Alcohol use among adolescents in Europe is still rather high. On average, four in five students reported lifetime alcohol experience and every second student reported alcohol use in the last 30 days. Nevertheless, countries vary to a large extent in the prevalence of lifetime and current use. The Nordic countries Finland, Iceland, Norway and Sweden are traditionally among the countries with the lowest rates of current alcohol use. However, low rates can also be found in Albania, the Faroes, the former Yugoslav Republic of Macedonia, Ireland and Ukraine and in the Baltic states Estonia and Lithuania. In countries with low consumption rates, the prevalence of heavy episodic drinking is also generally low. Conversely, high alcohol use prevalence generally coincides with high rates of heavy episodic drinking. Among the countries with the highest rates are Austria, Bulgaria, Croatia, Denmark, Hungary, Liechtenstein and Monaco.

Beverage preference is rather differently spread across the ESPAD countries. Countries where beer accounts for more than 50 % of total alcohol consumption are Albania, Belgium (Flanders), the former Yugoslav Republic of Macedonia, Poland and Romania. In Croatia, Georgia, Moldova, Slovenia and Ukraine, wine accounts for at least 30 % of all alcohol consumed. In approximately every second ESPAD country, spirits make up the largest share of total ethanol consumption. In a small number of countries, other beverages such as alcopops or cider account for more than 20 % of total alcohol consumption. In Cyprus, Finland, Italy and Liechtenstein, alcopops account for one fifth or more, and in Denmark, Estonia, the Faroes, Ireland, Norway and Sweden, the share of cider is 20 % or more.

Evidence for the existence of associations between beverage preferences, drinking patterns, alcohol-related consequences and the use of other substances is scarce. In a study among Swiss young men, beer preference was associated with risky drinking patterns and illicit drug use (Dey et al., 2014), and a study on youths in the United States revealed that a preference for hard liquor and beer was associated with riskier patterns of drinking and other health-risk behaviours (Siegel et al., 2011). However, it is difficult to draw firm conclusions from the limited evidence available. Improved surveillance of alcoholic beverage preference, particularly with regard to spirits, may help to elucidate the factors related to youth drinking and the negative consequences associated with beverage types.

Despite the still high rates of alcohol use and, in particular, of heavy use, temporal trends over the past two decades indicate a positive development, with an overall decrease in lifetime and 30-day use between 1995 and 2015 from 89 % to 81 % and from 56 % to 47 %, respectively. Interestingly, both lifetime and 30-day prevalence have decreased markedly from a peak reached in 2003. Unfortunately, changes in heavy episodic drinking have been less pronounced and only observed among boys (42 % to 37 %), with overall rates declining by one percentage point (36 % to 35 %) over the past 20 years. The reported decline in weekly alcohol use among 15-year-olds in the HBSC study between 2002 and 2010 supports the present findings (de Looze et al., 2015). Although changes in alcohol use prevalence varied in magnitude, there are only a few countries with stable or increasing lifetime prevalence (Croatia, Cyprus, the Czech Republic, the Faroes, Hungary and Slovenia).

Factors that may have played some part in the general decrease observed in alcohol consumption include changes in norms on drinking and intoxication, competing responsibilities and attractions that demand or favour sobriety, structural changes, external influences and the range of societal or cultural responses to alcohol problems (Room et al., 2009). Others have argued that the observed change in adolescent drinking behaviour was due to changes in adult prevalence, shifts in teen culture or parental control (Ryan et al., 2010; Simons-Morton et al., 2009; van der Vorst et al., 2006). More recently, based on age, period and cohort analyses, results clearly indicate that younger cohorts reported abstinence more frequently and drank less than older cohorts (Härkönen and Mäkelä, 2011; Kraus et al., 2015; Meng et al., 2014; Pabst et al., 2010). In all western European countries, policies are in place to limit underage access to alcohol (Brand et al., 2007). In addition, stricter prevention policies are emerging in many countries (Anderson and Baumberg, 2006; Anderson et al., 2012). Some countries in the east of Europe deviate from the generally observed decline in adolescent alcohol use. These exceptions have been explained by rapid increases in wealth since 1990 and opportunities for adolescents to acquire and consume goods, including alcohol, that were previously unavailable or difficult to obtain (Zaborskis et al., 2006).

Our findings on trends in alcohol use suggest a closure of the gender gap in heavy alcohol use among adolescents in Europe and support earlier findings (Kuntsche et al., 2011; Simons Morton et al., 2009). The gender convergence, however, is more visible for heavy episodic drinking than for more regular drinking behaviours, such as monthly alcohol use.