SoylentNews

Study: No "Safe" Level of Alcohol Consumption

posted by chromas on Saturday August 25 2018, @04:00PM   
from the this-news-is-terrible-and-I'm-gonna-need-a-drink dept.

takyon writes:

No alcohol safe to drink, global study confirms

A large new global study published in the Lancet has confirmed previous research which has shown that there is no safe level of alcohol consumption. The researchers admit moderate drinking may protect against heart disease but found that the risk of cancer and other diseases outweighs these protections. A study author said its findings were the most significant to date because of the range of factors considered.

The Global Burden of Disease [open, DOI: 10.1016/S0140-6736(18)31310-2] [DX] study looked at levels of alcohol use and its health effects in 195 countries, including the UK, between 1990 and 2016.

Analysing data from 15 to 95-year-olds, the researchers compared people who did not drink at all with those who had one alcoholic drink a day. They found that out of 100,000 non-drinkers, 914 would develop an alcohol-related health problem such as cancer or suffer an injury. But an extra four people would be affected if they drank one alcoholic drink a day. For people who had two alcoholic drinks a day, 63 more developed a condition within a year and for those who consumed five drinks every day, there was an increase of 338 people, who developed a health problem.

One of the study authors, Prof Sonia Saxena, a researcher at Imperial College London and a practising GP, said: "One drink a day does represent a small increased risk, but adjust that to the UK population as a whole and it represents a far bigger number, and most people are not drinking just one drink a day."

Related: The Truth We Won't Admit: Drinking is Healthy
Study Shows 3 Drinks a Day May Cause Liver Cancer
Even Moderate Drinking Linked to a Decline in Brain Health
American Society of Clinical Oncology: Alcohol Use Increases Risk of Cancer

---

Original Submission

• Questionable statistics (Score: 4, Informative) by gringer on Sunday August 26 2018, @09:39AM

  by gringer (962) on Sunday August 26 2018, @09:39AM (#726491)

  TLDR: too much uncertainty, and the average curves don't make sense.

  The relative risk graphs in the study are wonderful in that they show the individual data points (rather than just average lines), include individual uncertainties, and have the same X scale for all graphs; that's great stuff to do. They are a bit less wonderful in... most other areas:

  • X axis limit is too high; 5 drinks per day would be a better upper limit, because that seems to be where most of the data points finish
  • Y axis differs in all graphs
  • It is assumed that the uncertainty (of relative risks) for 0 drinks per day is 0 (not supported by the data points), and gradually increases up to the maximum drinks
  • The "total" graph has a kink at about 1 drink, and is essentially flat from that point to 0 drinks per day
  • The relative risk is relative to 0 drinks per day (which has high uncertainty due to lower sample sizes), rather than the population average number of drinks per day

  So, what about that average graph itself? Is it reasonable to say that, all things considered, there's a general trend for increased risk for any consumption at all?

  That depends on whether the average curves for each graph are being summed, or the individual data points. If the former, then there might be a case there. But those average curves have their own issues...

  • Most have a monotonic increasing (always going up) average curve, but the individual data points represented on the graphs suggest otherwise:
    • Atrial fibrulation and flutter
    • Epilepsy
    • Hemorrhagic stroke
    • Hypertensive heart disease
    • Interpersonal violence
    • Larynx cancer
    • Self-harm
    • Transport injuries
  • A few have obvious J-shaped relative risk curves (where the relative risk dips between zero drinks per day and one drink per day):
    • Diabetes mellitus
    • Ischaemic heart disease
    • Ischaemic stroke
  • There are only a few with data points that appear generally / consistently greater than 1 for relative risk:
    • Cirrhosis and other chronic liver diseases. This one is hard to tell, because the Y scale is too large.
    • Pharynx and nasopharynx cancer
    • Pancreatitis (again, Y scale too large)
    • Tuberculosis
    • Unintentional injuries
  • For the remainder (mostly cancer), most have a high uncertainty at the low end:
    • Breast cancer
    • Colon and rectum cancer
    • Esophageal cancer
    • Larynx cancer
    • Lip and oral cavity cancer
    • Liver cancer
    • Lower respiratory infections

  --
  Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]

  Starting Score:	1		point
  Moderation		+2
  Interesting=1, Informative=1, Total=2
  Extra 'Informative' Modifier		0
  Karma-Bonus Modifier		+1
  Total Score:		4