Drug harms paper: a summary

December 9, 2010

Drug harms in the UK: A multi-criteria decision analysis

The ISCD’s first published work, launched in the Lancet on 1 November 2010, has attracted considerable attention in the media worldwide and from the public. The summary below lays out the basics of what the papers shows and what it cannot. To read the paper in full, download it for free from the Lancet’s website.

Overview of paper
An assessment of the various harms of drugs used recreationally in the UK using multi-criteria decision analysis (MCDA) – a method that uses relevant experts’ knowledge and experience to assess the actual and relative harms.

What is its purpose?
To assess drugs in terms of their known harms to individuals, those around them and more broadly, to domestic and international society. Showing drugs’ harms relative to each other also enables more objective and informed assessment of their harms individually and in relation to their current controls.

What does it show?
The report found heroin, crack cocaine, and methyamphetamine to be the most harmful drugs to individuals (part scores 34, 37, and 32, respectively), with alcohol, heroin, and crack cocaine as the most harmful to others (46, 21, and 17, respectively). Overall, alcohol was the most harmful drug (overall harm score 72), with heroin (55) and crack cocaine (54) in second and third places.
It also found the legal status of most drugs to bear little relation to their harms.

What does it not show?
Because of the many factors that could change, the report does not take the legal status of a drug into account. This will clearly have some impact on its level of harms, for instance alcohol’s regulated availability means that many more people have access to it than most drugs, which will have an effect on its level of harm. Similarly, the controlled status of some drugs will make them more harmful as unsafe production of the drug introduces contaminants and risky practices by users puts them at risk of diseases such as Hepatitis C and HIV.

The rankings of harm to the user  would be unlikely to change drastically if all drugs were controlled or, conversely, not controlled, as those harms are intrinsically linked to their effects on the mind and body.  However, the harms to society could change if a drug’s legal status changed. There are examples in a number of countries of different levels of control, such as in Brazil, Mexico, Portugal, Spain, the Netherlands, Finland and parts of Australia and the USA, amongst others. However, it is not possible to expect identical results if replicated in the UK, given the wide variety of cultural and historical factors that can influence drug use and especially problem drug use.

Areas of development
This paper represents a step forward in our understanding of drug harms but is by no means the definitive answer. Repetition of the MCDA process with other relevant groups (such as police, teachers, doctors, politicians, users) and in other countries would help to build a fuller picture of the range of harms. Taking into account the various benefits of drugs would also enable better understanding of why people take drugs and how they might positively impact on users and society.


18 Responses to “Drug harms paper: a summary”

  1. John Ellis Says:

    Dave you know this! We campaigns know this! Unfortunately the bankers in charge still believe in a flat world.

    What do you make of the success of cannabis in the EU now that it has been given the go ahead to be taxed and regulated by euro parliament, does our government really have any ground left to stand on with this dogmatic archaic policy they now have on the table.?

  2. Rich and Co. Says:

    Gee, we didn’t know that the expectation of a “…definitive answer…” applied to any science or advanced area of knowledge.

    This is great work. Imperfect, of course. Hats off to the British. Wish the US would do more of this kind of, (mostly) ideology free, public social research. Likely it will not happen.

    We adhere to Feynamann’s view: “All science is wrong, some is just less wrong.” than other ways to make sense of experience.

  3. Mafficker Says:

    Thank you Sir for your efforts. Is it true that the ACMD chose the 16 harm criteria and also the MCDA process?

  4. Although you say legal status does not contribute much in the way of harm, you do also make it clear the control method – whether prevalence of alcohol due to law or the prohibition of others substances makes a difference – and this for many is the key factor in trying to draw attention to the actuality of what is going on; those who have a distant and blinkered outlook on this debate fail to address the issue of substances and drugs correctly.

    There really is a swings and roundabouts situation with trying to address prohibition on drugs vs alcohol prevalence and law. This is where the debate tends to break down into nonsense as you know first hand Professor. Are there progressive ways to address this I wonder?

    Thank you for continued work .

