Tobacco use is a common and preventable risk factor for a wide range of chronic, noncommunicable diseases [1]. China represents approximately 40% of the world’s total cigarette use [2], causing about 20% of all adult male deaths in 2010 [3]. According to the China Adult Tobacco Survey in 2018, there were 308 million smokers, and a relatively low percentage (16.2%) intended to quit smoking within 12 months [4]. This indicates that tens of millions of smokers needed help from medical and health institutions such as smoking cessation clinics, because most self-help quit attempts failed [5]. However, our previous nationwide survey of available smoking cessation support programs showed an imbalanced distribution of resources, such as smoking cessation clinics, and limited availability of medications in China [6]. Therefore, tobacco cessation interventions that are of low cost, are convenient, and have potential for wide reach are urgently needed in low- and middle-income countries (LMICs).

Text message interventions for smoking cessation have been proven to be efficacious and have great potential for serving geographically and economically diverse populations. On the one hand, a large number of randomized controlled trials (RCTs) have shown that text-based tobacco cessation programs can significantly increase smoking abstinence rates. However, the definition of the primary outcome of smoking abstinence in previous studies has varied, for example, biochemically verified continuous abstinence at 6 months in studies in the United Kingdom [7], the United States [8], Hong Kong [9,10], and mainland China [11]; self-reported, 7-day point prevalence abstinence (PPA) at 6 months in Norway and Hong Kong [12,13]; 1-month PPA in Sweden [14]; and self-reported, 30-day PPA at 9 months in the United States [15]. On the other hand, a mobile intervention could be achieved through WeChat (a popular, multipurpose messaging, social media, and mobile payment app). There are more than 1.1 billion WeChat users in China [16]. Therefore, the intervention could offer an opportunity to extend the reach of tobacco cessation interventions outside of traditional health care settings via a mobile phone app that most adults already own, is generally always on, and is already used many times each day.

However, the real-world effectiveness of text messaging programs is largely unknown, although great efficacy had been achieved in clinical trials [7-15]. Evaluation of effectiveness in real-world settings is of great significance, because the public health impact of an intervention crucially depends on not only its efficacy but also real-world uptake, which could be influenced by population-specific characteristics and other factors [17]. To our knowledge, only 3 previous studies explored the real-world effectiveness of a text messaging intervention [18-20]. Primary outcomes in 2 studies of smoking cessation by the Veterans Health Administration were short-term abstinence at 5 weeks [18] or self-reported 30-day PPA at 6 months [20]. In a program conducted in India, self-reported 30-day PPA at 4 months to 6 months was used [21]. However, no published studies have explored the effectiveness of self-reported 7-day PPA at 6 months as the primary outcome, even though this abstinence rate was commonly used in previous RCTs.

Therefore, to gain a more comprehensive understanding of the effectiveness of a text message intervention, we conducted a real-world analysis of 2 years of nationwide data from a message-based tobacco cessation (mCessation) program. The mCessation program is based on a library of 162 short text messages provided by the World Health Organization (WHO) and can be used by scanning a WeChat QR code. In this study, we aimed to provide real-world data about the utilization and effectiveness of the mCessation program in China.

This study examined the real-world use of mCessation, a low-cost, text messaging, WeChat-based smoking cessation program, among smokers in the general population in China. Even though nearly three-quarters of mCessation users were younger than 40 years old, the technology did not limit the use of mCessation by older smokers: The average age was about 35 years, and the oldest user was 71 years old. Most of the mCessation users were male smokers (96.2%) and with light tobacco dependence (89.1%). For telephone interview respondents, the primary outcome, the 7-day PPA rate at 6 months, was 21.9%. Those who were older than 40 years, single, and with light tobacco dependence were significantly more likely to successfully quit smoking at 6 months than the corresponding control groups. In addition, we also found that participants who worked in a public service unit or participated in mCessation after becoming aware of it through special publicity activities were more likely to not respond to the telephone interview than the control groups.

Our data support that real-world utilization of mCessation is related with age and gender factors. The abstinence rate was higher in older people (≥40 years) than in those aged <30 years, which might be explained by more underlying diseases and greater intention to lead a healthy life in the older adults. However, consistent with previous findings in Bangladesh [27] and China [28], younger participants had higher acceptance of an online intervention than older people, which was easily understandable because young people were more familiar with digital devices, which are ubiquitous in their lives [29]. Therefore, future research should pay more attention to improving the smoking cessation rates of young people, not just the acceptability, which would further improve the effectiveness of text messaging interventions. Similarly, increasing the acceptance of mCessation by older people would have great potential benefits, based on their positive abstinence rates in the real-world settings and high rate of mobile phone use and coverage of WeChat among Chinese citizens. There is a large number of older smokers in mainland China, and a nationally representative, cross-sectional study found that the number of current smokers was higher among those aged 40 years to 69 years than among those aged 20 years to 39 years in the general population (16,705 vs 5282) [30].

