Purpose: The purpose of the paper is to show that a syndrome of problem behaviours, i.e. early substance abuse, school and family problems and sexual promiscuity impairs normal development in adolescence. This comparative study looked for differences in the problem behaviour profiles of 15-year-old adolescents in the Pitkaranta district in Russia and in eastern Finland, in order to develop more effective strategies for adolescents' health promotion. Design/methodology/approach: Data from the Russian Pitkaranta Youth Study (n=385) and the Finnish North Karelia Youth Study (n=2098) were used. A K-means clustering algorithm was used to identify homogenous groups of adolescents based on variation in selected health behaviour variables. Findings: The paper finds that four different profiles including the variables of current smoking, first smoking experiments, first drinking experiments, experiences of drunkenness and sexual experiences were identified. The identified profiles, titled "Non- or late experimenters", "Middle experimenters", "Early experimenters" and "Child experimenters", were found to be distinct across gender, country and other external variables. Adolescents more often in Pitkaranta than in eastern Finland belonged to "Non- or late experimenters" of minimal risk-taking behaviours. Unhealthy dietary habits, use of illegal drugs, psychosomatic disorders and problems with parents were more common among "Early experimenters" and "Child experimenters". These findings added to the evidence of grouping of problem behaviours. Research limitations/implications: The paper shows that there is a need to develop and implement tailored and coordinated health promotion programs for specific target groups of adolescents. Obviously, adolescents with a high level of risk-taking behaviours would benefit from programs that acknowledge their cultural expectations in their everyday life contexts. Practical implications: This paper describes a cross-sectional comparison of health surveys among adolescents in two countries. It will be interesting to carry out a follow-up survey in, for example, ten years to see how health issues have changed, especially among Russian adolescents, of whom there is not much research available. Originality/value: Cluster analysis was a useful method in identifying adolescents' problem behaviours in a cross-cultural study. (Contains 2 tables and 1 figure.)