  5. > other relevant groups (such as police, teachers, doctors, politicians, users)

    Applauding the core of the initiative to get to grips with what is a drug harm and or benefit requires us to be very careful on who we engage with as the master status of the likes of MD’s and PC’s tends to take public and media precedence. On the question of what is a user, again great care has to be take to establish what we are taking about here. There is a vast difference between an otherwise productive law abiding (and largely invisible) cannabis consumer who mediates an alcohol dependency, offsets ‘that aching joint’ or elevates mood (sense of wellness) and a person who is, for want of a better description regailing ‘pot’ for destroying his/her life (be it true or otherwise) and quite probably captured by the treatment industry.

    I would commend the Tavistock Method (or its US equivalent the name of which escapes me for the moment) as a way of integrating the group dialog (and prevalence Jason) to ensure we capture the benefits and are thus better able to qualify (and thus quantify) the net benefits/deficits of current and proposed options.

    As I see it, national drug policy(s) as defined under UN conventions has never faced a cost benefit analysis or policy impact analysis which is mandatory in other spheres of civil governance.

    For a fraction of what has been spent politicising the Prop19 debate in California – and logically an infinitesimal amount currently frittered away on enforcement, this absolutely necessary process could be adequately resourced. Absent appropriate resourcing is to short change both due process and Ottawa Charter principals.

    An example of how flawed analysis can be trotted out as useful… can be seen in the New Zealand BERL report on drug harms (paid for unsurprisingly by Police).

    How we capture the cost of the war on drugs, and inparticular cannabis’s identified ‘impediments to effective health promotion’ I will leave to experts.

  6. […] This post was mentioned on Twitter by DrugScope, Rebecca Daddow. Rebecca Daddow said: RT @DrugScope: Drug harms paper: a summary [David Nutt's Blog, UK] http://bit.ly/ghFrv6 […]

  7. […] Drug harms in the UK: A multi-criteria decision analysis The ISCD's first published work, launched in the Lancet on 1 November 2010, has attracted considerable attention in the media worldwide and from the public. The summary below lays out the basics of what the papers shows and what it cannot. To read the paper in full, download it for free from the Lancet's website. Overview of paper An assessment of the various harms of drugs used recreationa … Read More […]

  8. Will Mathieson Says:

    I think you are doing an excellent job, Professor Nutt.Your findings confirm what I had already observed for myself. I hope you will not be discouraged by irrational opposition but continue to publish the truth as you find it. Remember, decent people knew slavery was wrong from the beginning, but it took hundreds of years to end it.
    It’s such a pity no-one stands to make a profit from the truth.

  9. sudon't Says:

    The analysis purports to tally the “harms caused by the misuse of drugs” when, in most cases, what it shows are the harms caused by prohibition. That “the report does not take the legal status of a drug into account” badly skews the results, so that we see heroin shown as far more harmful than tobacco, (by which, one hopes, you mean cigarettes, specifically). Opiates are one of the most studied classes of drugs known, and their harmlessness at normal dosages is well-established. To label them as more harmful than cigarettes is absurd.
    This data, for lack of a better word, is useless for making assessments about a drug’s legal classification, if the desire is to base that classification on the relative harms caused by the drugs themselves.
    I also have to question the terminology. What is meant by “misuse?” Are we only talking about those who over-indulge? Or is any non-physician prescribed use considered misuse? How does this then apply to those who use alcohol or tobacco?
    How are you judging addiction? Is it always, de facto, bad? Would it still be so, even if the artificial scarcity, with its attendant high prices, were not an issue? One wonders how the harmfulness of water would be rated, were it a controlled substance.
    I think it would be much more useful to do a thorough evaluation of the intrinsic harmfulness of drugs, while carefully defining what is meant by the terms used. Are you equating the addiction of opiates with the compulsive use of cocaine and speed? What about terms like misuse, or abuse? Just as it makes no sense to group the majority of non-problem drinkers in with the problem drinkers, does it make sense to throw everyone who gets high in with a minority of compulsive, or problem users? Surely there’s a different level of harm, there.
    Perhaps those drugs which do not give pleasure ought to be included? The drugs in this assessment are grouped together strictly for moral reasons. What about the misuse of antibiotics? Isn’t that a greater threat to society than most of the drugs in this list? Wouldn’t that put these questions into a more proper perspective? I think when doing something that purports to be science, (it is being published in a peer-reviewed medical journal), one really ought to be scrupulous about not letting morality color the process.
    While I applaud the intentions of this piece, if we expect objective decisions from politicians, we must at least give them objective data to base them upon.

