Estimating the populations at risk for dependence on marijuana versus cocaine

Oct 08 2012 Published by under Cannabis, Cocaine, Drug Abuse Science

The recent fax (yes, they still call it this despite it arriving via email attachment) from CESAR (Vol 21, Issue 40; October 09, 2012) puts us back on an occasional theme of this blog.
They have adapted data from the latest update from SAMHSA's National Household Survey on Drug Abuse. This figure shows the number of past year users of selected illicit/recreational drugs.
Interestingly, marijuana use continues to trend up from the approximate plateau of 2002-2007, while use of cocaine is trending downward. Even the nonmedical use of prescription drugs (which has been a big problem overdose-wise) is relatively flat. Rounding slightly, we're looking at some 30 million past year users of marijuana compared with 4 million past year users of cocaine.

So why is this interesting? Well, as we've covered in the past the notion of conditional probability of dependence is a key issue for parents and policy makers and yet we have really poor estimates on that. Direct studies are usually limited in scope and the big-scale epidemiological stuff is too imprecise- i.e., rarely are there good diagnostics of dependence. So we sometimes have to infer things based on, e.g., daily use rates versus annual rates. Something like that. Fortunately the more precise studies and the broader interpretive efforts tend to agree.

Roughly speaking the conditional probability of alcohol dependence is on the order of 4%, for cannabis on the order of 8-10% and for stimulants, including cocaine, on the order of 15%.

So, applying these rough estimates to the past-year data above, we end up with something on the order of 600,000 dependent on cocaine and 2,400,000 dependent on marijuana. If you dropped the estimate of conditional probability for marijuana to the 4% of alcohol, you still end up with 1.2M people dependent on marijuana.

My point, as always, is that the definition and scope of a "drug dependence problem" is going to depend on frame of reference. One important frame of reference in my view is the number of people who are affected. This, btw, is why we think of alcohol dependence as such a huge problem even though just about every estimate suggests the conditional probability of dependence is one of the lowest. Because the percentage of the entire population exposed to alcohol on a regular basis is so large, the number of people who are dependent is relatively large.

22 responses so far

  • physioprof says:

    It looks like heroin use went up sharply when the economy collapased in 2008.

  • drugmonkey says:

    There are two other inputs To consider, PP. first a supply issue having to do with the Afghan war. Second, a poorly understood conversion of people who started on OxyContin (and friends) converting to smack because of $$.

  • Virgil says:

    As a society, when we discuss dependence, we're also talking about the consequences of deprivation. The attention a drug gets from society in terms of dependence, is really driven by how big is the effect on society of dependent individuals being deprived. Hence the current focus on meth, which is a tiny fraction of people, but when they get deprived they go apeshit, so it's worth dealing with in the interests of public order. At the other end of the spectrum is the huge percentage of the population "dependent" on coffee, but they don't hold up people at ATMs just so they can buy a cup-o-joe.

    So long as the people tasked with defining the "problem" of dependence, are the law enforcement officials dealing with the consequences of such, dependence will always be defined at the societal level, rather than at the level of the individual who is actually dealing with the personal consequences of dependence.

  • drugmonkey says:

    Yep. Like I always say, some independently wealthy person with no dependents and a whomping smack habit really doesn't necessarily have a drug "problem".....

  • flip says:

    Or converting to smack because of supply constraint, as has occurred in my rural Appalachian home town. When I was a young'n, heroin was available but not widely abused, it was also higher priced than the usual fare offered. Then oxycodone and friends took over like gangbusters. Pill-mill closures have now flushed their clientele to the heroin market.

  • drugmonkey says:

    http://www.ncbi.nlm.nih.gov/pubmed/22465379
    OxyContin® as currency: OxyContin® use and increased social capital among rural Appalachian drug users.

  • drugmonkey says:

    Transition from first illicit drug use to first injection drug use among rural Appalachian drug users: a cross-sectional comparison and retrospective survival analysis.
    http://www.ncbi.nlm.nih.gov/pubmed/21883604

  • drugmonkey says:

    http://www.ncbi.nlm.nih.gov/pubmed/15370019
    Impact of the heroin 'drought' on patterns of drug use and drug-related harms.

    (in Adelaide....)

  • What's the conditional dependence probability of the 'nip?

  • becca says:

    ""dependent" on coffee, but they don't hold up people at ATMs just so they can buy a cup-o-joe."
    We'll see what happens when brick and motar banks disappear, and no free coffee is available. I expect to see more grad students (because the scientists will give up free coffee when you can pry it out of their cold dead hands), or people holding up folks at ATMs. Unintended consequences of online banking!

  • Grumble says:

    " some independently wealthy person with no dependents and a whomping smack habit really doesn't necessarily have a drug 'problem'"

    Yeah, except that the risk of overdose is high because the concentration of heroin in street smack is so variable. And overdose is a problem even with prescription opiates, where the dose is well known to the abuser.

    Comparatively speaking, the overdose problem is vanishingly small for marijuana and caffeine. So that really mitigates the supposedly greater conditional probability of dependence for those drugs.

  • DrugMonkey says:

    From the libertarian perspective, why do we care if some independently wealthy person with no dependents accidentally kills himself?

  • becca says:

    "From the libertarian perspective, why do we care if some independently wealthy person with no dependents accidentally kills himself?"
    The same reasons we care about intentional suicides:
    1) many are incomplete, thus costing us ER time (even if they pay their bills in full without insurance, that's some kid with a broken arm who isn't seen for another hour while they deal with the crisis)
    2) just cause you have no kids doesn't mean nobody relies on you
    3) just cause nobody relies on you doesn't mean nobody cares about you
    4) just cause nobody cares about you doesn't mean you'll never be worth caring about/waste of human potential

    I don't know that any of those are outweighed by FREEDOMZ! But I think they are worth considering.

  • drugmonkey says:

    "dependents", not "kids".

    And I'm not sure the state has an interest in the tender feelings of someone's momz....

  • Isabel says:

    Just to be clear: "the state" has no (legitimate) interest in dependent cannabis users either. As the state itself has determined -repeatedly. Or hadn't you heard?

  • drugmonkey says:

    That's not what Federal law says Isabel.

  • Isabel says:

    "This, btw, is why we think of alcohol dependence as such a huge problem even though just about every estimate suggests the conditional probability of dependence is one of the lowest. "

    Also, this certainly isn't true, or more accurately it is only a part of the story.

  • Isabel says:

    part of the reason, I should say. And what's with "That's not what Federal Law says"?

    If a government assigns a committee to look into a situation and then disregards the committee's recommendations, and then does this repeatedly, can we still say the government has a sincere, legitimate interest in the situation?

    what do we even call a situation like this?

  • Grumble says:

    From a libertarian perspective, we don't care about what anyone else does, ever.

    From a human, common sense, public policy, and just about any other perspective that has some basis in ethics, we do care about our fellow human beings.

    Of course, almost no one is "independently wealthy" and has no dependents, so the question is moot to begin with. We need to consider the entire range of negative (and positive) effects caused by drugs, whether we care about the user or the user's dependents. And those negative effects are only partially a function of the conditional probability of dependence.

  • drugmonkey says:

    Your characterization of libertarianism lacks nuance. We care when the rights of one interfere with the rights of another.

  • drugmonkey says:

    Also Grumble,
    When did I say anything about the subject of this post being anything other than part of any consideration? You seem in fact to be arguing that the numbers affected are not important. Care to defend that?

Leave a Reply