摘要：Aims The aim of the current study was to test empirically two predictions from Skog's theory of collectivity of drinking behavior, using time series data from Norwegian adolescents. The two specific predictions were: 1) A change in mean alcohol consumption is positively associated with a change in the proportion of heavy drinkers, and 2) A change in mean alcohol consumption is positively associated with a change at all consumption levels. Data & Methods The present analyses are based on ESPAD data collected from Norwegian adolescents (15-16 years) in 1995, 1999, 2003, 2007 and 2011. The relationship between mean consumption and the proportion of heavy drinkers was analyzed by regressing the proportions of heavy drinkers at each time point on the consumption means at each time point. In order to assess whether adolescents at all consumption levels, from light to heavy drinkers, changed collectively as mean consumption changed, we regressed log-transformed consumption means on the log-transformed percentile values (P25, P50, P75, P90 and P95). The analysis was restricted to adolescents who had consumed alcohol in the last 30 days (total n = 7554). Results The results showed a strong relationship between mean alcohol consumption and the proportion of heavy drinkers. An increase in mean consumption was also associated with an increase at all consumption levels, from light to heavy drinkers. Conclusion The results of the current study are in line with the theory of collectivity of drinking behavior. The findings of this study suggest that by reducing the total consumption of alcohol among adolescents, consumption and risk of harm may be reduced in all consumer groups.
