Introduction
You have been rehearsing it for months. The sit-down. The things you'd say and the order you'd say them in. The tears, the denial, the door slamming, the possibility that you make everything worse and they disappear entirely.
Here is the most useful thing this article can tell you: the confrontation you are dreading is not the intervention you should be planning, and the research says so fairly clearly.
The Evidence on Confrontation
For decades, two approaches dominated advice for families: the dramatic staged confrontation — the "Johnson Institute intervention," familiar from television — and referral to Al-Anon, which teaches that you are powerless over the other person's use and should detach.
In 1999, researchers ran a randomized trial comparing three approaches for concerned family members. The outcome measured was whether the loved one actually entered treatment.
Al-Anon facilitation: 13%.
The confrontational Johnson-style intervention: 30%.
A third approach, called CRAFT — Community Reinforcement and Family Training: 64%.1
Subsequent work with families of people using illicit drugs found CRAFT engaging around 74% in one evaluation, and in a later randomized trial, 58.6% for CRAFT alone and 76.7% for CRAFT with aftercare, against 29% for Al-Anon/Nar-Anon facilitation.
The confrontation you're afraid of is less than half as effective as an approach you have probably never heard of.
What CRAFT Actually Is
It is neither confrontation nor detachment, and that's the point.
It teaches the family member specific behavioral and communication skills, on the premise — contrary to the detachment model — that family members have substantial influence and can use it deliberately. Broadly, it involves: understanding the pattern of use, changing how you communicate, reinforcing periods of non-use, withdrawing rewards that accompany use, allowing natural consequences, improving your own life, and preparing a treatment invitation for the moment when the person is receptive.
It is unilateral. It works through you. It does not require the person who is using to agree to anything, or even to know it's happening.
The Part That Should Reassure You Most
CRAFT improves the family member's own depression, anxiety, and functioning — in trials, this held even when the loved one never entered treatment during the study period.
So the thing you would be doing has a benefit to you that does not depend on their cooperation. That is unusual, and it means the effort is not a gamble on someone else's behavior.
What Detachment Gets Wrong Too
Since CRAFT outperformed both, it's worth noting that the gentler traditional option didn't fare well either.
The detachment model teaches that you are powerless over another person's use, that you cannot influence it, and that your task is to step back and care for yourself. There is something genuinely valuable in that — it relieves an unbearable sense of responsibility, and the self-care component is real.
But the premise appears to be empirically wrong. Family members do have substantial influence. In the trial, Al-Anon facilitation engaged 13% of loved ones in treatment, less than a quarter of the CRAFT figure. Being told you are powerless is comforting and it may not be true.
This is not an argument against attending Al-Anon, which helps many people enormously with their own suffering. It's an argument against believing there is nothing you can do.
Why Confrontation Backfires
Understanding the mechanism helps you stop rehearsing the speech.
Confrontation reliably produces defensiveness, which produces argument, which produces a person defending their use out loud — and people tend to believe the arguments they hear themselves make. A cornered person does not update. They dig in and then they conceal more carefully.
It also positions you as an adversary at the precise moment you need to be the person they can come to. The staged intervention frequently ends with the family having said everything, the person having heard nothing, and the concealment tightening.
What to Do Instead of the Big Conversation
Some principles that follow from the evidence.
Many small conversations beat one enormous one. Low stakes, no audience, no ultimatum.
Describe, don't diagnose. "You've missed the last three Sundays and I've been frightened" lands differently from "you're an addict." The first is a fact about your experience. The second is a verdict that invites argument.
Ask, don't tell. Questions produce reflection. Statements produce defense.
Don't do it while they're intoxicated. Judgment narrows under the influence, and nothing said will be weighed properly.
Have the option ready. The window in which someone is receptive is often short and unannounced. Know the number to call, the clinic, the appointment that could be made today.
Say what you'll do, not what they must. "I'm not able to lend money anymore" is enforceable. "You have to stop" is a wish.
The Fear Is Also Information
One more thing worth naming.
If you are frightened of this conversation, ask what specifically you're frightened of. That they'll be angry? That they'll leave? That they'll confirm it and you'll have to act? That they'll deny it and you'll have to decide whether to believe them?
Some of those fears are about them. Some are about what you'll be obliged to do once it's said out loud, and those are worth working out before the conversation rather than during it.
The Bottom Line
Confrontation is not the best available tool and the research is fairly clear about it: in a randomized comparison, staged confrontation engaged 30% of people in treatment, Al-Anon facilitation 13%, and CRAFT 64%. CRAFT works through you, doesn't require the other person's agreement or even their knowledge, and improves your own mental health even in cases where the loved one never gets help. Stop rehearsing the confrontation you've been dreading. Look up CRAFT, and find a clinician trained in it.
Sources
- CRAFT 64% vs Johnson 30% vs Al-Anon 13% — Miller WR, Meyers RJ, Tonigan JS (1999). Engaging the unmotivated in treatment for alcohol problems: a comparison of three strategies for intervention through family members. Journal of Consulting and Clinical Psychology, 67(5):688-697. View source ↗