Should alcohol policies aim to reduce total alcohol consumption? New analyses of Canadian drinking patterns

Abstract

We investigated whether high-risk drinking patterns are restricted to a few high-volume drinkers or are evenly distributed across the population to inform discussion regarding the optimal mix of targeted versus universal prevention strategies. Drinking patterns reported in the 2004 Canadian Addiction Survey (CAS, n = 13,909) were assessed against various low-risk drinking guidelines. Under-reporting was assessed against known alcohol sales for 2004. Non-response bias due to the low response rate (47%) was investigated through comparisons with the 2002 Canadian Community Health Survey (CCHS). Self-reported alcohol consumption for the past week and past year accounted for between 31.9% and 37.0%, respectively of official alcohol sales data. Comparisons with the 2002 CCHS suggested only limited non-response bias. Many more respondents regularly placed themselves at risk of short-term harm (20.6%) than exceeded guidelines for avoiding long-term health problems (3.9%). Ten percent of respondents consumed more than 50% of total self-reported consumption. Most alcohol (73.4%) consumed by the sample in the previous week was drunk in excess of Canadian low-risk drinking guidelines - for 19 to 24 year olds this figure was 89.4%. These data provide support both for universal prevention strategies (e.g. reducing economic and physical availability of alcohol) as well as targeted interventions for risky drinkers (e.g. screening and brief interventions in primary health care settings). © 2009 Informa Healthcare USA, Inc.