Objective: A key assumption in Finnish alcohol policy is that the officially registered alcohol consumption (i.e., alcohol sales) is closely related to alcohol-related harm. During the last two decades, a sizable part of total alcohol consumption, however, comprises unrecorded consumption, which may potentially make alcohol sales less powerful as a predictor of alcohol-related harm. This article thus aims to estimate the relationship between alcohol sales and alcohol-related harm on the basis of more recent Finnish time-series data.
Method: Data on alcohol sales (liters of 100% alcohol/capita age 15 years and older) were obtained from the National Institute for Health and Welfare in Finland. As indicators of harm, we used police-reported assaults and three forms of mortality: alcohol-specific mortality, accidents, and suicide. Quarterly data on mortality and alcohol sales spanned the period 1995-2016, and data on police-reported offenses covered the period 1990-2016. Data were analyzed by SARIMA (Seasonal Autoregressive Integrated Moving Average) modeling.
Results: A positive and significant association between alcohol sales and all harm indicators was found. A 1-L increase in alcohol sales per capita was associated with a 20% increase in alcohol-specific mortality, a 12% increase in assaults, and a 5%-6% increase in accidents and suicide. These estimates are in line with earlier findings estimated on data for the period when unrecorded alcohol consumption was less common in Finland.
Conclusions: The results provide support for a continued strong relationship between alcohol sales and alcohol-related harm in Finland. Policy measures aimed at lowering alcohol sales were supported from these results.