Statistics

Alcohol More Harmful Than Heroin Or Cocaine?

Belly up to the bar, boys. Have a snort! Reach into that pile of needles and raise them high. I’d like to propose a toast to Professors David Nutt [sic], Leslie King, and Lawrence Phillips for proving, via sophisticated statistical modeling, that heroin and cocaine are less harmful than alcohol.

Cheers! Sniff! Snort! Syringe Plunge! Followed quickly by a mix of manic and stuporous applause.

It is true: no drug is more harmful than alcohol. Nutt [sic] and his pals have proved it scientifically, submitted their proof to peer review, and published it in the Lancet, one of the top two medical journals in the world. Nothing beats peer review!

The article is Drug harms in the UK: a multicriteria decision analysis, which, if I read it correctly, means there is still a chance for us in the States, our frozen brothers to the North, you guys Down Under, and for our cousins elsewhere to find another drug more harmful than booze.

Nutt’s [sic] paper is science (right, Luis?). We know this because it appears in a science journal, has been reviewed by peers, has footnotes to other sciences papers, a table, plenty of scientific pictures, and prose thick enough to choke a whale. Just take a look at this excerpt and see if you do not marvel at its scienceness:

MCDA distinguishes between facts and value judgments about the facts. On the one hand, harm expresses a level of damage. Value, on the other hand, indicates how much that level of damage matters in a particular context. Because context can affect assessments of value, one set of criterion weights for a particular context might not be satisfactory for decision making in another context. It follows then, that two stages have to be considered. First, the added harm going from no harm to the level of harm represented by a score of 100 should be considered—ie, a straightforward assessment of a difference in harm.

And then there’s this picture. Look at all that science!

Nutt Alcohol is harmful

Alcohol has the highest harm score; ipso facto, alcohol is the most harmful. Yes, it’s even worse than LSD, tobacco, ecstasy, crack, and a slew of others.

By now you’re wondering how did science prove such an amazing result, one so far from the one expected by common sense? Why, through multicriteria decision analysis modeling (MCDA), of course. And what is that impressive scientific sounding process?

Well, Nutt [sic] and his co-authors invited some of his like-minded pals over for a drink (tea, presumably), and asked them, “Boys, what drugs cause the most harm to people and to society? What say we create a list and rank these drugs according to our subjective opinion? We should only consider harm and not benefit.”

Somebody in the crowd (almost certainly a beer drinker; an untrustworthy lot) must have asked, “Fine idea, Nutt [sic]. But what criteria do we use to make this rating? How about we score badness across several factors and then add those scores up to create a final badness score? Just to be sure we get the correct results, let’s also weight the individual factors to give some factors extra weight, some less.”

A third party (bottled water drinker) must have chimed in. “I like it. It sounds scientific. Some factors we should consider are injury, crime, environmental damage, loss of tangibles, and of course international damage, by which I mean deforestation.”

This proposal was surely put to a vote—just as the list of harmful drugs would be—and the motions were carried. The gang then elected a leader, who we are assured was “an independent specialist in decision analysis modelling. He applied methods and techniques that enable groups to work effectively as a team, enhancing their capability to perform, thereby improving the accuracy of individual judgments.” What could possibly go wrong?

When they finished, the cabal of voters realized they were on to something big. Their scientific findings were so shocking that they “correlate poorly with present UK drug classification, which is not based simply on considerations of harm.” Press releases would have to be written! Policies would have to change!

The results would be taken seriously because the language used to describe them was serious, because the people who created these results had credentials multitudinous, and because they would appear in a respected science journal. Heads would bow in awe.

————————————————————

Anybody who does think I summarized the paper unfairly is welcome to read it for themselves. I first learned of this paper from a link to the Daily Telegraph, which ran a story of Nutt’s [sic] achievement. A Telegraph reader called “Q46” was kind enough to link to my story on flawed medical research, and his link appeared in my logs. Thanks, Q46!

Categories: Statistics

39 replies »

  1. Briggs,

    “Anybody who does think I summarized the paper fairly is welcome to read it for themselves”

    Assuming that your intention wasn’t to suggest that if someone thought this piece insufficiently misconstrued they could take a shot at it themselves, maybe there’s a “not” needed.

  2. Years ago Consumer’s Reports had an analysis that showed that tobacco was more addictive than heroin. And it cited a case that showed long term heroin use of pharmaceutical grade heroin (he was a doctor) did not impair function or length of life.

