Most families dealing with a loved one’s addiction spend years doing everything they know how to do, and watching it not work. The CRAFT approach for families of addicts offers something different: a structured, evidence-based method that actually moves the needle, not just on treatment entry, but on the family’s own wellbeing along the way.
What is the CRAFT approach?
Community Reinforcement and Family Training, or CRAFT, is a behavioral intervention designed to help family members and close friends get a loved one into treatment for substance use disorder, reduce that person’s use even before treatment begins, and improve the family member’s own quality of life in the process. It was developed by psychologist Robert Meyers at the University of New Mexico, building on the Community Reinforcement Approach that had already shown strong results in direct treatment contexts.
What makes CRAFT distinct is its starting premise. Where Al-Anon encourages family members to detach with love and accept powerlessness over a loved one’s behavior, CRAFT teaches that family members actually have significant influence, and gives them specific skills to use it. Where the Johnson Intervention model assembles a surprise confrontation designed to break down denial, CRAFT recognizes that confrontation frequently escalates resistance rather than reducing it. CRAFT is non-confrontational by design, and it puts trained skills in the hands of the person who is already showing up every day.
The scale of the problem these families face is significant. According to the Substance Abuse and Mental Health Services Administration, roughly 46 million Americans met criteria for a substance use disorder in 2021. Behind each of those individuals is at least one family member trying to figure out what to do.
Why traditional family approaches fall short
The most widely cited clinical comparison of family approaches comes from the landmark CRAFT trials conducted by Meyers and colleagues in the late 1990s and early 2000s. In a study published in the Journal of Consulting and Clinical Psychology in 1999, Meyers et al. randomly assigned Concerned Significant Others (CSOs) to one of three conditions: CRAFT, Al-Anon facilitation therapy, or the Johnson Intervention. Treatment entry rates among the loved ones were 64% for CRAFT, compared to 17% for Al-Anon facilitation and 30% for the Johnson Intervention. These findings have been replicated across populations including partners of alcohol users and parents of drug users, consistently showing CRAFT outperforming the alternatives by a wide margin.
What the numbers describe, in plain terms, is that the most commonly recommended approach for families (finding a support group and learning to let go) produces treatment entry in fewer than one in five cases. That is not a personal failure on the part of families who have tried Al-Anon. It is a structural gap: detachment-based models offer emotional support without a strategy for change.
The enabling trap
Enabling looks specific in a household. It is calling in sick on someone’s behalf after a night of heavy use. It is paying a bill that the substance use disorder created. It is absorbing the social damage of missed events and offering explanations to everyone who asks. None of this happens because the family member is naive or weak. It happens because the alternative, allowing consequences to fall, feels unbearable when you love someone.
A 2012 review by Orford and colleagues, published in Drug and Alcohol Dependence, examined the family stress and coping literature and found that families that primarily use protective or shielding coping behaviors (what clinicians often call enabling) report significantly worse personal wellbeing over time, and their loved ones show lower rates of treatment entry. The mechanism is direct: when consequences are consistently removed, the internal calculus that might move someone toward help never tips.
CRAFT does not ask families to stop loving the person. It asks them to redirect that energy in ways that are actually more likely to work. The shift is from absorbing damage to strategically responding to it.
Why confrontation often makes things worse
The Johnson Intervention, despite its cultural familiarity (it is essentially what most people picture when they hear the word “intervention”), has a weaker evidence base than many assume. A 2004 clinical review by William Miller and colleagues, examining the evidence for confrontational approaches in addiction treatment, found that therapist confrontation consistently predicted worse outcomes, including dropout and continued use. The pattern holds in family-initiated interventions as well: surprise confrontations frequently trigger shame, defensiveness, and disengagement rather than openness.
The catch is not that honesty is bad. It is that honesty delivered through pressure, without an intact relationship to hold it, gives the person an easy exit: walking away from the conversation. CRAFT preserves the relationship as the vehicle for change, which keeps the conversation open over time.
The core goals of CRAFT
CRAFT pursues three outcomes simultaneously, and the fact that all three are explicit is part of what distinguishes it from older models. The first goal is getting the person with a substance use disorder into treatment. The second is reducing their substance use even in the period before treatment entry occurs. The third is improving the wellbeing of the family member doing the work.
