That's why people started coming to MI training back in the 90s. "How can I deal with the most impossible, difficult, intransigent people?" We used to start off our trainings with demonstration and practice on "resistant" cases, and students always wanted to role-play the most impossible clients. There are no clients as difficult as those role-played by addiction therapists. What we discovered over time, though, is that if you start with the foundational methods of MI - engaging, focusing, evoking - if you listen well, "resistance" is not that common a problem. In our third edition we recommended abandoning the term "resistance" because it's an unhelpful way of thinking about working with clients. It sounds like a competition: "Here I am doing this great therapy, and you're just resisting me." Most of what we had called resistance in our first two editions was just sustain talk - one half of the client's normal ambivalence. Nothing pathological or unusual about that. The other component we called discord, which is signals about dissonance in your working alliance. Both are highly responsive to counseling style. It takes two to "resist." It's an interpersonal phenomenon, and thus you can influence the extent to which it will occur and persist. I could teach you how to have highly resistant clients. Unfortunately, client resistance predicts poor outcome, and it's much better to counsel in a way that reduces sustain talk and discord.

Let me add my favorite response – explicitly honoring the client’s autonomy to make that choice. After all, the truth is that no one can make a person give up an addiction if they are determined not to. MI, like everything else, will fail if the person is determined not to change. But after all, how big a percentage of your caseload is people who are determined not to change? Some, but most people you see are ambivalent, conflicted and tortured by their addiction at some level. What they really are is the “cake” of ambivalence covered by the “frosting” of defiance. Once you cut the cake, you can see what flavor it really is under all that frosting.

The fact that you call it an addiction is a foregone conclusion that the person may not share. How do you raise a sensitive topic? A good start is to ask permission, like "Would it be all right if we talked for a few minutes about your alcohol use?" That usually works well in primary care settings where the person didn't come in to talk about alcohol. In MI we begin with curiosity. I've often told people that I don't care about labels. What I am concerned about the person, and what, if anything, he or she might choose to change. Let the person tell you about their own life and situation. Most people don't respond well to being told what's wrong with them and what they should do about it. That's why MI seemed such a radical departure from the confrontational tactics of the 70s and 80s.

First of all, post-treatment changes in addictive behavior tend, on average, to sustain very well in the long run. Individuals bounce around a bit, but if, for example, clients show on average 90% days abstinent at the end of treatment, it will probably still be 80% days abstinent 1, 2, or 3 years later. That's true for most any addiction treatment that produces significant change. In Project MATCH we directly compared cognitive-behavior therapy (12 sessions), motivational enhancement therapy (4 sessions), and twelve-step facilitation therapy (12 sessions). Overall improvement was huge, was the same for all three treatment groups (randomly assigned), and sustained very well at 1-3 year follow-up. That's the normal outcome of treatment for alcohol use disorders, and has been since the 1970s, replicated in many studies. So how do we develop such pessimistic beliefs about treatment outcome? It's a product of perfectionism. We have even used "survival curves" on which, once you have a drink, you fall off and can never get back in the "success" category. Imagine if we used that standard in treating hypertension or diabetes: one elevated blood pressure or blood sugar and treatment failed. The very idea of "relapse" is misleading, because it implies that there are only two possible outcomes: perfection or disaster. In fact, most outcomes fall somewhere in between, and overall post-treatment outcomes are very good.

About as well as the other effective, evidence-based treatments we’ve got.

The strongest evidence for MI so far is in clinical trials focused on alcohol, tobacco, or marijuana use. There's also some evidence with methamphetamine use disorder. But stand-alone MI is not the usual approach. Rather MI is being used in combination with other psychotherapies or with pharmacotherapy. With behavioral addictions, there's solid evidence with pathological gambling. There are significant effects with MI alone, but usually it's used in combination now. It could be a stepped care approach: Try MI first, and if it's not sufficient add other methods.

The most common length of treatment for someone receiving treatment for addictions is….one session. We tend to conceptualize treatment as us helping people with the how of changing. But if you only have one session, maybe it is smarter to focus on the why of changing, since many people already have good ideas and skills for how to change, once they decide to (though of course not all do). I would like to see “intake” of addiction clients reconceptualized away from gathering facts to encouraging motivation. That might be all we need to do, or have a chance to do, anyway. Both Bill and I have reorganized addiction programs we worked in to follow that kind of model with a reduction in number of million-dollar work-ups for clients who never come back.

Often that's a matter for agency policy. In group sessions we typically asked the person to come back sober, and arranged transportation. With individual sessions you have more discretion unless there's an absolute agency policy. I always remember the research done in emergency departments, where the people being treated (selected for study) are all intoxicated, have been stitched up or otherwise treated following an injury, and it's usually the middle of the night - not exactly ideal conditions for psychotherapy. Nevertheless, a 20-minute person-centered conversation about drinking in this situation can often have a substantial effect. The research goes back to the 1960s with Morris Chafetz, who found return for treatment (after an ER conversation) increased from zero to about 60% in randomized trials. So don't automatically assume that a conversation will be fruitless because there is alcohol present. Usually it's on the downside of the intoxication curve, and there is good evidence that it can be effective.