In addition, the fact that nearly all mCessation participants were male could be explained by the following 2 reasons. On the one hand, the 2018 China Adult Tobacco Survey estimated the prevalence of current smoking was 50.5% in men and 2.1% in women [4]; therefore, the number of male smokers was much higher than that of female smokers. On the other hand, earlier studies in China had shown that women were less likely to accept an online approach than men [31,32], which was also reported in a study with veterans [20]. This might be because men are more task-orientated and more interested in new technology [32,33]. Moreover, inconsistent with the findings of earlier studies, we also found an association between acceptance of a text messaging intervention and educational levels, which was reflected by the high proportion of those with at least a college or university education in our study [16,34].

Among mCessation users in the real-world setting, the prevalence of tobacco dependence was higher than in our previous cross-sectional study in the general population [30]. This suggests that tobacco dependence was associated with an individual’s acceptance of the text messaging intervention, because smoking dependence was the main reason for challenges with stopping smoking [35]. Consistent with the findings of a cluster RCT with Chinese adolescents [36], we believed that a mobile phone text messaging intervention could encourage those dependent on cigarettes to quit smoking, especially for those who were mostly light smokers. The proportions of individuals with a light tobacco dependence were about 50% in the general population [30] but nearly 90% in mCessation participants, which indicates that these lightly dependent populations have a higher willingness to quit smoking than those with a moderate or severe dependence. For those with a severe tobacco dependence, it is difficult to quit smoking on their own initiative, because they tend to have a complex and severe condition involving physiological, psychological, and behavioral processes, which is considered a mental disorder by the ICD [37].

We first reported that, among mCessation telephone respondents, the real-world 7-day PPA at 6 months was 21.9%, which is higher than in previously reported RCTs (12.1%-17.7%) [12,13]. We believe the difference in the abstinence rates between our real-world study and previous RCTs was likely due to the underlying population differences: The proportion of individuals with a light tobacco dependence (89.1%) was higher than in RCT studies in Hong Kong (51.3%) [13] and Norway (29.4%) [12]. A cluster RCT conducted in the Chinese population supported that a mobile phone text messaging intervention could inhibit cigarette dependence in adolescent smokers who were mostly light smokers [36]. In addition, the 21.9% abstinence rate in our study was similar to that in a study in India (19.1%) [21] but higher than those in 2 studies conducted with veterans (13% in 2016 [18]; 3.7% in 2021 [20]). This might be due to differences in follow-up methods (telephone interviews vs message interviews) and the targeted population (general population vs veterans).

In the multivariate analysis, we also found that those with a heavy dependence were less likely to quit smoking. On the contrary, older adult or single smokers had higher chances of stopping smoking, which may be because older adults prefer to make behavioral changes due to high health awareness, while single individuals pay more attention to giving the impression of a healthy lifestyle to find a partner [38]. Notably, a follow-up telephone interview to determine abstinence rates was of great importance to evaluate the effectiveness of a text messaging intervention in the real world. Published real-world data had shown quite low response rates of subscribers to text messaging programs [20,21], which could lead to bias in smoking cessation rates. For example, with the primary outcome of self-reported 30-day PPA at 6 months, the abstinence rate determined by telephone interview was 19% in a study in India [21] but only 3.7% in a study with veterans without a telephone interview [20]. Therefore, we also explored the influencing factors of nonresponse to the follow-up telephone interview, and the results suggested that, to improve the response rates to telephone interviews, we should add more interventions that target those working in public service units and who found out about mCessation through special publicity activities.

The strengths of this real-world study were that our target population included current smokers in the general population and answering the smoking questions in mCessation was mandatory, which minimized the bias related to the populations generally included in previous studies. For example, some participants who enrolled in research conducted with veterans [18,20] or pregnant individuals [19] were likely not veterans or pregnant women, respectively; even nonsmokers subscribed to the smoking cessation program in the study in India. More importantly, a long-term abstinence rate, the 7-day PPA rate at 6 months, was selected as the primary outcome; this was necessary to evaluate the smoking cessation intervention since tobacco dependence is a chronic, relapsing condition. However, there were still some limitations. First, data for the primary outcome of the abstinence rate were self-reported and collected by telephone interview, which may lead to information bias affecting our results. Therefore, sensitivity analysis could increase the robustness of our results. Second, response rates to the telephone follow-up were lower than 40%, indicating that strategies to improve health awareness and public cooperation were critical to the evaluation of the effectiveness of the smoking cessation program in the real-world setting. Finally, our data collection did not include receipt of additional tobacco cessation care such as the use of smoking cessation medications. It was possible that additional interventions received by the subscribers may have influenced abstinence rates, which would lead to overestimation of our results in this study.

Analysis of mCessation provided important information on the demographics of real-world users, suggesting that the program was more likely to serve a population that has the following characteristics: younger age, men, and light degree of tobacco dependence. The main finding of this analysis was that mCessation could help 1 in every 5 smokers aged 18 years to 67 years quit smoking, but the real-world effectiveness was less significant among young, married, and heavily dependent smokers compared with the control groups. Tobacco control is of great public health significance because tobacco use is one of the main risk factors for noncommunicable diseases (cancer, cardiovascular disease, and respiratory disease), which are a major health threat globally [37]. Thus, a low-cost, efficacious, and easily and publicly accessible smoking cessation modality such as text messages or a mobile app is suitable in China or LMICs. In the future, more studies of how to improve the popularity of mobile cessation programs and their effectiveness in insensitive populations are needed.