:it appears to be lower in the middle-aged (45-65years) compared with the young (<45 years) and the elderly (>65 years). This result remained statistically significant except for SASEH.2. Fractional polynomials (FPs) regression was used to explore the relation of SASE to age and the result demonstrated that the relation of SASE to age showed a U-shaped curve with the nadir of about 60 years old.3. The result of partial correlation analysis showed that five predictors involved in our analysis were correlated with each other after adjusting for potential confounders of age, occupation, educational level, and marital status. As for the total score of SASE, TCSQP presented positive while the other four predictors presented negative correlation with it. As for the three context-specific situations of SASE, some different correlations were presented between predictors and dependent variables as follows:different from the correlation between five predictors and SASET, SASST and GSEST were positive correlated with SASEP and SASEH respectively. FTND was the most highly correlated predictor with each dependent variable while SASST was almost the least correlated predictor with each dependent variable.4. The result of dominance analysis showed that FTND was the most important predictor of SASE, its dominance weight was significantly higher than that of the other four predictors. While SASST was almost the least important predictor of SASE. In this analysis, the relative importance of five predictors ranked differently for SASET and its three situations. For SASET, the second important predictor was TCSQN, followed by GSEST, TCSQP and SASST. For TCSQP, it ranked second for SASEP but fourth for the other three dependent variables. It was worth noting that the dominance weight of TCSQP, TCSQN, GSEST and SASST were significantly lower than that of FTND for each dependent variable in this analysis.ConclusionsThis study indicated that SASE was distributed inconsistently in different age groups, and it was significantly lower in middle-aged people compared to younger and elders. Therefore, improving SASE in middle-aged people can promote smoking cessation and reduce the possibility of relapse, which is worth noting for smoking intervention research in the future. Meanwhile, FTND was the most important predictor for SASE, and prevention and treatment which focus on FTND could improve SASE of smokers so as to reduce smoking level and promote health among population.