Web- and text-based interventions for smoking cessation: Meta-analysis and meta-regression

Background: In the past decade, several smoking cessation interventions have been developed and implemented through information and communication technology (ICT). Evidence suggests that they might be suitable for large-scale public health interventions, based on updated communication media characteristics in terms of interplay between technology and graphical user interface, reaching high numbers of individuals. Objectives: We aimed at estimating web/text-based interventions effectiveness as compared with approaches routinely used for general population, that is smoking assessment or non-electronic self-help materials. Methods: A systematic review and meta-analysis was performed searching through PubMed, Embase and PsycInfo databases, as well as references of relevant papers. Heterogeneity and risk of bias were evaluated following standard methods. In addition, we performed meta-regression analyses testing if candidate covariates moderate the overall effect. Results: Slight but significant effectiveness was found for eHealth interventions over control conditions (RR = 1.28, 95% CI: 1.14–1.45). Meta-regressions showed similar findings for web- and text-based interventions. The effect seemed moderated by the follow-up period, being higher at 3 months and lower at 6/7 months follow-up. Conclusions: Our results outline moderate effectiveness of web/text-based interventions. However, a paucity of properly controlled studies and lack of information on several effect modifiers still hamper the development and implementation of smoking cessation interventions through ICT