E-cigarette, cannabis, hookah and tobacco use patterns in fee-paying South African high schools

BACKGROUND: Monitoring adolescent substance use is crucial for informing public health strategies. However, in South Africa (SA), recent large-sample data on the use and co-use of tobacco, nicotine and cannabis among youth remain scarce. OBJECTIVES: To describe the use and co-use of cannabis, hookah, tobacco cigarettes and electronic cigarettes (e-cigarettes/vapes) among SA high-school learners, and to examine how these patterns vary by school-based characteristics, including grade, school fee category and school gender composition. METHODS: A cross-sectional survey was administered to 25 149 learners in grades 8 - 12 from 52 fee-paying high schools across eight provinces. Learners reported their past-30-day use of cannabis, hookah, tobacco cigarettes and e-cigarettes/vapes. Key outcomes included current use of each individual product, any one of the four products, as well as dual-usage patterns. Multilevel logistic regressions examined associations between school grade, fee category (lower-, mid-, or high-fee), gender composition (co-educational, all boys, all girls) and the odds of single, any and dual product use. RESULTS: Among sampled learners, 19.39% (95% confidence interval (CI) 18.91 - 19.88) reported current use of any product. Vape use was most prevalent (16.83%, 95% CI 16.37 - 17.30), followed by cannabis (5.13%, 95% CI 4.86 - 5.41), hookah (3.16%, 95% CI 2.95 - 3.39) and tobacco cigarettes (2.08%, 95% CI 1.91 - 2.27). Dual use was especially common among vape users, with more than one-third (34.31%, 95% CI 32.88 - 35.77) reporting concurrent use of at least one other product: 22.06% (95% CI 20.83 - 23.35) cannabis, 13.50% (95% CI 12.50 - 14.58) hookah and 10.13% (95% CI 9.25 - 11.08) tobacco cigarettes. Usage rates were highest among learners in Grade 12, in co-educational schools and in lower-fee schools. Multivariable regression analyses showed that advanced grade level was significantly associated with increased odds of current use across all product types. Compared with learners in all-boys schools, those in co-educational schools had significantly higher odds of cannabis use (odds ratio (OR) 1.53, p