摘要：■ Aims In 2004, quotas on travelers' imports of alcoholic beverages from other EU member states for personal use were abolished, Estonia became a member of the EU and taxes on alcohol were decreased by 33 per cent on average. These changes resulted in a 10 per cent increase in alcohol consumption. This study set out to explore which population groups accounted for this increase and what happened to self-reported alcohol-related harm. ■ Data The data used in the article consists of a panel study where the same individuals completed a postal questionnaire before and after the changes. The questionnaire was mailed in autumn 2003 to a random sample of 4 000 Finnish persons aged 15–69 (response rate 60%) and in autumn 2004 to 1 786 persons who had completed the questionnaire in 2003 and who also agreed to participate in the follow-up (response rate 76%). ■ Results, Discussion & Conclusion A central thread in the results is that the respondents themselves reported no increase either in their own consumption or in alcohol-related harm from 2003 to 2004, although this was expected on the basis of the increase observed in sales statistics. One possible explanation for the contradiction between the statistical evidence and our respondents' self-reports is that the price cuts have led to an increase in consumption mainly among existing heavy consumers who were not covered in the questionnaire. This hypothesis is supported by reports from the treatment sector and by police records on intoxicated persons taken into custody. Another possible explanation is selective non-response, i.e. that heavier drinkers have dropped out more often than others. The non-response analyses indicated that part of the results that conflicted with our expectations was due to selective non-response. The results of our study do not warrant the conclusion that there will be no adverse consequences in terms of alcohol-related harm from the changes implemented in 2004. If consumption increases most of all among heavy consumers, that will lead to a greater increase in serious alcohol-related harm than would a growth of consumption among moderate drinkers. If consumption levels settle at a higher level, it is possible that harms will increase in the mainstream population as well.
