Text to Quit China: an mHealth Smoking Cessation Trial

Augustson, Erik, Michael M. Engelgau, Shu Zhang, Ying Cai, Willie Cher, Richun Li, Yuan Jiang, Krystal Lynch, and Julie E. Bromberg

American Journal of Health Promotion 31, no. 3 (2017): 217-225


One-third of the adult smokers in the world reside in China. As smoking is a risk factor for health issues such as cancer and cardiovascular disease, the country could see more of these diseases in the future. Smoking has also had a significant impact on the Chinese health system, contributing to at least $6.2 billion in health care costs in 2008. Despite these effects, intention to quit is quite low among Chinese smokers. In 2010, only 15% of adult smokers reported an intention to stop smoking in the next year, and 36% said they had tried to quit in the year prior. Programs that increase awareness and support smoking cessation are limited in the country. Cell phones are promising platforms for such programs, as cellular network access in China is widespread; however, only one text message smoking cessation study has been conducted in the country, and it showed a reduction in the number of cigarettes smoked by adolescents after intervention exposure. The current study aims to assess the impact of a text message smoking cessation intervention on smoking status over a period of six months in China. 


The program was a six-week text message smoking cessation intervention containing text messages adapted for the Chinese population from a text message library designed by the National Cancer Institute for use in the United States. The high frequency text contact (HFTC) group received the full intervention and the low frequency text contact (LFTC) group served as controls. All participants received eight encouraging messages in the week before their quit date. This was then followed by the six-week intervention when the low frequency group were sent one text message per week (six messages total) and the high frequency group received 91 messages at a decreasing frequency. Messages included motivational content, tips on how to resist smoking, and information on the health impacts of smoking.     


Design: Recruitment was done via text messaging using Nokia Life Tools, a program for Nokia phones that sends users text messages with content of their choice. Subscribers to the software received a message inviting them to participate in a free program that would share text-message context for one week about smoking’s impact on health. Those who completed the program were then invited to participate in the study. The first 8,000 who opted in were randomly assigned to either the high-frequency or low-frequency groups (randomized controlled trial). 

Participants self-reported their smoking behaviors in the seven days before the end of the intervention, and then at one-, three-, and six-months post-intervention. A small reimbursement was given in exchange for their reply. These data were then used to calculate the outcome of interest, prevalence of smoking cessation, at the four time points. Geographic data on the participants were provided by Nokia. In addition, a sub-sample of participants in the high-frequency group were contacted via phone post intervention and asked about their satisfaction with the program. 

Sample: The 276,000 Nokia Life Tools subscribers in the Zhejiang, Heilongjiang, and Shaanxi provinces who completed the initial tobacco education program were eligible for the study. Of the 251,359 (91%) subscribers invited to participate, the first 8,000 (3%) who opted in were randomly split between the high- frequency and low-frequency groups, with 4,000 participants each. Non-response for the zero-, one-, three-, and six-month assessments ranged from 52%-59%. 


The authors first looked at the results using an intent-to-treat approach, which assumes that participants who chose not to receive messages after they were randomized and those who did not report their smoking status during the assessment continued to smoke. For overall quit rates, about 27%-30% of participants reported that they quit smoking at all the follow-up time points for both the high-frequency and low-frequency groups. No difference was observed between groups in the percentage of participants who quit smoking. The authors then looked at the results excluding participants who chose not to receive messages after randomization and those who had missing smoking status reports. Using this approach, participants in the high frequency group were more likely to quit smoking than the low frequency group immediately, at one month and at six months after the intervention. 

Limitations: One limitation of the study is the small number of variables collected during the assessments; information on smoking history and demographic characteristics could have provided more context and strengthened the analysis. The study also relied on self-reported smoking status (which tends to yield higher quit rates than biologically verified data) and had high opt-out and non-response rates. Lastly, the study design made it difficult to tease out the effects of text frequency and message content on smoking status. 


This study demonstrates that a smoking cessation intervention can be successfully adapted and delivered via text message to a large number of people in China, potentially helping to decrease smoking prevalence in the country. However, further research is needed to tease out the intervention attributes and assess their unique effects on smoking cessation and its long-term maintenance.