    The same type hype was made over some studies by Bruce Ames which indicated grilled hamburgers and peanut butter were major cancer sources. Here is a link to one of his lists where beer is extremely high ranking in causing cancer. http://potency.berkeley.edu/text/maff.table3.html

    And here is a letter from Dr. Ames stating that “60 Minutes” distorted what he said http://www.fortfreedom.org/n16.htm.

    It is apparent that “scientists” have now taken the part of “60 minutes” in both climate gate and “harm” indexes.

    And isn’t it interesting that Al Gore is a failed seminarian and yet he is preaching typical “end of the world” stuff using semi-scientific rot and making a ton of money off of it.

  3. Nutt’s [sic] paper is science (right, Luis?)

    Just as calling Obama a muslim is science.
    Just as a bad restaurant is still a restaurant.
    Just as Justin Bieber’s album is music.

    Really, you are having some troubles with the obvious, mr Briggs.

    Science is what we make it to be. If we don’t make good science, it is nevertheless science. You find it ludicrous and completely asinine? Tough luck! Work harder! Ridicule the “scientific” basis of this paper. Etc.

  4. Don’t you just love this logic? Using the same logic one can show that knives as weapons are more harmful than any randomly selected WMD. How is it they can include butane (I think. Darn small print) but leave out petrol? Surely more crime is committed using petrol not to mention its involvement in traffic related deaths. But of course they are only interested in ingested substances. Why is crime one of the factors? Drugs (themselves) don’t cause crime, do they? Remove a given substance and the problems associated with it will evaporate.

    Briggs,

    It would appear they got you again. When did omitting something involve insertion?

  5. DAV,

    I’ll tell you how: by inserting blanks. So there!

    Luis, my old friend, we don’t do any of that “Obama/Muslim” nonsense here, and you know it. And I hardly ever name bad restaurants. Plus, I don’t know who Justin Beiber is: I’m guessing I’m thankful.

    The only point with which we agree “Science is what we make it to be.” How unfortunate.

  6. Wonder why they ignored deaths due to mis-prescribed prescription drugs? From what I read that is higher than most of these other items.

  7. I have to say that the paper has answered several questions that I had after reading this post. For examples, how is “harm” defined? Hmmm, after reading the paper, I would like to learn more about multicriteria decision analysis.

  8. Oh… What I like about the paper is that it addresses the shortcomings of the study.

    Limitations of this approach include the fact that we scored only harms. All drugs have some benefits to the user, at least initially, otherwise they would not be used, but this effect might attenuate over time with tolerance and withdrawal. Some drugs such as alcohol and tobacco have commercial benefits to society in terms of providing work and tax, which to some extent off set the harms and, although less easy to measure, is also true of production and dealing in illegal drugs. Many of the harms of drugs are affected by their availability and legal status, which varies across countries, so our results are not necessarily applicable to countries with very different legal and cultural attitudes to drugs. Ideally, a model needs to distinguish between the harms resulting directly from drug use and those resulting from the control system for that drug. Furthermore, they do not relate to drugs when used for prescription purposes. Other issues to explore further include building into the model an assessment of polydrug use, and the effect of different routes of ingestion, patterns of use, and context. Finally, we should note that a low score in our assessment does not mean the drug is not harmful, since all drugs can be harmful under specific circumstances.

  9. DAV’s nailed it. Harsh words are more harmful than atomic bombs, the common cold is more harmful than ebola etc.

  10. RE: JH

    And they left the door open to the grant gravy train:

    “Other issues to explore further include building into the model an assessment of polydrug use, and the effect of different routes of ingestion, patterns of use, and context.”

    Their funding came from this organization:

    http://www.crimeandjustice.org.uk/mem.html

    Who say:

    “We are honest. We strive to base our work on good evidence, presented honestly.”

  11. Briggs,

    Oh, it just keeps getting better…

    http://www.crimeandjustice.org.uk/wilkinsoninequality.html

    “What greater equality brings
    In societies where income differences between rich and poor are smaller, the statistics show that community life is stronger and more people feel they can trust others. There is also less violence – including lower homicide rates; health tends to be better and life expectancy is higher. In fact most of the problems related to relative deprivation are reduced: prison populations are smaller, teenage birth rates are lower, maths and literacy scores tend to be higher, and there is less obesity.”

  12. Luis, my old friend, we don’t do any of that “Obama/Muslim” nonsense here, and you know it.

    DAV, not my intention to say you ever did. It’s just one more example of “politics”. I also didn’t say you liked Justin Bieber.

