Binge drinking at University: a social network study in Belgium

Health Promotion International, Vol. 30 No. 3 doi:10.1093/heapro/dau007 Advance Access published 12 March, 2014 # The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: Binge drinking at University: a social network study in Belgium VINCENT LORANT* and PABLO NICAISE Institute of Health and Society, UCLouvain, Clos Chapelle aux Champs 30 /B1.30.15.05, Brussels 1200, Belgium *Corresponding author. E-mail: SUMMARY Many university students engage in risky alcohol consumption behaviour during their stay at university. So far, however, most studies have relied on cross-sectional surveys and paid little attention to the role of social ties. University students, however, are socially connected, so it is likely that their alcohol consumption behaviour is also connected. We hypothesized that university students’ social positions within their networks are related to their drinking behaviour. We carried out a social network analysis within a whole network approach with undergraduates in two faculties (n ¼ 487), those of Engineering and Psychology, in a Belgian university. All students filled out a questionnaire recording their drinking behaviour and their social ties (friendship, working with, partying with and room-mate). For each individual, indicators of centrality, social capital, and crossgender relationships were computed. We found that being socially close to binge drinkers was associated with a higher frequency of binge drinking. The risk of binge drinking increased with centrality but decreased with social capital. Having cross-gender relationships decreased the risk of binge drinking. We found indications that the effect of centrality and gender on binge drinking depends on the composition of the network. We conclude that social position has important effects on risky drinking behaviour and that the composition of the network may affect these factors. Those developing health promotion strategies could investigate the benefits of targeting central individuals in order to prevent binge drinking among university students. Key words: alcohol drinking; school health; social network analysis INTRODUCTION Excessive alcohol consumption accounts for an important share of morbidity and mortality among teenagers and young adults (Rehm et al., 2009). The university is a key player in this issue: an increasing percentage of younger people are heading to university after secondary school (Organisation for Economic Co-operation and Development and Centre for Educational Research and Innovation, 2009) and this transition is often associated with more frequent and risky drinking behaviour (Schulenberg and Maggs, 2002). In Europe, the consumption of alcohol among teenagers is a rather recent topic of research. In recent research, more than 40% of students aged 17– 30 in several European countries reported having drunk heavily (Dantzer et al., 2006). Risky drinking has also been found to be a common practice (Stock et al., 2009). Although some of this behaviour may become less frequent as teenagers become adults, cohort studies suggest that excessive alcohol behaviour, including problematic alcohol consumption, may persist into adulthood, leading to a higher risk of dependence (McCambridge et al., 2011). Thus, reducing alcohol drinking during adolescence and young adulthood is important for preventing long-term adverse consequences in adulthood. Programmes to limit excessive alcohol drinking have relied on a wide range of preventive 675 676 V. Lorant and P. Nicaise strategies, including educational or school approaches, community-based programmes and treatment within the health sector, regulations on drinking, driving and advertising, and policies affecting alcohol availability or price. Policies influencing the availability and the regulation of alcohol marketing are more effective than information or educational approaches (Anderson et al., 2009). A meta-analysis of individual-based approaches confirmed that their effectiveness in terms of reducing the quantity of drinking is small, with an average effect size of ,0.20 of a standard deviation (Carey et al., 2007). This has led to something of a shift towards more community-based educational approaches (Saltz et al., 2010) rather than individual-based ones. More recently, the cutting edge of community health policy has relied on approaches that use social norms (Pischke et al., 2012) and social ties as the main preventive vehicle (Valente, 2010), with promising results (Webel et al., 2010). To date, alcohol consumption has been analysed from an individual perspective, with the main focus on individual risk factors such as gender, age, socio-economic status, personality, psychological factors and drinking motives. Alcohol consumption among university students, however, takes place in a very specific social environment that includes independent living, reduced social control, increased social homogeneity, wide availability of social activities such as pre-gaming (i.e. drinking alcohol prior to going out in order to prime oneself for the social occasion ahead) (Read et al., 2010), drinking games (Borsari et al., 2003) and other student recreational activities. These social activities and context affect alcohol consumption, partly through norms, as the perception of the drinking norm has been shown to be a potent predictor of actual drinking in the USA (Borsari and Carey, 2003; Perkins et al., 2005) and in Europe (Lorant et al., 2013). Because of this social dimension, several studies have investigated the role of peers in alcohol-drinking behaviour (Delk and Meilman, 1996; D’Alessio et al., 2006; Keller et al., 2007; Mcalaney and McMahon, 2007; McMahon et al., 2007). However, most of these studies have relied on cross-sectional surveys in which one’s own drinking and peer drinking behaviour are reported by the same individual (Alva, 1998; Rose, 1999; Durkin et al., 2005). A couple of recent social network surveys have investigated the role of ties among the adult population (Rosenquist et al., 2010) or at high school (Ennett et al., 2006) in the USA. These studies have shown that alcohol use increases when people’s (best) friends drink more alcohol. With the exception of one ego network study (Reifman et al., 2006), there is a clear paucity of work about the influence of social networks on drinking habits among university students. Our study applied Social Network Analysis (SNA) to the study of risky drinking behaviour among university students. We set out to analyse the role of peers and of social position within a university network in drinking behaviour. In particular, we hypothesized that social position in the network affects drinking behaviour and that cross-gender relationship is an important component of this drinking-network effect. METHOD Design and data source This study is part of an important multi-method investigation into alcohol drinking among university (...truncated)