Skip to content
Raising the bar for excellence in CBT. Enroll in Beck Institute CBT Certification today!
Substance Use Disorders

CBT for Alcohol and Substance Use Disorders

By Cory Newman, PhD, ABPP

Therapists who treat patients struggling with alcohol and other substance use problems are familiar with the problems of under-reporting, minimizing, and denial. Under-reporting means that the patient is willing to disclose that he or she partakes of alcohol and/or other substances, but does not report the full amount either because of low self-awareness, self-serving bias, shame, and/or fear of anticipated consequences (e.g., being told by the therapist to stop, or else leave therapy). Minimizingrefers to patients’ spuriously maintaining that their level of drinking and/or using is not hazardous, and/or not any worse than their self-described comparison group (e.g., “Everybody in my frat drinks, and my drinking is pretty tame compared to the others. It’s all expected and normal.”). Therefore they believe that their drinking and/or using is a non-issue. Outright denialrefers to patients who state that they have not used or lapsed, when the evidence shows otherwise. Denial also refers to disavowing a label without adequate self-reflection, such as a patient reflexively stating, “I’m not an alcoholic.” The above terms overlap to some degree.

So what is a good CBT practitioner to do?


When you believe that someone is under-reporting (e.g., they say “I had one or two drinks,” as if they could not remember which is which)1, it is helpful to suggest some formal, written self-monitoring assignments between sessions. Therapists can give the patient the following instructions and rationale: “Let’s get some baseline data on your [drinking, smoking, using] to see what we’re up against. If you can keep track every day, we’ll get a much better idea about where we stand on the issue, and what sorts of changes we might be able to agree to pursue in therapy.” While there is no guarantee that patients will follow through or be accurate, self-monitoring does not require immediate change, and therefore is less imposing of a task. If patients succeed in documenting the numbers, there is an improved chance that the data will be less affected by a self-serving memory bias.


If patients minimize, this usually reflects a dysfunctional belief about the substance(s), such as a “permission-giving” belief (Beck, Wright, Newman, & Liese, 1993; Newman, 2004). For example, they may believe that their drinking or using is not a problem because they:

  1. “Only” drink after work and on weekends
  2. Function well and get things done
  3. Have never had a serious medical issue
  4. Know “plenty of people” who are much worse
  5. Know how to “handle themselves” so they don’t seem too impaired
  6. Only use “safe” substances or stay away from the “hard stuff,” to name a few.

Therapists must be adept at detecting and flagging such beliefs when patients suggest them in passing, and put them on the therapeutic agenda as beliefs to be evaluated, rather than as established “facts.” Again, therapists should be collaborative, presenting their concerns as areas for exploration, rather than as arguments.


When patients outright deny that they have used even though the evidence against them is formidable (e.g., the patient’s last two urinalyses were positive for cocaine), it does little good to engage in a power struggle, such as confronting them in a non-empathic way. Instead, therapists may silently hypothesize that the patients are ashamed and afraid of the consequences of acknowledging a relapse. Yes, there are some patients simply looking to avoid taking responsibility, but many are ambivalent about admitting their drinking and/or using, and would be more willing to admit a relapse if given reason to believe that the therapist would be empathic and willing to continue treating them (see Newman & Strauss, 2003). One way of approaching this problem is by asking the patients if they had any “close calls” or “spikes in [their] craving” this week. Patients may be more willing to couch their problem in these terms, and the therapist can ask pertinent questions about high risk situations, the patients’ decision-making, how they “coped,” how to steer clear of such close calls in the future, and other topics that are germane to an authentic agenda in providing CBT for substance misuse, even if the patients do not make a full admission.

If therapists decide to be more direct about the evidence for a relapse, they can discuss their own struggle in how to address the issue while still maintaining the therapeutic relationship. For example, the therapist may say, “Part of me wants to believe you when you say that you have not relapsed in any way, because I know you have worked hard in therapy, and I appreciate your efforts. Another part of me feels I owe it to you to give you the benefit of my best clinical judgment, and right now that judgment tells me that you’ve had a slip and need help acknowledging it so you can get back on track. It’s quite a dilemma, because I value our therapeutic relationship and I don’t want to have a disagreement with you, but I also don’t want to be so agreeable that I miss something important and shortchange your treatment.” Therapists can emphasize that “trust is not always a perfect thing,” and that an episode of doubt can be worked out if both parties value the relationship.

When a patient denies that he or she is an “alcoholic” or “addict,” it is not necessary to get into an “all or none” battle between “Yes, you are,” and “No, I’m not.” Instead, the therapist can note that regardless of how apropos the label may or may not be, they can look at the behaviors and the consequences to determine if the patient is trapped in a pattern of drinking and/or using that causes problems on an ongoing basis. The therapist can say, “Let’s just examine cause and effect in your life to see if alcohol and other drugs are harming you, and to determine how you perceive that, and what you could do about it. If you’re suffering consequences but you’re having difficulty making changes, that’s a red flag that there’s a problem we should probably work on.” This sort of rationale does not require a labeling of the patient, and therefore may be more acceptable.

1The answer “once or twice” is rarely accurate. When someone says, “I used once or twice,” it probably means “three or more,” just like when a teenaged boy says he’s had sex “once or twice,” it probably means zero. Either way, they would certainly remember the difference between “once” and “twice” if either number were accurate!


Beck, A. T., Wright, F. D., Newman, C. F., & Liese, B. S. (1993). Cognitive therapy of substance abuse. New York: Guilford Press.

Newman, C. F. (2004). Substance abuse. In R. L. Leahy (Ed.), Contemporary cognitive therapy:Theory, research, and practice (pp. 206-227). New York: Guilford Press.

Newman, C. F., & Strauss, J. L. (2003). When clients are untruthful: Implications for the therapeutic alliance, case conceptualization, and intervention. Journal of Cognitive Psychotherapy: An International Quarterly17, 241-252.