That third goal matters more than it might appear at first. Most family-focused approaches treat the CSO’s wellbeing as secondary, as something that improves once the loved one gets better. CRAFT takes the position that the family member deserves support and improvement regardless of what the person with the SUD decides to do. This reframing changes the experience of the process entirely. The family member is not just a vehicle for getting someone else into care. They are a client in their own right.
How CRAFT works: the key skills families learn
CRAFT is a skills-based model. The techniques it teaches are learnable, practical, and grounded in behavioral science. They are not personality traits or spiritual dispositions. They are specific behaviors that can be practiced, adjusted, and improved. Here is what that skill set actually looks like.
Functional analysis of behavior
Functional analysis is a structured way of mapping what happens around substance use: what triggers it, what the behavior itself looks like, and what consequences follow, both short-term and long-term. A family member doing a functional analysis is essentially becoming a careful observer of a repeating pattern.
In practice, this means tracking specific incidents. What was happening in the hours before use began? What emotional state was the person in? Who else was present? What happened immediately after use (relief, social connection, conflict avoidance)? And what happened later (regret, financial cost, relationship damage)? This information is not just useful emotionally. It reveals the structure of the behavior, which is exactly what CRAFT uses to design effective responses.
A 2019 study by Roozen and colleagues, reviewing CRAFT outcomes across multiple trials, found that the functional analysis component was consistently associated with CSOs developing a clearer, less emotionally reactive understanding of their loved one’s use patterns, which in turn supported more effective use of later CRAFT skills. The immediate action is simple: start a brief daily log this week, noting one specific incident and recording what came before it and what followed.
Positive communication training
CRAFT teaches a communication style built around three elements: expressing feelings without blame, making specific and positive requests, and staying engaged rather than shutting down. The goal is to make conversations more likely to stay open and less likely to escalate into the kind of exchange that ends with one person leaving the room.
A 2010 study by Manuel and colleagues, evaluating CRAFT’s communication component specifically, found that CSOs who completed communication skills training reported significantly higher relationship satisfaction and lower rates of conflict, independent of whether the loved one had entered treatment yet. The communication shift benefits the family member directly, not just as a strategy.
The practical move: before the next difficult conversation, try one structured statement in the form of “I notice / I feel / I’d like.” For example: “I noticed you came home late and didn’t have dinner. I felt worried. I’d like us to have a plan for checking in when your schedule changes.” That is the shape of a CRAFT communication, specific, non-accusatory, and actionable.
Rewarding sober behavior
One of the most counterintuitive elements of CRAFT, for families accustomed to waiting for rock bottom, is its emphasis on actively reinforcing sober or reduced-use behavior. The instruction is not to praise sobriety in general. It is to make the experience of not using noticeably and immediately better than the experience of using, in concrete ways that the family member controls.
This is behavioral science applied directly to a household. A 2006 study by Smith and Meyers, examining the reinforcement component of CRAFT, found that CSOs who consistently applied positive reinforcement to sober behavior reported faster reduction in their loved one’s substance use compared to CSOs who focused primarily on consequence-setting. The mechanism: behavior that gets reinforced increases in frequency. That principle holds in addiction contexts the same way it holds everywhere else.
In practice, this means identifying what the person genuinely values (connection, a particular activity, physical affection, a calm home) and making those things accessible during sober time. Specific and immediate beats general and delayed every time.
Allowing natural consequences
CRAFT also teaches when to step back. Allowing natural consequences is not punishment and it is not cruelty. It is the deliberate decision to stop absorbing consequences that the substance use disorder has created, and to let those consequences land where they belong.
A 2015 review by Roozen and De Jong, published in Drug and Alcohol Dependence, found that CSOs who learned to differentiate between protective behaviors (enabling) and genuine support reported significantly reduced personal distress over time. The distinction CRAFT draws is practical: consequences the family member absorbs become invisible to the person using. Consequences that land become information.
The action this week is identifying one consequence the family is currently absorbing that belongs to the person using. Not removing it dramatically, not announcing it as a punishment. Just stopping one specific protective behavior and letting the natural result occur.
