Polydrug use among nightclub patrons in a megacity: A latent class analysis.

BACKGROUND: Nightclubs are places with a high prevalence of binge drinking and illicit drug use. The aim of this study was to evaluate the characteristics of polydrug use, including licit and illicit drugs, among 2420 nightclub patrons in a probabilistic sample in the city of São Paulo, Brazil,
METHODS: The study was conducted in 2013. A latent class analysis (LCA) of polydrug use, accounting for binge drinking (BD) and other drug use (cannabis, cocaine, ecstasy, tobacco, ketamine, inhalants and hallucinogens) in the past 12 months was performed using Mplus. Multinomial logistic regression was used to evaluate latent class associations with sociodemographic characteristics and variables that characterise type of nightclub and frequency of attendance.
FINDINGS: A three-class LCA model best described polydrug use patterns. We found a "no polydrug use" class (55%), a "moderate polydrug use" class (35%) and a "high level polydrug use" class (10%). Compared to "no polydrug use", patrons in the two "polydrug use classes" were more likely to be men, young adults (<34 years), have attended nightclubs three times or more per month and have attended hip-hop and rock music nightclubs. Patrons in the "high level polydrug use" class were more likely to attend electronic (aOR=9.9, 95% CI: 5.4-8.1, p<0.001) and hip-hop music nightclubs (aOR=10.1, 95% CI: 6.2-16.5, p<0.001).
CONCLUSION: LCA is a useful method to identify groups of polydrug users among nightclub patrons. The three groups identified represented the diversity of patrons of São Paulo nightclubs. Frequency of attendance and the nightclub's musical style were highly correlated with polydrug use.