Skip to main content

Verified by Psychology Today


Self-Medication or Self-Destruction?

Much of the pain that needs medicating comes from the addiction itself.

The self-medication model portrays addictive behaviors as attempts to diminish the feelings of anxiety, loss, shame, and loneliness left in trauma’s wake. For me, this connection is too obvious for words. I was lonely, depressed, and constantly on the lookout for personal attacks while at boarding school as a teenager. Within a year of leaving, I had shot enough heroin to end up unconscious in a bathtub, appearing to my friends to be dead. I don’t for a moment doubt the connection between these episodes of my life. And I say that, not as a scientist, but as a regular person, trying to make sense of it all.

I recently spoke with a reader who has serious problems with alcohol, but only when things go wrong in her personal life. Eleven months of the year she has no craving, no attraction to alcohol. She doesn’t even have to be on guard because there’s no urge to get drunk. However, when her (now, thankfully, ex) husband became abusive, and later threatened to block her custody of her son ... that's when she drank to excess. How can you be an “addict” only when things get tough, and then become a non-addict when life goes back to normal? The disease model simply can’t explain that sort of pattern, whereas the self-medication model predicts it. Threat and anxiety lead her to take alcohol to soothe the hurt.

But there are problems with the self-medication model that need to be addressed.

First, although trauma may lead to addiction, it isn’t necessary. Some people fall into addiction without any evident history of trauma—although you might find it if you dig deep enough.

Second—the issue I'll focus on in this post—is that self-medication doesn't work so well. The things we take or do to diminish bad feelings actually increase them in the long run, or even in the not-so-long run. Maybe we’re not very good doctors. We prescribe for ourselves treatments that do more harm than good. Or they work for a little while—a month, a week, an evening—and then we get hit by the after-effects. Our dopamine-powered beam of attention cares only about the immediate, not the long run. Pretty short-sighted for a doctor.

Such iatrogenic (more harm than good) effects don’t actually contradict the argument for self-medication. Many medications produce iatrogenic effects—side-effects, dependency, systemic damage, and so forth. But the balance between help and harm is crucial. It should be the first thing your doctor considers when prescribing medication. In the case of addiction, we have to ask: how much of the trauma being “medicated” comes from the medication itself? If the answer is "most of it", that's one hell of a vicious circle. In fact, it challenges the very idea that trauma causes addiction—rather than the other way around.

So let’s imagine a causal story that goes completely opposite to that proposed by the self-medication model.

As I noted above, some people start down the road to addiction without having lived through serious trauma. But even following trauma in childhood or adolescence, one’s PTSD or depression or anxiety might be under control. When I first tried heroin, I wasn’t terrifically happy, but I wasn't in great psychic pain at the time, either. The heroin felt good. It made me warm and cozy, body and soul. Enter the choice model: I want to do that again, because it’s more valuable to me than any alternative on a particular day or evening. After a while, the substance or activity is a presence in one’s life. And that presence takes on increasing value: it’s sorely missed when it’s gone. Now the source of my anxiety wasn’t so much my historical injuries (e.g., my mother’s depression, my stint at boarding school). Rather it was my present fear of going without dope, and wanting it badly, and not being able to stop thinking about it. Now we’ve got at least two of the most common outcomes of trauma—loss and anxiety—both caused by present rather than past events.

Then along comes outcome number three: shame. The loss of self-control—whether due to dirty underwear at age 4 or slavering desperation to get high at age 24—is contemptible. That’s how others see it, so that’s how we see it. So addictive thoughts and deeds lead directly to shame—often intense shame. And guess what? Shame is one of the most common sequelae of trauma. In fact, shame is one of the few emotions that is directly, viscerally painful.

Now, combine the loss you feel after running out or stopping, with the anxiety you get from craving what you can’t seem to get or know you shouldn't get, with the shame that comes from your lack of self-control, and you’ve got a feast of negative emotions. The need for self-medication is now screaming—so the addiction itself is the trauma.

The vicious circle—whereby addiction causes psychic pain that leads to further addiction—may well be the causal engine we’ve been searching for. Yet self-medication is only one part of this story. The story, as mapped out here, doesn't show that traumatic life events produce a special, intrinsic need for self-soothing. What it shows is that, for some period of time, we believe there’s one thing in the world that can make us feel better—and we end up being wrong.

Of the three models of addiction—disease, choice, and self-medication—the self-medication concept still works best for me. As long as we acknowledge that trauma is an ongoing progression, with its roots in our childhood but its branches still growing, still advancing, sometimes wildly, out of control, with each addictive act.

More from Marc Lewis Ph.D.
More from Psychology Today