    • From admin;- Many very good points from sudon’t. Some were raised by academics in response to the original Lancet paper, and David has answered these, not always to contradict them, but to say that the paper is indeed imperfect, but it is better than nothing. The scientific process is conducted by people building on previous work, it would be great to see some new attempts to rank drugs which improve on the method.

      Some specific points;- you are right that many of the harms result from prohibition and other contextual issues, but no alternative way of measuring is possible or meaningful. The paper was specifically called ‘drug harms in the UK’, and indeed the harms would be different in other cultural and legislative contexts. But this is, I would argue, the most objective way to look at it, as it is not meaningful or possible to measure an intrinsic harmfulness of a drug sitting on a lab bench, as the harmfulness only exists in the context of use in the real world.

      The choice to limit the range of drugs to recreational ones is not a moral choice, it is simply the scope of this paper. It is for other scientists to publish info on the harms of antibiotics etc.

      The question of whether addiction is always harmful is also not important for this paper, which focused on harms specifically. Aspects of addiction which are not harmful would not be counted.


  10. sudon't Says:

    “…not meaningful or possible to measure an intrinsic harmfulness of a drug sitting on a lab bench, as the harmfulness only exists in the context of use in the real world.”

    I would argue that that is the only way to measure a drug’s intrinsic harmfulness. In many cases, such as when a drug is also used in medical practice, harmfulness, (or harmlessness), has already been established, and by those very methods. And, I believe there are other ways to tease out intrinsic from extrinsic harm. For instance, we see quite clearly in the case of opiates how harm disappears once the addict is given a legal, pure, and affordable supply.
    In the way that data was gathered for this paper, I think it would be particularly easy to set some criteria for sorting intrinsic and extrinsic harms, and to ask your participants to offer their evaluations based on those. It would make an interesting and instructive contrast to the original paper.
    Of course, if your scale so flew in the face of people’s received wisdom, possibly they wouldn’t credit it at all.
    I realize that the authors of this study aren’t responsible for the grouping together of these drugs, but I think it’s worth pointing out the reasons they are grouped together, and contrasting the attitudes between these drugs, and certain other drugs, (which are actually doing harm to society), to provide some perspective to your intended audience. Again, definitions of use, abuse, and misuse would be helpful here. They are conflated in this discussion far too often.
    I don’t think the debate will change, nor can it be meaningful, unless those assumptions, based in ignorance, are challenged, and the terms of the discussion clearly defined.

  11. Tim Says:

    I find the conclusions wrt alcohol very confusing. Maybe it’s because only harm is considered in this study. Corrao et al (http://bit.ly/WX0KFD) found that alcohol consumption for males reduces harm up to 72g per day (that’s 9 UK units!), mostly due to the impact on coronary heart disease and ischaemic strokes, which are major rich world killers. That is a widely cited paper, but for some reason, the benefits of alcohol mentioned in it are less frequently mentioned. Meanwhile, Channel 4 quotes an average consumption of 11.5 units per week across all adults (http://bit.ly/WX17jn).

    So, surely, on average, alcohol is unlikely to cause much harm. In fact, consumption ought to be encouraged among below average drinkers – I cannot remember the source, but I did hear a reference to 40% of drinkers consuming 2% of the alcohol.

    Have I missed something, or are the conclusions meant to be misleading?

