I believe it's the same study that was on here a few weeks ago, it's from brain scans + tests. 550 people over 30 years, same number, same years. Self-reported consumption.
The study is pretty detailed, but there are some oddities in the results, higher IQs drank more on average and lost more, people who drank heavily lost less than those who drank moderately.
This was based in the UK, in one community, middle-class, civil servants, predominantly men, average age 43 at start of study (and I would guess almost all white, though they don't say it).
Unsafe drinkers differed from safe drinkers by having a higher premorbid IQ, a higher percentage of men and smokers, and higher Framingham risk scores
Some of the interesting results:
Higher alcohol consumption over the study predicted faster decline on lexical fluency but not semantic fluency or word recall.
People drinking has a % greater reduction from their baseline in lexical fluency per year:
7-<14 units 0.5% (14% over 30 years)
14-<21 units 0.8% (17% over 30 years)
>21 units 0.6% per year (16% over 30 years)
!! These figures are odd and seemingly don't add up. Is it just bad rounding? Are their graphs based on these wrong figures? !!
Their conclusion doesn't highlight that it doesn't seem to matter how much you drink in terms of lexical fluency, it's that you drink at all?
I don't really understand the section titled "Grey matter". 3.4 odds in lighter drinkers, 5.8 odds in heavier drinkers? They don't really explain what that means.
Edit: having now listened to the audio it seems there was some technical problem with the white matter measurement, but they say they saw a decline.
Actual study: http://www.bmj.com/content/357/bmj.j2353
The study is pretty detailed, but there are some oddities in the results, higher IQs drank more on average and lost more, people who drank heavily lost less than those who drank moderately.
This was based in the UK, in one community, middle-class, civil servants, predominantly men, average age 43 at start of study (and I would guess almost all white, though they don't say it).
Unsafe drinkers differed from safe drinkers by having a higher premorbid IQ, a higher percentage of men and smokers, and higher Framingham risk scores
Some of the interesting results:
Higher alcohol consumption over the study predicted faster decline on lexical fluency but not semantic fluency or word recall.
People drinking has a % greater reduction from their baseline in lexical fluency per year:
!! These figures are odd and seemingly don't add up. Is it just bad rounding? Are their graphs based on these wrong figures? !!Their conclusion doesn't highlight that it doesn't seem to matter how much you drink in terms of lexical fluency, it's that you drink at all?
I don't really understand the section titled "Grey matter". 3.4 odds in lighter drinkers, 5.8 odds in heavier drinkers? They don't really explain what that means.
Edit: having now listened to the audio it seems there was some technical problem with the white matter measurement, but they say they saw a decline.
Previous discussion: https://news.ycombinator.com/item?id=14508201