INTRODUCTION AND AIMS:Alcohol consumption among Swedish adolescents has halved during the last decade. We aim to: (i) investigate whether the overall decrease in drinking may conceal an underlying heterogeneity in drinking trajectories across at-risk groups that differ with respect to risk for drinking and; (ii) assess to what degree alcohol-related harm has responded to this decrease. DESIGN AND METHODS:Data were obtained from the nationally representative annual school survey of alcohol and drug habits among Swedish ninth-grade students covering the period 2000-2012 (n ≈ 5000/year). Respondents were divided into five at-risk groups ranging from low to high based on their relative ranking on a risk scale for drinking. Alcohol consumption was measured by beverage-specific quantity and frequency items summarised into a measure of overall drinking in litres of 100% alcohol per year. Alcohol-related harm was measured by eight items asking about whether the respondent had experienced various alcohol-related negative consequences. RESULTS:Drinking and alcohol-related harm decreased in all five at-risk groups. There was a marked relation between the overall consumption and the mean consumption in each of the five at-risk groups. Self-reported alcohol-related harm decreased during the study period to an extent that was expected from the decrease in alcohol consumption. DISCUSSION AND CONCLUSIONS:Alcohol consumption among Swedish youth has declined in five groups that were delineated based on their relative ranking on a risk factor index. The findings are consistent with Skog's theory of the collectivity of drinking behaviour. [Norström T, Raninen J. Drinking trajectories of at-risk groups: Does the theory of the collectivity of drinking apply?.