Let’s all be safe out there.

If by "intervention" you mean what the Johnson Institute used to do (but no longer does), or what is featured on Arts & Entertainment, no. MI is almost an opposite style of working with people. However, the original reason I developed MI was to do what "interventions" hoped to do: to help people look in the mirror, take a good look at their drinking (in a safe and supportive environment), and decide what if anything they might like to do differently. I originally thought that it would be a prelude to treatment, like priming the pump to respond to treatment, and indeed MI seems to do that. When MI is added to other evidence-based treatment methods, both work better. The surprise, though, was how often MI along was enough to prompt change similar in size to the effects of other forms of treatment. Once you get people over the hump of ambivalence, sometimes that's all that is needed. If you're asking whether MI is a specific kind of behavioral health intervention, yes and no. We can and do specify what constitutes good practice of MI that is associated with subsequent change. What seems to be happening, though, is that MI is usually being combined with other evidence-based treatment methods. MI is like a way of doing what else you do.

I can say definitely, yes. We weren't sure before we did the first clinical trial with an adolescent SUD group, because teenagers can defeat almost anything. We were having good results with adults, but what about teens? Well, we got the largest effect size of MI that we had ever seen. Outpatient session completion increased from 8 (without MI) to 20 sessions in a randomized trial, and polydrug abstinence rate doubled even though they got the same treatment program- the only difference was whether or not they had an MI session at the beginning. So few adults actually listen to teens, that you get a big contrast effect just with the engaging skills.

Well, first of all let me say that I favor scrapping the term "relapse" from our vocabulary. It carries so much moralistic and black-and-white-thinking baggage that I believe it does more harm than good. When a person with diabetes comes into the emergency department, no one says, "You relapsed!" The focus is on life-saving first and then figuring out what adjustments are needed in treatment and recovery. Imperfection is human nature, and "relapse" implies that there are only two outcomes: perfection or disaster. Recovery and treatment outcomes are mostly somewhere in between - successive approximations. The answer to your question, though, apart from my relapse sermon, is Yes, of course. When a person who intended to abstain comes back to treatment (thank goodness) having used alcohol or another drug, MI still applies and is a good tool for focusing on how the person will choose to proceed.

Yes, but you need a different model for thinking about relapse as Bill is saying. In particular, if people are ambivalent about changing after their relapse (for example, they’ve decided that giving up smoking isn’t that important after all), MI is optimal. If they are ready to move forward with a plan (I have to get right back to quitting!), MI might not be as important.

First of all, MI is not needed when a person is already ready to change. Actually it's the evoking process that isn't as important. We now teach MI as comprising four processes: engaging, focusing, evoking, and planning. If someone is ready to change, don't spend much time evoking motivation for change because it's already there. Engage, have a clear focus, and go on to planning. If a client seems to be in "precontemplation" they are not ambivalent, and there the task is actually to foster some ambivalence. It's what we now discuss as developing discrepancy (chapter 18 in our 3rd edition), a phrase that we formerly used in a more generic sense. We haven't yet found a diagnosis where MI is unhelpful, though there may be some. Probably children need to reach a certain level of cognitive development for MI to be effective, since it is intended to trigger self-regulation, but for younger children you really need to be working with the parents anyhow. Perhaps the simplest answer is: when it's not working for you.

I am increasingly saying that MI is a way of doing what else you do, whether it's addiction counseling, diabetes education, teaching, coaching, whatever form of helping relationship is yours. In my own training I learned a person-centered way of counseling before I began learning and practicing behavior therapy, and they always fit together well for me. A common combination has been MI and cognitive behavior therapy.

What I will also say here is that MI cannot be your only tool. MI is not appropriate for every client, so it does not replace your other skills or expertise. There are times it must be combined with other approaches. If you like the idea of combining MI with cognitive-behavioral treatments, have a look at the work of Henny Westra (Westra & Arkowtiz, 2010) who has a very sophisticated way of blending these two approaches.

That's easy to answer. We have never found any relationship between years of education and the ability to learn and use MI. Even doctoral-level people can learn it! It has been used effectively by peer health visitors in Africa, brand new therapists in training, and paraprofessionals. It seems to be about human relationship, not degrees or licenses.

Even more good news! The type of profession you practice and your years of experience do not seem to predict who is good at learning MI. Of course, when you are a licensed professional, you must abide by the guidelines of your chosen field. But there is nothing about any particular profession that bodes well or poorly.

