Alcohol use is considered as one of the risk factors that contribute to social and physical harm and disease development globally. Alcohol use onset usually takes place long before adulthood and the surrounding environment has a central role in the development of alcohol use behaviour among children. Thus, targeting parents in prevention programmes has become more common in the past 20 years. This thesis aimed to investigate the relationship between parental attitudes towards children’s alcohol use and their children’s alcohol use and to evaluate a universal parent-oriented alcohol use prevention programme. A systematic review and meta-analysis (k=29 studies) of observational and experimental studies was conducted to assess the former. The findings showed that children whose parents had less restrictive attitudes towards children’s alcohol use had 1.45 (95% CI=1.17–1.80) times higher odds of initiating alcohol use, 1.52 (95% CI=1.24–1.86) times higher odds of drinking alcohol frequently, and 1.58 (95% CI=1.35–1.85) times higher odds of getting drunk. A similar relationship was present between perceived parental attitudes and children’s alcohol use frequency (OR=1.76, 95% CI=1.29–2.40), but no evidence was found for a relationship between perceived attitudes and alcohol use initiation (OR=1.65, 95% CI=0.93–2.94). When data from the Effekt trial was included, the change in pooled effect sizes was mostly marginal. The only exception involved the relationship between perceived parental attitudes and children’s alcohol use initiation, which became statistically significant (OR=1.55, 95% CI=1.14–2.11). Parental attitudes were only weakly positively correlated with perceived parental attitudes (r=0.27, p≤0.001). These findings confirmed and extended the earlier findings presented by Swedish researchers 20 years ago who developed a prevention programme (ÖPP, later called Effekt) to delay and reduce alcohol use among 13–16-year-old children by targeting parental attitudes. The programme was adapted and implemented in Estonia in 2012–2015 among the parents of 11–13-year-old children. The format (i.e. two meetings and two newsletters twice a year, agreements between parents) and the main messages focusing on parental attitudes, alcohol supply and communication were kept unchanged; but more emphasis was put on authoritative parenting, other risk behaviours besides alcohol use and active communication with parents. A matched-pair cluster-randomised controlled trial was carried out among children (n=985) and their parents (n=790) in 66 schools (34 intervention, 32 control) in 2012–2015. While no evidence for an intervention effect was found for students’ alcohol use initiation, past year use or drunkenness, the parents in the intervention group had double the odds (OR=1.92, 95% CI=1.31–2.83) of having restrictive attitudes by the end of the programme than the parents in the control group. A qualitative study at the end of the programme with the facilitators (n=8; focus group), teachers (n=12; individual interviews) and parents (n=24; individual interviews) indicated that the participants perceived the programme to have a possible effect on parents’ attitudes and behaviour, but were unsure as to whether this would have an impact on children’s drinking behaviour. One of the main perceived reasons behind reduced effectiveness was the issue of low participation rates at the meetings throughout the programme. Suggested reasons behind that included the general practice of not attending meetings in higher grades, the parents’ perception of the importance of their role decreasing as children get older, having enough knowledge on the topic, having minimal information about the programme before it started, perceiving that their parenting skills were questioned and having liberal views on the matter. In general, participants supported the idea of the long-term continuation of the programme; although they recommended changes in the content (e.g. reducing the repetitive content, including additional relevant topics, more practical examples, tailored content) and format (e.g. more emphasis on engaging teachers, including children). The findings of the thesis will provide valuable insights for the development of family-based programmes that target substance use prevention by focusing on a broader set of factors that influence behaviour and by conducting qualitative assessment in parallel with quantitative to better understand potential values and weaknesses and participants’ experiences and attitudes towards the programmes.