Aim: The aim of this master thesis was to explore contexts of initiation of substance use and to describe the social, physical and mental health situation of women injecting drug users (WIDU) in Tunisia. Relevance: In North Africa, people who inject drugs are an important risk group in the HIV epidemic. There is evidence that WIDU are even more vulnerable. However, in Arabic Muslim societies, rehabilitation, treatment for this subpopulation and research on this issue are scarce. Method: A qualitative study was conducted. Six in-depth interviews with WIDU in Tunis were recorded, transcribed and inductivelyanalyzed using thematic analysis. Findings: Four themes were identified. The first theme describes situation of women before dependence. Women were brought up within patriarchal cultural norms where they were victimized. The second theme is related to the circumstances of initiation of drug use characterized by the wish for escape and pursue of pleasures. The third theme describes the dependence, withdrawal and consequences on everyday life. Final and fourth theme is related to the way out of addiction. Conclusion and recommendations: The socio-ecological model and the gender relational theory helped to interpret the findings.WIDU in Tunisia suffer from marginalization and from social and health inequalities due to their gender and to their dependence. This puts them at a higher risk of violence, abuse, health hazards and blood-borne and HIV infections. Thus, awareness and policies should be designed in order to alleviate the stigma and bring services closer to this subpopulation.

Men who have sex with men (MSM) account for a majority of all men currently diagnosed with AIDS. MSM is also recognized as the largest risk category of all AIDS cases. Drug use has been shown to have a synergistic effect on the prevalence of HIV in the MSM population. The study aimed to examine the association between injection drug use, non-injection drug use, and non-drug use with sexual risk behaviors for HIV in men who have sex with men. Secondary, cross-sectional data procured from the National HIV Behavioral Surveillance System for the Atlanta Metropolitan Survey Area were used for the study. The study population was 960 participants. Using binary logistic regression analyses, the drug use categories were studied for unprotected intercourse, unprotected receptive anal intercourse and HIV status. Strong associations were seen independently for unprotected intercourse, unprotected receptive anal intercourse, and HIV status with injection and non injection drug use, but the association weakened for drug use categories when controlled for other independent factors. Taking into account current findings and findings from previous research, the importance of clinical significance over statistical significance was considered. Racial disparities were evident, in that, although the Black participants showed no increased odds for sexual risk factors or drug use, it had a higher odds for being HIV positive compared to Whites. Combining Viagra/Levitra with drugs was significantly associated with unprotected intercourse (AOR=1.9), and each individual drug showed a different degree of association with Viagra/Levitra. Further research is recommended to identify sub-populations at risk and appropriately allocate resources and channel programs and interventions.