Recognizing when to suggest treatment
CRAFT teaches families that windows of receptivity exist, and that they are more predictable than they feel in the moment. After a difficult consequence. During a moment of physical discomfort from use. When the person has just expressed regret or shame. These are not random. They follow patterns that become visible once a family member has done enough functional analysis to recognize them.
The skill is timing the conversation about treatment to coincide with low resistance rather than with high emotion on the family member’s part. A 2019 analysis by Roozen and colleagues found that CSOs who had practiced the treatment-entry conversation in role-play scenarios were significantly more likely to successfully invite their loved one to seek help during a genuine window of receptivity. The action: identify one recurring moment in the weekly pattern when your loved one tends to be more reflective and less defended, and plan the conversation for that window rather than the next crisis.
What the research shows about CRAFT outcomes
The headline finding, consistent across more than two decades of trials, is that CRAFT produces treatment entry rates of 64 to 74 percent, compared to 17 to 30 percent for detachment-based and confrontational approaches. The original Meyers et al. trials have been replicated with partners of alcohol users, parents of adult drug users, and in international contexts, with consistent results.
Beyond treatment entry, the CSO outcomes are just as significant. A 2002 study by Miller, Meyers, and Tonigan found that family members who completed CRAFT training reported significant reductions in depression, anxiety, and physical health complaints, and significant improvements in relationship satisfaction, regardless of whether the loved one had entered treatment by the study’s end. This is not a secondary finding. It confirms that the evidence on family involvement improving recovery works in both directions: helping the family member is not just ethically right, it is clinically effective.
The practical summary: families who complete CRAFT training get their loved one into treatment at roughly twice the rate of families using Al-Anon facilitation, and they feel measurably better themselves while doing it.
CRAFT vs. other family approaches
Comparing CRAFT directly to the alternatives reveals clear differences in both philosophy and outcomes.
Against Al-Anon facilitation therapy, CRAFT differs fundamentally in what it asks the family member to do. Al-Anon teaches acceptance of powerlessness and emotional detachment from the loved one’s behavior. The emotional support Al-Anon provides is real and valuable, but the model does not equip families with strategies for influencing treatment entry. The outcome data reflects this: 17% treatment entry versus 64% for CRAFT. For families who want to take action rather than accept the situation as it is, CRAFT is the evidence-based choice.
Against the Johnson Intervention, CRAFT differs in approach rather than intent. Both models aim to get the person into treatment. The Johnson model uses a planned, often surprise confrontation with multiple family members and prepared statements of impact. The research shows this approach, while emotionally compelling, produces lower treatment entry rates than CRAFT and carries meaningful risk of relationship damage and increased resistance. CRAFT’s non-confrontational approach preserves the relationship that makes future conversations possible.
Against ARISE, a more recent model that falls between Johnson and CRAFT in its confrontational intensity, CRAFT still outperforms on treatment entry in direct comparisons, while also producing better family wellbeing outcomes.
The research points clearly in one direction. That does not mean Al-Anon has no value. For family members who need community and ongoing peer support, Al-Anon fills a real need. But as a primary strategy for getting a loved one into care, CRAFT is the strongest tool the evidence supports.
Who CRAFT is designed for
CRAFT is designed for the Concerned Significant Other, the person in the family who is most engaged with the loved one’s situation and most willing to do structured work. That person is often a spouse or partner, but CRAFT has been studied and applied effectively with parents of adult children, adult children of people with SUD, and close friends who function in a family-like role.
One of CRAFT’s important features is that it works even when the person with a substance use disorder refuses to engage. The entire model is built around working with the family member, not requiring the loved one’s participation as a precondition. The family member learns, practices, and applies skills in their own interactions, and the cumulative effect of those interactions shifts the environment in ways that increase the probability of treatment entry over time.
CRAFT is not designed for situations involving active domestic violence or significant safety risk. When a household includes violence or credible threat of harm, safety planning takes priority over treatment-entry strategy, and a different clinical pathway is appropriate. For families dealing with the specific dynamics of supporting a spouse through treatment, CRAFT provides the pre-treatment foundation that structured family therapy then builds on.
