Mobile phone-based smoking cessation interventions increase long-term quit rates compared with control programmes, but effects of the interventions are heterogeneous
- 1Science of Research and Technology Branch, National Cancer Institute, Rockville, Maryland, USA
- 2Tobacco Control Research Branch, National Cancer Institute, Rockville, Maryland, USA
- Correspondence to: William Riley, PhD
Science of Research and Technology Branch, National Cancer Institute, 6130 Executive Blvd, MSC 7365, Rockville, MD 20852, USA;
Commentary on: Google Scholar
Implications for practice and research
Meta-analysis of five studies shows that smoking cessation delivered predominantly via automated text messaging increases quit rates 47–99% compared with minimal intervention.
Mobile phone cessation interventions provide quitting tips, social support and motivational messages, tailored to quit stage, multiple times per day at limited cost.
Although healthcare providers are moderately adherent with practice guidelines for asking, advising and assessing tobacco use, adherence to assisting and arranging cessation services remains poor, often due to limited time and training.1 ,2 Referral to quit lines has been an established means for providing these cessation services,3 but the growing ubiquity of mobile phones now provides the opportunity to deliver cessation services with exceptional reach to nearly everyone at any time. Automated text messaging and other computerised mobile phone features allow these services to be fully scalable with minimal human resource costs.
The authors performed a meta-analysis of five randomised or quasi-randomised studies of mobile phone-based interventions for smoking cessation. Of these, three studies used the same text messaging intervention which provides one or more automated text messages daily (greater frequency per day around the quit date) with content tailored to participant characteristics and quit stage (eg, preparation, action or maintenance). Messages include tips for quitting and motivational messages as well as some distraction/general interest messages. Most messages are one-way, ‘push’ messages but participants can also request additional messages on demand, and text a ‘quit buddy’. The interventions evaluated in the other two studies follow a similar text messaging format, and one also includes links to short videos modelling quit attempt strategies. The primary outcome of interest was an intent-to-treat prolonged abstinence at 6 months (ie, no smoking since quit date, allowing for up to three lapses or five cigarettes).
The five studies reviewed represent 9100 participants in New Zealand, Australia and the UK. Across the five studies, the risk ratio was 1.71 (95% CI 1.47 to 1.99) which indicates that the intervention participants had quit rates 71% greater than controls. One large, well-controlled study of 5792 smokers accounted for over half of the meta-analysis participants, and this study had a risk ratio of 2.14 (95% CI 1.74 to 2.63) based on biochemically validated quit rates (9.2% intervention vs 4.3% controls).4
This meta-analysis shows that smokers assigned to mobile phone interventions are significantly more likely than controls to remain abstinent at 6 months. Although encouraging, this finding is based on only five studies with heterogeneous outcomes. This heterogeneity highlights the need for additional research not only on the nature, frequency and timing of messages, but also on how these programme components interact with participant characteristics and clinical or policy contexts to produce cessation outcomes.
To isolate the effect of mobile phone interventions, this meta-analysis excluded two well-controlled studies included in the previous review because the intervention included both mobile phone and desktop internet components. In practice, however, mobile phone interventions are likely to be part of more comprehensive cessation programmes including non-mobile technologies, counselling and/or medications. Determining the additive effect of mobile phone interventions on current cessation services is a clinically relevant research question to pursue.
Delivering intervention content within 160 characters can be challenging. To assist in developing and evaluating cessation text messages, the National Cancer Institute has developed a freely available, open source library of core text messages and associated timing algorithms (QuitNowTXT message library).5
Since text messaging can be delivered to nearly any mobile phone, it has been the predominant modality for delivering cessation interventions, but as smartphone adoption increases, interventions will be able to leverage additional modalities and capabilities to further improve intervention delivery. Regardless of modality used, mobile phones provide a unique ability to deliver interventions throughout the day in the context of the behaviour. Once developed, these interventions also provide near infinite scalability at minimal cost and can continue indefinitely, cycling between initiation and maintenance as needed. Although a nascent field with numerous remaining research questions, the results of this meta-analysis show that mobile phone interventions are beneficial for smoking cessation.