The NY Times carried an op-ed titled “Crazy Pills” about a prescription anti-malaria drug that is widely used and appears to be uneventfully beneficial for most people, but has radically psychotic effects on a small fraction of patients. The piece was written by one such patient, David Stuart MacLean.
I have no issue with the facts MacLean presents, but it isn’t clear what he recommends should be done, precisely. However, it seems he’s hinting at banning the drug (Lariam), or, at least, many readers will infer that recommendation. That propensity to ban a drug that is widely used beneficially is what I take issue with (and MacLean may have intended something more along the lines of what I would propose: more careful prescribing, monitoring of users, or – preferably – pre-testing prospective users for the conditions that lead to significant negative side-effects). He cites a study where 67% of patients experienced some kind of negative side-effect. 61% of these were, apparently, clinically minimal, as only 6% of those with side-effects required follow-up treatment. I’m not fond of a drug that requires as much as 6% of its users to seek treatment for side-effects (unless it is very good at achieving it’s primary benefit – which this one is, apparently). But these numbers have a very clear and important inverse meaning that we often ignore all too often: 33% of recipients have zero side-effects, whatsoever (and all the benefits, presumably).
Banning drugs that work for billions ( 33% of 7 billion would be more than 2 billion, after all) is a brutish response. We are entering an age of “personalized medicine” – where the expectation of one-size-fits-all solutions are not only difficult or impossible – but myopic – a foolhardy dream. We have the tools now to approach the question, “Why does this drug cause significant negative effects in 6% (or, for the truly psychotic effect, something like 1 in 10,000), but not the other 94% or 9999?” Is the difference genomic? Proteinomic? Environmental?
Banning Lariam might be the best solution for now, especially if there are “good” substitutes. But as our society grows to include so many people that every statistically unlikely outcome becomes a mathematical certainty, we have to stop approaching our judgements in a binary fashion – to ban or not to ban. Rather, we need to recognize that most things – even widely “harmful” things – are safe and even beneficial to at least some people, or in some circumstances. Merck’s Vioxx fiasco now appears to be a situation where a drug that is widely beneficial for many people was pulled from the market because it can cause deadly cardiac side-effects in some people. Lawsuits were filed. Millions were paid out. Lives on both sides of the issue were drastically altered.
But lost in the din was the more important lesson: something that is predictably deadly for some people can still be absolutely beneficial to others – and we should only outright ban things when the risks are both inscrutably random and significantly exceed the aggregate benefits. This is a radical concept because we’ve all come to accept laws and norms that fail this test.
To jump straight into the fire of disagreement, for instance, consider that the propensity for addiction varies between people. As a result, for many people, opioids are useful and safe – but destructively addictive for some. Yet we’ve practically banned opioids, made merely possessing them without a prescription a crime, and are even moving to further restrict access to them, even as the amount of them prescribed has grown tremendously. Hidden within the stories of accidental overdoses and violent thefts that seem to be the only side of opioid usage that is ever discussed in our media anymore are millions of people that are obviously using them safely, and obviously benefiting from that usage (else usage would be down or addiction treatment would mirror increases in consumption). Even many people that use opioids illegally are doing so in a way that is effectively safe, else everyone that used them would be requiring addiction treatment or dying.
This type of relationship – widespread “usage” of a substance (or engaging in some “risky” behavior) while those suffering the conspicuously negative effects are only a small subset of users or doers – needs to be recognized as a clear-cut signal of a conditionally “bad” or “risky” substance or behavior, not an absolute. When something is conditionally bad we know we need to treat it differently than things that are always bad. For instance, shooting your neighbor simply because they irritated you is always bad. It should always be punished. But banning or criminalizing drugs and behaviors that are only conditionally bad would be like always punishing everyone who shoots their neighbor as a murderer.
If you have to ask when shooting your neighbor would ever not be murder then you aren’t being very creative! What if your neighbor is running at you with a machete? What if your neighbor is dying of cancer and asked you to shoot them? What if your neighbor has literally become a real-world brain-eating zombie? In each of these cases – depending on your moral sense – you may feel that shooting your neighbor in that situation is still “bad,” but we can all agree these examples are significantly different from the immorality of shooting your neighbor simply because they irritated you, or (worse?) premeditated murder for personal gain.
This distinction is why our legal system judges assaulting-thy-neighbor based on the conditions of the assault. But in many other cases we don’t apply the same reasonable pragmatism. All too often the underlying basis for this failure is sheer laziness – unfamiliarity with the positive conditions and an unwillingness to be open-minded enough to see them for what they are. It is surprisingly common for us to decide that all the uneventful instances somehow have no real net-benefit to anyone – which then makes it easy to judge on the matter since the negative instances “obviously” tip the scales against the thing or behavior. This is essentially the rationale that led to Prohibition. The social (and psychological) benefits of moderate alcohol consumption by millions of people every day were overlooked, and judgement was passed on the grounds of the ill-effects seen in those with a permanent propensity to become destructively addicted, as well as those that either accidentally or intentionally become temporarily at risk for any kind of self-destructive behavior, such as due to temporary depression. For the latter group, alcohol isn’t even the problem – it’s just the means of destructive behavior.
We (sort of) learned our lesson on Prohibition, but the truth is that many were against re-legalization, and many still are – and alcohol consumption is just one of the more frivolous of conditional evils we pass judgement on every day. Consider all the things that people (even you) regularly engage in that are either illegal or taboo (ie, not yet illegal). What about speeding? Marijuana? Texting/phoning/discussing-while-driving? (especially driving rules in school zones).
Don’t get me wrong – I’m not saying these should all be fully decriminalized. These are simply the most currently-debated cases of conditional risk where millions of people “beat the odds” every day. Many people want to (continue to) ban these – many even as they, themselves, commit the “crime” in question (or something worse). What I’m suggesting is that we come to recognize that when many people somehow “beat the odds” on a risky thing or behavior – especially when this good fortune is predictable, we should recognize the issue is conditional and seek, first, to remedy the conditions that lead to “bad” outcomes, and/or craft laws and taboos to address the truly “bad” conditions. This needs to be our default way of thinking – and a “constitutional” test of law.
Said a different way, stereotyping behaviors and things can be just as evil as stereotyping people. But stereotyping, itself, is a conditionally bad behavior. There are many cases where stereotyping people is a good thing to do. For instance, guessing that a person with a swastika tattooed on their forehead might either be a dangerous neo-nazi or a dangerous fan of Charles Manson together comprise the bulk of statistically likely possibilities (that is, once you’ve ruled out that the swastika is an optical illusion or a temporary tattoo as part of temporary role-play (eg Halloween)). Likewise, if a 18 year old blonde girl wearing a bright yellow sundress is walking right at you sporting a broad smile and a hefty black bible, stereotyping is the perfectly valid way we decide that
- 1. she is effectively not a physical threat, but that
- 2. (unless you know her and are a member of her congregation) it is very possible you are about to have an awkward exchange on metaphysics, or
- 3. be asked for directions or some bit of trivia
All other possibilities represent such a small likelihood that you will be rightfully surprised if any of these prejudices turn out to be wrong.
And all I’m really saying is that locking yourself indoors for the rest of your life because #2 happened one day makes about as much sense as our current laws on speeding, use of mobile phones, marijuana, and many other issues that would surprise each of us, once we see them in the proper light. Let’s rethink instances of bans or criminalization and consider whether technology is ready to offer us a better way to handle the bad conditions.