How to access CRAFT: finding a trained therapist or program
CRAFT is delivered by trained therapists in individual and group formats. The CRAFT clinical training program certifies therapists specifically in the model, and the Robert J. Meyers Research and Training Institute maintains resources for families seeking trained providers. Searching for a CRAFT-certified therapist through Psychology Today’s therapist directory with the CRAFT filter is a practical starting point.
Some residential treatment programs, particularly in the Western United States, incorporate CRAFT-based family training into their treatment continuum rather than treating family involvement as a one-time family day event. This matters because what family therapy in structured treatment actually involves varies significantly from program to program. Asking a specific question directly is more useful than reading a program’s general description.
When calling a program, ask directly: “Does your family program include CRAFT-based training for family members, or is family involvement limited to education sessions and family weekends?” The answer will tell you quickly whether the family component is integrated into the clinical model or supplemental to it. One call this week, with that specific question, is a concrete starting point.
What to try this week
Before any formal program, before finding a therapist, before making a single call, one piece of CRAFT is available to any family member right now: the functional analysis log.
Pick one incident from the past week involving your loved one’s substance use. Write down what was happening in the two hours before it started. Note what emotion or situation seemed to trigger it. Write down what happened immediately after, what seemed to relieve or reward the behavior in the short term. Then write down what the longer-term consequences were, for the person using and for you.
That is one entry. Do it for seven days. At the end of the week, read what you have written and look for a pattern. You will almost certainly see one. That pattern is the beginning of a functional analysis, and it is the foundation on which every other CRAFT skill is built. You do not need a therapist to do this first step. You need a notebook and seven minutes a day.
When you are ready to move further, and the data will make you ready, the full CRAFT model is waiting.
Frequently asked questions
How long does CRAFT training typically take?
Most CRAFT programs run between 12 and 16 sessions with a trained therapist, though some structured formats are shorter. The model is designed to be delivered in a focused timeframe, not as ongoing indefinite support. Many families begin seeing behavioral changes in their loved one’s substance use before the program is complete, often within the first several weeks of applying the skills consistently.
What if my loved one finds out I am doing CRAFT training?
CRAFT does not require secrecy, and in many cases, transparency about the process is not harmful. The skills CRAFT teaches, better communication, strategic positive reinforcement, allowing natural consequences, are not manipulative. They are relationship skills grounded in behavioral science. Some families find that being open about the fact that they are working with a therapist to improve how the household functions actually reduces defensiveness rather than increasing it.
Can CRAFT work if my loved one has already refused to go to treatment multiple times?
Yes. CRAFT was specifically designed for situations where the person with a substance use disorder is not yet willing to seek help. Prior refusals do not reduce CRAFT’s effectiveness, because the model works by changing the environment the person lives in over time, not by relying on a single decisive moment. Families with loved ones who have repeatedly refused treatment are exactly who CRAFT was built for.
Is CRAFT the same as enabling with extra steps?
No, and the distinction is important. Enabling means absorbing consequences, removing barriers the substance use disorder has created, and shielding the person from the results of their behavior. CRAFT is the opposite of this in several of its core components. It teaches families to allow natural consequences to occur rather than preventing them, while simultaneously reinforcing sober behavior in specific and strategic ways. The model is built on behavioral science, not on making the person comfortable in their addiction.
Do both partners need to participate in CRAFT for it to work?
No. CRAFT is designed to be effective with a single Concerned Significant Other. The research trials that established CRAFT’s outcomes were conducted with one family member doing the work, not requiring participation from the person with the substance use disorder or from multiple family members simultaneously. One trained family member, applying the skills consistently, is sufficient to produce the outcomes the research documents.
How is CRAFT different from what a family learns in al-anon?
Al-Anon offers community support, a peer network, and a framework centered on accepting that you cannot control another person’s choices. For emotional sustainability over the long term, that peer community has real value. CRAFT differs in that it teaches active behavioral strategies for influencing treatment entry, rather than focusing primarily on the family member’s acceptance and detachment. The two approaches are not mutually exclusive, but they operate from different premises and produce different outcomes on the specific question of getting a loved one into treatment.
