Someone will have to parse the study to find out what's really happening. My guess is that most of the deaths that are being called alcohol-related are not chronic alcoholics; most are probably accidents and violence in which alcohol had been consumed at some point. So, I would be cautious about reading causation into those statistics.

There was a document published not too long ago about Auckland City Hospital, New Zealand which would possibly support what you're saying. A huge percentage of the emergency department admissions had been drinking. The below link is similar, but related only to accidents. The gist being that people admitted to hospital following an unintentional injury are often under the influence of alcohol. Small study, 2000ish participant, 20% ish screened for alcohol, 50% were under influence. http://journal.nzma.org.nz/journal/120-1249/2417/content.pdf