More drinking stats

The WHO report on alcohol shows that Kiwis aren't that heavy of drinkers. Kiwiblog covered the highlights, so I hadn't hit it here. But there are a few things there yet worth noting.

First, the report tallies both official and unrecorded consumption, so the figures are a little bit higher than the official stats I tend to rely on here. The report doesn't explain particularly well how they record the unrecorded consumption. It doesn't matter a lot for New Zealand: they list average recorded consumption per capita aged 15+ at 9.3 litres, and unrecorded consumption at 1.6 litres, so the increase isn't that substantial. But here's the sum total of their description of the method, in Appendix IV:

An additional survey on unrecorded consumption was used to improve estimation of unrecorded consumption in some countries. In this survey, sent to 42 countries with at least 10% unrecorded consumption as part of the total consumption, the nominal group technique was used to solicit five expert judgements per country. In addition, a systematic search was conducted on all published literature (Rehm et al., 2014). The data obtained were analysed and fed back to the experts who used them to arrive at a final estimate.

So it sounds like expert guesses on consumption of home-made alcohol. It wouldn't surprise me if home production were increasing with excise, but does it seem plausible that home-brew has increased from an average of 0.5 litres [2003-2005] to an average of 1.6 litres [2008-2010] of pure alcohol per person aged 15+ in New Zealand? Say that 1% of us do any home brewing or home distillation: that would be 166 litres of home production per still or per brew kit. Seems high. If the beer is around .05, that's over three thousand litres per year per home brewer, if about 1% of the population are home brewing. Note that my 1% guess is based on a rough estimate of the number of people I know and the number of people I know who do any home brewing or distillation. Even if it's 10% of the population doing it, the numbers still seem implausibly high.

The Press notes that, per drinker rather than per capita, our consumption figures are higher, though they get the figure wrong: the WHO reports that we're at 13.7 litres pure alcohol per drinker. But they get this right:

But despite warnings about a binge-drinking culture, the figures show rates of "heavy episodic drinking" are well below many other countries.
Less than 6 per cent of the drinking population admitted to consuming at least six standard drinks on one occasion in the previous month.

Summarising from the WHO report:

• Per capita (aged 15+) consumption, in litres of pure alcohol, was 10.9 litres on average from 2008-2010. 
• Per drinker (aged 15+) consumption, in litres of pure alcohol, 2010, was 13.7 litres.
• 4.5% of the population aged 15+, and 5.6% of drinkers aged 15+, consumed at least 6 standard drinks on at least one occasion in the past thirty days. Six standard drinks (60 grams pure alcohol) is a six pack of 4% beer, 3 RTDs at 8%, or three-quarters of a bottle of wine. 

Sellman notes:

National Addiction Centre director Doug Sellman said the data showed the average New Zealander, of 15 years and over, was drinking about 16.5 standard drinks a week.

This suggested more than half were drinking in a risky manner, with health guidelines recommending no more than 15 standard drinks a week for men and 10 for women.

Alcohol consumption per head had increased by 13.5 per cent in New Zealand from 2005 to 2010, despite factors including a recession and an ageing population, he said. "This suggests the heavy drinking culture in New Zealand is being driven by strong factors which can overcome these."

These included "relentless" alcohol marketing, cheap availability, and a drinking age of 18, Sellman said. "As long as we have about 10 New Zealanders dying every week as a result of drunkenness, we deserve the unhealthy reputation of being a wild-south binge-drinking country."

I'll quibble a bit here. 10.9 litres does work out to 16.5 standard drinks per week on average, but consumption seems skewed: the median will be rather below the average in this kind of data, and especially where non-drinkers are included, so I'd be pretty reluctant to say that more than half are drinking in excess of health guidelines. I'd also be a bit nervous about relying on estimates that imply pretty substantial levels of home alcohol production based on "expert assessment" rather than inferred from things like distillation supply sales.

I also note that while total per capita consumption is up a bit from 2005 to 2010, it is substantially down on the 80s and early 90s, and consumption declined again after 2010.

If "relentless" alcohol marketing, cheap availability, and a drinking age of 18 are to blame for everything, why are youth drinking rates down? Why does the Productivity Commission note New Zealand's particularly high alcohol prices?

Again, here's the Ministry of Health's 2011/2012 survey findings:

• Since 2006/07, the level of hazardous drinking among past-year drinkers has significantly decreased for men (from 30% to 26%), but not among women (13% to 12%).

• People aged 18–24 years (particularly men) are at higher risk of hazardous drinking. Among past-year drinkers, about 44% of men and 26% of women aged 18–24 years have hazardous drinking patterns. However, the rate of hazardous drinking has decreased significantly in past-year drinkers aged 18–24 years from 2006/07 (49%) to 2011/12 (36%).

Emphasis added above. Prices haven't abated; marketing hasn't abated. The drinking age has stayed 18. But hazardous drinking has decreased, and has particularly decreased among youths aged 15-17. I'd summarised the broader stats here and here and here.

Update: Farrar seems right on our rank ordering: we're about the 96th heaviest drinkers when ranked on consumption per drinker. My spreadsheet gives me 97th, but the dataset had some strange country duplication. In 2003-2005, we were about 91st. So we're falling in the "heaviest drinker" league tables.