Excessive alcohol use is a significant public health problem globally. Alcohol use typically begins in adolescence or early adult life, and effective prevention strategies focused on this age group are needed to avoid initiation of harmful drinking. The aims of this PhD are to understand the epidemiology of alcohol use and alcohol use disorders (AUD) in East Africa, to validate self-reported alcohol use among young people in Tanzania, and to use the findings to recommend strategies for the prevention or reduction of harmful alcohol use among young people. Specific objectives include (1) to systematically review the evidence on alcohol use among young people in eastern Africa; (2) to estimate the prevalence of alcohol use, AUD and associated risk factors among different groups of young-people in northern Tanzania; (3) to assess the validity of self-reported alcohol use against the blood biomarker phosphatidylethanol (PEth); and (4) to assess the validity of the Mini International Neuropsychiatric Interview questionnaire (MINI) for the diagnosis of alcohol dependence against PEth, in Mwanza (northern Tanzania). The main findings are a high prevalence of reported alcohol use among diverse groups of young people in eastern Africa, especially among university students and sex workers, but that few studies used standardised alcohol screening questionnaires. In our survey of almost 2000 young people in northern Tanzania, the prevalence of reported alcohol use was higher among males (20-45% for current use) than females (12-47% for current use). Alcohol use was also associated with being in a relationship, greater disposable income, and a higher number of sexual partners. There were significant positive correlations between reported total alcohol intake and PEth concentration in males (Spearman correlation (rs)=0.65 among college students and rs=0.57 among casual labourers; p<0.001). Self-reported alcohol use in the past month was also a sensitive marker of having a positive PEth result (overall sensitivity 89%, 95%CI 81-94%), and was similar in all groups. The MINI dependence criteria (positive responses to ≥3 questions) were met by 79/202 (39%) casual workers and college students. The high prevalence was mainly due to two non-specific questions (on tolerance and compulsion to drink). Both sensitivity and specificity of the MINI were low when compared to PEth, raising questions about the validity of the tool for this population. This thesis finds that alcohol use is a significant problem among young people in northern Tanzania, and the Timeline Followback calendar (TLFB) and Alcohol Use Disorders Identification Test (AUDIT) are sensitive measures of alcohol use. Most affected groups are college students and casual labourers. These two groups need urgent interventions addressing both individual and societal risks to reduce hazardous/harmful alcohol use.