Unravelling the web of addictions: A network analysis approach

Common elements across different forms of addiction suggest the possibility of comorbid addictions, as well as the transition/replacement of one form of addiction with another. This study aimed to conduct a Network analysis of symptoms of 10 forms of addictive behaviors to examine their behavioral commonalities/ interrelations. Methods: To address this aim, an online community sample of 968 adult participants (33.6% women, 66.4% men) completed self-rating questionnaires covering a range of addictive behaviors including alcohol, drugs, tobacco, sex, online gambling, internet use, internet gaming, social media use, shopping, and exercise. Their responses were examined with regularized partial correlation network analysis (EBICglasso) and a community detection algorithm (Walktrap) to identify: (a) specific links between neighboring forms of addiction; and (b) clustering of symptoms of addiction. Results: Findings showed positive network connections across different addictive behaviors, with addictive tendencies towards gambling showing the highest centrality, sequentially followed by addictive tendencies towards internet use, internet gaming, alcohol, shopping, social media use, drugs, sex, smoking, and exercise. Conclusion: Symptoms associated with disordered drug use and gambling are suggested to maintain severity of addictive disorders and increase the likelihood of developing cross addictive behaviors. Clinical implications for the assessment and treatment of addiction comorbidities and the replacement of one form of addiction with another are discussed considering these findings.