  13. Wow, I’m really sorry, I messed up who said what. I was obviously talking to mr Briggs.

  14. Bruce,

    Is there anything statistics can’t prove? Equality!

    Great find, thanks. I wasn’t aware of who funded the study until your comment.

  15. Briggs, it is actually a little more complicated than pure statistics and science. You see, Nutt is designing his own alcohol substitute using benzos instead of grain. The grain can then be put to useful purposes like fuel for our cars.

    http://www.timesonline.co.uk/tol/life_and_style/health/article7061095.ece

    “David Nutt, chairman of the government’s Advisory Council on the Misuse of Drugs until he was dismissed last year, is developing his synthetic alcohol at Imperial College London.”

    So there is purpose behind this science. That makes it alright.

  16. I have not read the article. The approach on its face is silly and bizarre. Since there is objective data on mortalities, illnesses, injuries, car accidents how come he didn’t use it?

  17. I don’t see any scientific evidence presented in this screed by Mr. Briggs. On the contrary, he constantly uses false ad hominum attacks such as the snide misuse of the author’s name (Nutt [sic] ) and calling the researchers a ‘gang’ and ‘cabal’.

    As a refutation of a paper that is peer reviewed by scientists, attacks on the person, morals of the presenters (almost certainly a beer drinker, an untrustworthy sort), and ridicule (Sniff, Snort, Syringe plunge)–are appeals to the prejudices of like-minded thinkers (or perhaps unthinkers).

    Since non-scientific data such as the above are being employed by the self-proclaimed Statistician to the Stars, I will venture a few anecdotal items drawn from 50 years of life experience with users of alcohol, users of narcotics, cocaine and meth and various other hallucinogens.

    1. Lots more alcoholics die from the physical ailments of their habit than the total of cocaine, meth and marijuana users combined. I believe this is a statistical fact as indicated by the specific mortality factors attributed to alcohol above, but like Mr. Briggs I won’t bother to document my argument. Why use facts and statistics after all?

    2. Lots more innocent people die at the hands of alcoholics than from all the users of cocaine, meth and marijuana combined. Again, I present no empirical evidence, but lots of drunk driving deaths and bar and household murders are caused by alcohol.

    3. In my experience, more users of cocaine, meth and marijuana seem better able to function as productive members of society than alcoholics. Perhaps this is a function of the fact that there are more alcoholics than chronic abusers of other drugs.

    4. As a personal preference I’d rather be around someone who is happy and high on other substances than a sloppy, morose, antagonistic or crying drunk.

  18. Self-inflicted frontal lobotomies are even worse than that other stuff.

    PS to wild bill — your defense of junkies and crackheads makes me tired and paranoid.

  19. Wild Bill:
    If it is as you say, then the case should be readily made with objective data. As Wegman so pithily remarked: Answer Correct + Method Wrong = Bad Science

  20. Wild Bill,

    I write “Nutt [sic]” else some comedian—and here there are plenty—would accuse me of making the name up. As for the English word “gang”, I use it in the harmless sense, “Sally and the gang went to lunch and then shopping.” What were they shopping for? Synthehol, produced by Nutt’s cabal; previously thought only to exist in the fictional Star Trek universe.

    I’m glad you didn’t bother to produce any statistics yourself (see your #2). This way we can go back and forth and cry Nyah! nyah! nyah! at each other.

    You missed the joke, which was, in fact, a logical point. It is very likely that were these other drugs legal as alcohol is and has been, and customary and ubiquitous as alcohol is, than their death tolls would far exceed that of the grape and grain. And with alcohol (and perhaps with marijuana), more good than harm is had. I would rather be at a wedding party thrown by Jesus (water into wine! and the best wine!) than around sloppy drunks, too. Such a story cannot be told for heroin.

    Your fifty years of experience has been with people ingesting illicit substances. People who routinely flout the law to gain momentary pleasure—at the expense of a great deal of pain and cost—are people who do not think like those who eschew illegal substances. You have not seen a representative sample.

  21. wildbill2u,

    Some of society’s problems stems from willingness to shift blame to inanimates. The problem of addiction comes from within. Some drugs (perhaps all in one way or another) offer an initial reward that makes them attractive. Those who are prone to addiction will become so at nearly every opportunity. The actual path is incidental.

