1. Cannabis use and its socio- demographic correlates among in-school adolescents in Zambia Emmanuel Rudatsikira, MD, DrPH Dean & Professor School of Health Professions Andrews University

2. Outline Background Aim Methodology Results Discussion Conclusions

3. Background Cannabis dependence in adolescents predicts increased risks of using other illicit drugs, poor academic performance and psychotic symptoms. The prevalence of cannabis use was estimated two decades ago in Zambia among secondary school students

4. Aim To estimate the prevalence of cannabis use and its socio-demographic correlates among in- school adolescents in Zambia.

5. Methodology Secondary analysis of data from the Zambia Global School-based Health Survey (GSHS) GSHS uses a two-state probability sampling technique The outcome variable was lifetime cannabis use

6. Methodology Data analysis was performed using SPSS versions 14.1. A weighting factor was used in the analysis Frequencies were obtained as estimation of prevalence of the outcome variable

7. Methodology Proportions were compared using the Yates’ corrected Chi-square test at a 5% significance level Backward logistic regression analysis to estimate the association between relevant predictor variables and the

8. Ethical Considerations De-identified data that are in public domain with global authorization for re-use. Both the Ministries of Health and Education provided ethical oversight of the Zambia GSHS. Informed consent obtained from students and parents. Confidentiality was upheld through anonymity in completing the questionnaire

9. Results Overall, 37.2% of the respondents (34.5% of males and 39.4% of females) reported having ever used cannabis; p=0.037). In multivariate analysis, males were less to have ever used cannabis than females (OR=0.92; 95% CI [0.89, 0.92]).

10. Results Compared to adolescents 16 years or older, adolescents aged 14 years were 45% (AOR=1.45; CI [1.37, 1.55]) more likely to use cannabis. Other factors that were significantly associated with cannabis were history of having engaged in sexual intercourse (AOR=2.55; 95% CI [2.46, 264]), alcohol use (AOR=4.38; 95% CI [4.24, 4.53]), and having been bullied.

11. Results Adolescents who reported being supervised by parents during free time were less likely to have smoked cannabis (AOR=0.92; 95% CI [0.88, 0.95]).

12. Discussion Our prevalence is higher than the 26% reported in Nigeria and 16.9% reported in KwaZulu-Natal in South Africa, and much higher than the 6.2% reported in Zimbabwe. Differences in definitions of use of cannabis may partly explain the observed variations in the prevalence of cannabis use in Africa. Our finding that girls were more likely to smoke cannabis than boys was different from previous studies. It appears that it had become fashion for girls to smoke, thus surpassing the prevalence for boys.

13. Discussion Our finding that respondents aged 14 years or younger were more likely to have smoked cannabis than those who were 16 years or older is different from previous studies. The fact that those who reported cannabis use were also more likely to have used alcohol and engaged in sexual activity confirms the clustering of harmful lifestyles

14. Discussion The association between cannabis use and bullying victimization raises important issues. In large of study of 78,333 middle and high school students in USA, Rafliff et al (2012) observed a link between involvement in bullying and substance abuse.

15. Discussion The fact that adolescents who were supervised during their free time were less likely to use cannabis underscores the relationship between parental supervision and risk behaviors among adolescents.

16. Limitations Self-report No biomarker studies were carried out About 25% of the questionnaire did not respond to the questionnaire

17. Conclusion In this study of Zambian in-school adolescents, we found a lifetime prevalence of cannabis use of 37.2%. Parental supervision and being male were associated with lower likelihood of reporting cannabis use. This study confirms observation from other studies of the clustering of harmful lifestyles among adolescents.

18. Conclusions Interventions aimed at promoting health among adolescents may benefit from taking this findings into consideration.