Cognitive behavioral therapy for substance use: how it works

Cognitive behavioral therapy, or CBT, is a structured talking therapy that links what a person thinks and does to their substance use. NIDA describes it as helping people recognize, avoid and cope with the situations in which they are most likely to use drugs. It is one of the most studied psychological treatments in this field, and the evidence is encouraging but more modest than people often assume.

What CBT is for substance use

NIDA says behavioral therapies like CBT can help a person gain control over stressful emotions and thoughts that lead to wanting to use, and strengthen resolve when facing cues such as people, places, things and moods linked to past use. A 2010 review by McHugh, Hearon and Otto in Psychiatric Clinics of North America describes CBT for substance use disorders as combining operant learning strategies, cognitive and motivational elements, and skills building, aimed at overcoming the powerfully reinforcing effects of substances.

Trials have tested it for alcohol, cannabis, stimulants, opioids and combinations of drugs, alone or alongside other treatment such as medications.

Core ideas: triggers, thoughts and coping skills

• Functional analysis. The review says this step focuses on identifying and preventing high-risk situations, such as favorite bars or friends who also use, in which a person is more likely to use.
• Triggers. A trigger is a cue that sets off craving or use. Therapy looks at what comes before use and ways to avoid or manage those cues.
• Thoughts. NIDA says CBT aims to alter negative or unhealthy thought processes that lead to substance use. The review lists cognitive restructuring among the techniques.
• Coping skills. The review describes building problem-solving, drug refusal and coping skills, including for emotions and relationships.

The same review notes that CBT often borrows from other approaches, including motivational interventions, contingency management and relapse prevention, which is covered on the relapse prevention page.

What sessions are like in outline

McHugh and colleagues say CBT sessions use a regular structure, including agenda setting, identification of goals, and the assignment and review of homework. The length of treatment varies widely, and the review mentions formats ranging from a single session to courses of 12 sessions. The details of a particular program should be asked of the provider.

What the research says

A 2019 meta-analysis in the Journal of Consulting and Clinical Psychology, by Magill and colleagues, pooled 30 randomized controlled trials with 5,398 participants. The trials covered alcohol (15), cannabis (3), opioids (2), stimulants (6) and several drugs together (6). The researchers compared CBT against three kinds of comparison and reported results as Hedges g, a standardized measure in which larger numbers mean larger differences.

• Against minimal treatment or a waitlist: g of 0.58 for use frequency at early follow-up and 0.44 at later follow-up.
• Against nonspecific therapy: a smaller difference, g of 0.18 for frequency at early follow-up, and no significant difference at later follow-up (g of 0.05).
• Against another specific therapy: essentially no difference (g of negative 0.02 for frequency at early follow-up).

In plain terms, CBT did better than doing little or nothing. It did only a little better than general supportive therapy in the short term, and it performed about the same as other specific treatments. The authors also reported that three influential studies drove the result against nonspecific therapy: removing them made it non-significant.

Earlier, McHugh and colleagues wrote that evidence supports the durability of treatment effects over time, giving one cocaine study in which 60 percent of patients in the CBT condition had clean toxicology screens at 52 weeks.

Limits to keep in mind

• Magill and colleagues found that reporting of therapist training (44 percent of studies), supervision (70 percent) and fidelity (7 percent) was variable, so it is hard to know how closely trial therapy matched the intended model.
• McHugh and colleagues note that most substance use treatment happens in service settings far from the controlled conditions of trials, and that matching people to treatments from their characteristics has not yet produced clear guidance.
• Results differ by substance and by the comparison used, so no single number describes how well CBT works for everyone.
• NIDA states that treatment should address the whole person. CBT is often one part of a plan that can include medication and support for other health and life needs. See co-occurring mental health conditions.

Another motivation-focused option is motivational interviewing. For the bigger picture of what to expect, see what happens when you ask for help.

Questions to ask a clinician

• Is the therapy I am offered structured CBT, and how will it deal with my triggers and thoughts?
• How many sessions are expected, and what homework is involved?
• How will we measure whether it is helping?
• Would medication or another approach alongside CBT suit my situation?
• What training and supervision does the therapist have in CBT for substance use?