    You seem to have fallen into the some-abuse-it-so-no-one-should-have-it trap. Are you against sales of peanut butter because there are those to whom it is truly harmful? Should I ban oyster eating because for me to eat one is a near-death experience? Or is the use of alcohol for enjoyment that gets your dander up the most?

    And, no, at no time does alcohol use cause murders (assuming the murder wan’t over the alcohol itself). Believing so is fallacious reasoning. I consume alcohol and so do many that I personally know and have yet to hear of any of us committing a murder. Alcohol does reduce inhibitions, though. A mean drunk is really a mean person undisguised.

    “As a personal preference I’d rather be around someone who is happy and high on other substances than a sloppy, morose, antagonistic or crying drunk.”

    Amen to that.

  22. Question for wildbill:

    Since you seem to agree with the Lancet article’s authors that alcohol is a greater evil than these other drugs, does that suggest a particular public policy action to you?

    For instance, do you favor prohibition of alcohol?

    Or would you prefer decriminalization/legalization of the other drugs?

  23. Is it possible that more people die at the hands of people who take neither alcohol, nor any of the other substances indicated? And therefore the most dangerous people of all are the people who do none of these things?

    Did Professor Nutt (sic) include consideration of “none of the above” in his census of calamity causers?

  24. Matt,

    I dare say I give up my opinions by saying that I agree with you on drugs. Having had a very close family member engage in drug use, I know for a fact that coke is NOT good news.

    That being said, I suspect that pot as a legal drug would do less harm than it does as an illicit drug. This is based on my observations, not my statistical or otherwise academic analysis.

  25. Bruce Foutch,

    Darnit! You just poured cold water on the most arousing part of the paper! ^_^

    I don’t know if this kind of research can be lucrative, but I am inclined to think that it’s not. There are usually strict rules governing how grant money is to be spent.

    The authors of this paper give an honest evaluation of their own study, and they (sellers) actually tell the readers (buyers) to be aware of the problems. It could be that the authors simply don’t know how to resolve them but to address them as future research in the concluding section.

    So, such kind of concluding section is likable, don’t you think so?

    Always, honesty is the best policy in scientific research… and Confucius didn’t say this.

  26. When I was 14 my older sister married the nicest guy I had ever met. As it turns out he is still in the running as the all around nicest guy I had ever met 54 years later and 20 years after he passed away. However he was an alcoholic and when he drank he dad things we are all familiar with, that is he did what drunks do: drive their car, go to bed with women other then their wives, get into fights, spend time in jail, and spend all the food money on booze. My sister threw him out and he did not raise or contribute to the raising of his three boys. He was a good looking guy, a friend and likeable guy, a guy people enjoyed being with who had the bad luck to have a gene that made it impossible to resist alcohol. Alcohol probably contributed to his death at 49 and it clearly contributed to his sad and unfinished life. He wasn’t a bum or someone who deserved what he got. He was truely the nicest guy I ever met and if you were to have met him when he was 21 you would have felt the same way. I have seen many lives destroyed by alcohol (and drugs).
    The good news is his three boys, my nephews, are three great kids, I see their father in all of them especially the oldest. They only know the reality of what their father was (a drunk who wasn’t there for them). I am left with the reality of what their father could have been. When I go to lunch or hang out with his boys I am overwhelmed at the terrible waste of a life. He should have been here to see them grow up. His oldest boy is clearly the runner up and perhaps the winner for the nicest guy I have ever met. I am thankful for that at least. I never see booze or a drunken person without thinking of this story. I can not imagine anything that has negatively touched so many lives in the 20th century as much as alcohol has. Not the 45,000 car deaths a year. Not the deaths in war. Not even the murders and assaults we read about in the papers. Alcohol has destroyed more lives then anything else we have encountered in our lifetime.

  27. It sounds like they summed up the total harms to society for each drug, not per user or per use, but absolute. I imagine alcohol users outnumber other drug users by several orders of magnitude, so it wouldn’t surprise me in the least that they could find a way to show that alcohol is the “most harmful” drug out there. Even if crack kills all its users and alcohol only caused minor liver damage, if you have 1000 alcohol users for every crack head, it’s not a particularly large leap of reasoning to come to the conclusion that alcohol does more total damage.

    The fact that such methodology is disingenuous, misleading and often self serving should be pretty obvious to all but the most credulous people.

    w00t! I got a chance to use big words and sound all smartish! =)

Leave a Reply

Your email address will not be published. Required fields are marked *