Smoking prevention, peer support and social networks: diffusion of the ASSIST programme beyond the school setting

Abstract

Background: A Stop Smoking in Schools Trial (ASSIST) is a social network intervention designed to disseminate non-smoking messages among secondary school children via student selected peer supporters. Message diffusion is restricted to the school year, with limited understanding of the extent and context of message diffusion in peer supporters’ wider social networks. The aim of this thesis was to address this gap by examining whether ASSIST had the potential to influence smoking behaviour, attitudes, and knowledge of smoking related harm, beyond the school year.

Method: Two literature reviews and a systematic review were conducted to inform the empirical data collection phase, which employed a qualitative social network design using a mixed methods analytical approach. A novel research method to collect data from young people (egocentric sociograms using coloured dots and post-it notes to prompt further discussion) and recruitment approach were piloted with 16 peer supporters from two secondary schools. Network data was quantitatively analysed using Excel and qualitative data from the interviews was analysed thematically using NVivo and Framework.

Results: The potential reach of message diffusion among the 16 peer supporter networks was large, with the combined number of people nominated being 151. The actual number of reported conversations was 103. Two-fifths (41%) of these conversations were with school friends and three-fifths (59%) with peer supporter family and friends outside of school. Thematic analysis of reported conversation content revealed three main types of conversation: protecting non-smokers from starting to smoke; encouraging smokers to stop; and protecting people from secondhand smoke in wider social networks. Peer supporters perceived impact with 37 people from their social networks, with examples of both positive and negative impact, focusing on the dynamics of a child speaking to an adult.

Conclusions: Smoking prevention message diffusion was not limited to the school year, reaching into the wider networks of peer supporters. Future interventions using the ASSIST model should consider changing the guidance that message diffusion should be limited to students in their year group only, as this may reduce the extent of message diffusion. Peer supporters should be better supported to think about whom they could speak to in their social network and the associated potential risks and ethical implications carefully considered.