This feels like it was written by a temperance researchers who is manipulating his result. I understand drinking's a problem for a subset of our population and results in early deaths, but they've clearly identified it's the subset of heavy drinker but constantly preach to the people who aren't in that subset.
In this instance, he lied twice.
First the article content doesn't support the title's claim at all.
In idiomatic english, something being "not so good for you" implies it's bad for you. For example the synonyms "not so good" on thesaurus.com are "bad" & "poor".
That errant "so", obviously deliberately chosen, is a lie.
They didn't find that at all. They're actually found moderate drinking has no discernible consequence on health and only heavy drinkers are at risk.
But even that is a lie, that they admit themselves. There's this gem:
In women, moderate alcohol use was still linked with a health benefit, but given the few women drinking moderately (4%) this effect was questionable.
So the researcher has actually decide to ignore his own results! Either it's mathematically significant or not significant. You can't just discard results that don't agree with your temperance stance.