Harm reduction is what we all hope to do in helping professions, in addition to promoting positive heath and growth. The phrase "harm reduction" in addiction treatment usually refers to a willingness to work with people to make changes that are less extensive than a clinician might wish. I did ten years of research on helping people moderate their drinking, at a time (1970s) when that was highly controversial. Now it's mainstream, at least in health care, to screen for heavy drinking and help people reduce their use. One thing we found, by the way, was that most people trying this moderation approach ultimately decided to quit drinking. It's not that they tried and failed. They were moderating their drinking, and found it wasn't worth the effort that it required. Working with people where they are at present is central to MI.

I don't know the context in which you work, but MI has been used effectively in relatively brief contact settings like primary care. In part, the question is more time consuming as compared with what? If what is needed is a change in the person's behavior or lifestyle, it is not more efficient to just tell the person what to do. Practitioners have told us that MI actually saves them time. MI is a particular way of doing that tasks you already do.

Maybe MI will take longer. Sometimes giving people time to say why they might want to change does take longer than just asking questions to get facts. But we have all had the experience where it feels like pulling teeth trying to get answers to simple questions from our clients. They need to be listened to before they are ready to move forward. Sometimes when you behave as if you have all day, it just takes minutes; whereas if you act like just have minutes, it can take all day.

Definitely. I usually recommend developing your MI skills in individual conversations first, because there is more to manage in group settings, but the spirit and method are the same. Wagner and Ingersoll wrote a book specifically on Motivational Interviewing in Groups (Guilford Press) and there are clinical trials of MI in group format.

To me one of the most fascinating things about MI in groups is how to think about change talk. Is it the average frequency of change talk in the group that makes an impact on the individual or do they have to offer their own? Does change talk from people who talk a lot get “count” less than change talk from people who almost never speak? One of my favorite studies about change talk in groups comes from Shorey, Martino, Lamb, La Rowe and Santa Ana (2015). They study the synergistic exchange of change talk among and between group members, which they call “relatedness”. Group MI is associated with more of this kind of language than other kinds of groups. So maybe there is a special kind of exchange of words that happens in MI groups that is important.

The most honest answer is that I don't know. Steve and I could not anticipate the changes in our second edition from the first, or in the third edition when we wrote the second. Developments happen rapidly. MI is being applied with an ever broadening range of problems and populations, now in at least 55 languages around the world. My own thinking has been toward thinking of MI not as a competing treatment technique, but as a particular kind of therapeutic relationship, a way of doing what else you do.

I really want to see what will happen when we reconceptualize addiction treatment as shorter and more frequent episodes, like the model for diabetes care. Once we realize that “treatment” is going to last for the lifetime of the person (like diabetes) we can start being smarter about it. I think MI is especially well suited for shorter and more frequent episodes of focused care.

"Convince" or "persuade" is an unhelpful way to think about it. It sounds like a win/lost competition. The method of MI is about helping people talk themselves into change. That sounds unlikely at first, and it's amazing to watch it happening. There is also a very reasonable question as to whether "rehabilitation" is the right choice. Better to consider a menu of possible options. Most people change addictive behaviors on their own without treatment. Intensity of treatment (in randomized trials) is unrelated to outcome. I think a better question is what this person is ready and willing to do, and readiness is very much influenced by MI.

Remember, many people can change without our help. I know, I know – it’s hard to believe, but the data are clear here. Maybe what they need most is to leave their ambivalence behind. Maybe our biggest job is helping people figure out how to let go of their ambivalence so they can change themselves.

Well, it depends on the presenting issues. When treating someone with a substance use disorder, for example, you probably haven't helped much if substance use doesn't change. With that said, though, movement one stage forward in the transtheoretical stages of change IS progress. Moving from precontemplation to contemplation, or from contemplation to preparation would increase the chances of subsequent action and change. I don't think you need to have clients fill out a stages questionnaire for clinical purposes. You can hear the stage of change readiness from a client's in-session language. The ratio of change talk to sustain talk is a pretty good predictor of proximal movement toward behavior change, and therapists definitely influence that ratio.

When a client asks for your advice, you have permission to offer it, but here are some caveats. First I usually want to know what ideas the client has. I have also found it wise not to offer one suggestion, because the natural client response is to tell you what's wrong with your suggestion. Instead, offer a menu of options. A great thing about addiction treatment is that we have a variety of evidence-based approaches, none of which works for everyone. I usually describe several possibilities and ask for the client's hunches about which might be best to try first. It also helps to acknowledge the client's autonomy, in essence giving the person permission to disagree. "I don't know if this will make sense to you," etc.

Think of Giving Information and Advice like chili powder when you are cooking a stew. Add a little and see how it tastes (how the client responds). You can always add more, but if you put in too much it is a big setback. Like chili powder, a little direction goes a long way.