Commentary: Alcohol use disorder: as usual, prevention is better than cure.

In a very comprehensive review in this issue of the International Journal of Epidemiology, Roerecke and Rehm show a very high cause-specific mortality among alcohol use disorder (AUD) patients, especially regarding liver cirrhosis, mental disorders and injury. Several mortality estimates presented were higher than those previously found, and it appears that the newer (and better?) studies demonstrate higher risk estimates. In addition to the well-known consequences of a high alcohol intake (cirrhosis, mental disorders and some cancers), this study also reported a higher risk for heart disease in this population. This is in line with data from Russia, that show a strong relationship between high alcohol intake and high all-cause mortality, together with a high risk for cardiovascular disease (CVD) in heavy drinkers. It is, however, in contrast to many studies of general populations suggesting that a light to moderate intake of alcohol has a beneficial effect on coronary heart disease (CHD); the study does suggest that there is an upper limit for the beneficial effect of alcohol intake on coronary heart disease. In fact, this work by Roerecke and Rehm is so convincing that one might question the need stated in their own conclusion, that ‘there is a lack of research’. From my point of view, we need no further evidence before action is taken. We already have evidence from many large studies: from ecological studies in countries with high alcohol intakes and high frequency of cirrhosis, cancer, etc.; from case-control studies of the effect of alcohol on rare cancers; from large cohort studies of samples of the general population showing consequences of a heavy intake on a number of health outcomes; and now from this very thorough review—all pointing to the fact that alcohol in high doses is one of the largest killers in the Western world. Thus, if more research is needed, it is not research documenting the harms done by alcohol, but research into the interventions that may reduce the harm, or even better the interventions that may prevent us from beginning to drink hazardously, since prevention is always better than cure. All the studies included in the present review were cohorts of AUD patients. Although most of these presumably have had some treatment, many died too early compared with the background population. In this commentary it is not my aim to point out inadequate or insufficient methods of treatment, but rather to conclude that in these populations it was somehow too late. Of course, any treatment given may have been of some help: we may have observed more than 6420 deaths in this population of 28 087 patients, had they not been treated, but had they been included in the treatment earlier, or even better had their alcohol use disorder never developed, the numbers could have been lower. With regard to prevention, what measures could be taken? There are, of course, many reasons for people to drink heavily, but one of them is the easy availability of alcohol. Availability is a contributory cause for drinking too much, rather than a necessary cause. In some countries, such as the UK and Denmark, young people start to drink at a very early age. This has been shown to imply a higher intake later in life. In some countries, such as my own (Denmark), restrictions for selling alcohol to youth are not completely clear. A uniform age limit of, for example, 18