Title: Is Sexual Content in New Media Linked to Sexual Risk Behaviour in Young People? A Systematic Review and Meta-Analysis
Publication: Sexual Health
Author: Smith, Lucy Watchirs, Bette Liu, Louisa Degenhardt, Juliet Richters, George Patton, Handan Wand, Donna Cross, et al
Background**: Social networking and digital media increasingly have an impact on the lives of young people. We undertook a systematic review and meta-analysis of studies that examined the relationship between exposure to sexually explicit websites (SEWs) and ‘sexting’ (i.e. sending semi-nude or nude photos from a mobile phone) and the sexual attitudes and practices of young people. *Methods*: In accordance with the Preferred Reporting Items for Systematic Reviews and Meta Analyses statement, Medline, EMBASE and PsycINFO were searched for papers that described the statistical association between viewing SEWs or sexting by young people (defined as 10–24 years) and their sexual attitudes and behaviours. *Results*: Fourteen studies, all cross-sectional in design, met the inclusion criteria. Six studies (10 352 participants) examined young people’s exposure to SEWs and eight (10 429 participants) examined sexting. There was substantial variation across studies in exposure and outcome definitions. Meta-analyses found that SEW exposure was correlated with condomless sexual intercourse (odds ratio (OR) 1.23, 95% confidence interval (CI): 1.08–1.38, two studies); sexting was correlated with ever having had sexual intercourse (OR 5.58, 95% CI: 4.46–6.71, five studies), recent sexual activity (OR 4.79, 95% CI: 3.55–6.04, two studies), alcohol and other drug use before sexual intercourse (OR 2.65, 95% CI: 1.99–3.32, two studies) and multiple recent sexual partners (OR 2.79, 95% CI: 1.95–3.63, two studies). Most studies had limited adjustment for important potential confounders. *Conclusions*:** Cross-sectional studies show a strong association between self-reported exposure to sexual content in new media and sexual behaviours in young people. Longitudinal studies would provide a greater opportunity to adjust for confounding, and better insight into the causal pathways underlying the observed associations.