Writing down what you drink: a trial in Japanese workplaces

What we know

Even when people want to cut down, few remember exactly how much they drank this week. A Japanese research team (Sunami and colleagues, 2022) tested whether a simple web program that asks people to record their drinking every day would change how much they drink.

The participants were 100 workers from six workplaces in Japan who scored 8 or higher on the AUDIT, a standard questionnaire for problem drinking. A score of 8 or more is a common sign that someone’s drinking deserves attention. They were randomly split into two groups. Fifty people logged their daily drinking on the web for four weeks. The other fifty received nothing.

At week 12, compared with the group that did nothing, the logging group showed:

• less drinking in the previous seven days (effect size Hedges’ g = 0.53, a medium-sized effect)
• more alcohol-free days (g = 0.43)
• lower AUDIT scores (g = 0.44)

In weekly pure alcohol, the logging group went from 329 g to 246 g, while the comparison group went from 294 g to 276 g. The logging lasted four weeks, yet the difference was still there at week 12.

How to read it

This is a randomized controlled trial (RCT). Because chance decided who logged and who did not, the design lets us say that logging reduced drinking. It was also carried out in Japanese workplaces, so it speaks directly to people working in Japan.

Still, with 100 people it is a fairly small trial, and the true size of the effect could be larger or smaller. Drinking was self-reported, and only people whose drinking already called for attention took part. We do not know whether the same result would appear for light drinkers, or for people whose drinking needs medical treatment.

Why we do not link alcohol to income

UPSHIFT is a habit app about being on the side that raises your income. Even so, it treats “didn’t drink too much” as a health and self-tracking habit, not an income habit.

The reason is that research does not point in one direction. A study using the US General Social Survey (Peters and Stringham, 2006) found the opposite of what you might expect: people who drink earned 10 to 14% more than people who do not. Men who went to a bar at least once a month earned another 7% more. The authors explained this through social connections built over drinks.

That is a correlation. Other explanations fit just as well: people with higher pay can afford to go out more, or work in jobs with lots of socializing. It does not show that drinking pays off. In the same way, we have not found a study showing that cutting back on alcohol leads to higher income.

So the app does not tie your drinking to your income. What we can say is narrower: in a trial in Japanese workplaces, people who logged their drinking drank less and had more alcohol-free days.

A small step for today

• Log “didn’t drink too much” every day. Record days without alcohol and days you kept it light. Like the study, try it for four weeks first.
• On days you drink, note how many glasses. Writing “one beer, two highballs” lets you see your weekly total.
• Put alcohol-free days into your week. Choosing fixed days of the week makes them easier to keep.

If you find it hard to cut down, or feel uneasy without a drink, please don’t handle it alone. Talk to a doctor or a local support service.

Sources

• [B10] Sunami et al. (2022) A randomized controlled trial of the web-based drinking diary program for problem drinking in multi workplace settings. Journal of Occupational Health.
• [B9] Peters & Stringham (2006) No Booze? You May Lose: Why Drinkers Earn More Money Than Nondrinkers. Journal of Labor Research.