    • From admin;-
      Hi Tim,
      Thanks for your comment, it’s an important issue. I think there is something you might have misunderstood about the Corrao paper, it is a very common misunderstanding that is perpetuated often in the media with assistance from the alcohol industry, in headlines suggesting that wine for example is practically an essential health supplement. You and Corrao are right in the sense that the protection alcohol gives on coronary heart disease and ischaemic stroke is real. If you imagine a graph of the risk of these illnesses against consumption of alcohol, you get a ‘j-shaped curve’, i.e. the risk goes down for the first few units of consumption, (moderate drinkers having a lower risk than non-drinkers) before rising sharply (very high drinkers having a much increased risk of these diseases compared to non-drinkers). This means that FOR THESE PARTICULAR DISEASES, moderate drinkers are better off. If there was a strange individual who said “I don’t mind dying young of cancer, or becoming an alcoholic, or destroying my liver, but I would HATE to get coronary heart disease”, then you could recommend that they have 2 or 3 pints a day to reduce this risk. However, most people care about overall health and overall risk of harm, and a graph of overall harm is not j-shaped, it starts rising slowly from either 0 units or very close to zero, and rises more and more steeply as units increase. Breast cancer is a good example, it seems that the more you drink the higher the risk, and vice versa, so even very careful drinkers will probably have a slightly increased risk of breast cancer compared to non-drinkers. The following document shows calculations (estimates extrapolated from data) for the amount of deaths and hospital admissions caused by alcohol. The health benefits of alcohol in a handful of specific conditions like ischaemic stroke and gallstones show in the tables as minus figures, for example on page 48 you can see that alcohol prevented about 1506 women being admitted to hospital for ischaemic stroke, but this number is swamped by admissions from dozens of other alcohol harms, from assault, falls, epilepsy, cancers, car crashes, miscarriages, etc etc. Also, on a population level (and the ISCD work considered the UK as a whole, not ‘moderate’ users only), the benefits of those at the bottom of the ‘J-shaped curve’ can be cancelled out by the costs of those on the heavy-drinking right of the curve. You can see this for ischaemic stroke in that same table, where in men (who drink more overall) alcohol does more harm than good, because the damage done by very high drinkers outweighs any benefits to the moderate drinkers.

      For this reason, the idea that below-average drinkers should be encouraged to drink more is not correct. That is not to say that the ISCD is against drinking. Rather people should be aware of alcohol as an unhealthy but potentially enjoyable drug, so they can make informed decisions about how much consumption is worth the risks.

      To comment more on the actual ISCD paper, we should bear a couple of things in mind. Firstly, the paper shows the harm of alcohol to the country as a whole, so it is not too useful as a guide to whether you can personally use it responsibly. Alcohol is easier to use responsibly than some other drugs which scored lower overall. For example, in the paper, sniffing solvents gets a lower score of total harm than alcohol, whereas we would certainly not recommend people choose glue-sniffing over a glass of wine! It’s just that whilst solvents scored highly in only a couple of harm categories (i.e. it can kill you instantly and unpredictably, but, say, doesn’t cause much economic cost to the nation), alcohol scored high or pretty high in most categories (i.e. it kills through overdose AND through years of use, it causes crime, community problems, family adversity etc etc.) So the conclusions an individual might take from alcohol coming top should not be “I should take solvents instead”, but ‘there should perhaps be a proportionally higher investment in reducing the harms of alcohol in the UK than the harms of, say, LSD”.

      • Sarah Turner Says:

        I can only speak from experience, as a specialist only in alcohol misuse among middle aged women, who most certainly are constantly taken in by hype, marketing, and crucially that they only ever measure consumption in glasses, not units. Time and time again I am told by them, that drinking wine is not the same as drinking alcoholically. Even if two or three bottles are being consumed a day. Being told that red wine is good for you, showed a massive shift in habit from white to red. Few are aware of any of the 60 medical conditions that can result from heavy drinking, vaguely aware of liver damage. I see the outcomes of this misuse, and it isn’t what the government keep banging on about. It’s respectable, middle class people, drinking alcoholically with their lives and that of their families in chaos hidden behind closed doors. There needs to be far more done to highlight this demographic and the woeful help that is out there for the over 45s who now cost the NHS far more than the 16-24 year old age group. They don’t need to score drugs, it is totally normal for them to be blitzed by 10pm every night on what they can by with an online shop. I have women coming from all over the country for my help now, with a colossal 90% recovery rate. Quite a few are clinicians.

        • From admin;-
          Thanks Sarah. Your experience is exactly what we find. That demographic is being let down, especially as the Government’s alcohol strategy also helps people maintain the illusion that the problem is ‘them’, the binge drinkers who are shamed by the tabloids, collapsed in city centres or bloodied in A&E, not ‘us’ the general voting population of all ages.

          • Sarah Turner Says:

            I would love to know where the great and the good in power believe these binge drinkers come from David, Mars? They are the children of drinkers, the vast majority in this demographic who, just as they have taught them to read, write, eat and wash, have taught them that drinking is acceptable, and wine o’clock is just so relaxing, to be cool their children take it to another level. I get very very frustrated with it all. Common sense is not so common any more.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: