The available data on CBT for adult ADHD suggest that these interventions, as a group, show promise as efficacious interventions—however, more studies of the same treatment and more methodologically rigorous trials are needed. Despite the small overall number of trials, a range of distinct but related approaches have emerged. Can any preliminary conclusions be drawn from existing data regarding the most effective approaches at this stage? If so, common features might suggest promising directions for further treatment development and support for specific clinical recommendations. Caution must be exercised in comparing treatments to one another at this early stage because, 1) most programs have only been tested in a single study, 2) a host of factors could account for differences in effect sizes across studies besides features of the treatment itself, and 3) measures used vary widely across studies. However, this necessary caution does not preclude, at this stage, some well-placed critical thinking about what may work best for adults with ADHD.
To generate hypotheses about the most effective emerging psychosocial treatments, we calculated effect sizes (standardized mean differences; Cohen's d
) for ADHD symptom measures from pre- to post-treatment as reported in the eight published treatment trials described above representing seven distinct treatment packages (). All were structured, skills-based programs where statistics necessary to calculate effect sizes were reported. Outcome measures varied considerably across studies and thus we report the most comparable measures—self- or investigator-report of ADHD total and/or inattentive symptoms using either DSM
-based or other established rating. Because uncontrolled trials using pre-to-post data may overestimate the effect of the intervention compared to controlled trials, we only report pre-to-post effect sizes from the active treatment group from the three randomized controlled trials [20
]. Effect sizes and descriptors for each study are displayed in , rank ordered by magnitude of effect size on total symptoms.
Effect Sizes and Study Characteristics for Published Trials of Psychosocial Treatments of Adult ADHD Arranged by Magnitude of Effects on Total ADHD Symptom Scores
From this preliminary examination, nearly all treatment packages (with the exception of Virta et al. [25
]) resulted in “large” effect sizes (0.8+) [34
] on total ADHD symptoms and results were similar for studies that reported inattentive ADHD symptoms. Overall, these data provide support for these skills-based, psychosocial approaches in the treatment of adult ADHD, with a mean effect size for total symptoms of 1.12 and for inattentive symptoms of 0.99. However note that, as discussed above, this conclusion is based on uncontrolled pre-to-post findings from intervention groups, which may overestimate efficacy. In comparing treatment packages, note that neither number of treatment sessions nor format of intervention (group vs. individual) appeared to be associated with treatment effects. The program by Virta et al.[25
], which showed small effect sizes on self-reported symptoms, covered a broad range of topics, with a new topic or broad skill area introduced each session. It appears that less emphasis was placed on the acquisition, repeated practice, and reinforcement of specific compensatory skills directly targeting core symptom-related deficits (i.e., 5 of the 11-12 sessions). In contrast, the group intervention developed by Solanto and colleagues [22
], showed a much larger effect size and focused each session upon compensatory skills and their repetition. The authors state their belief that, “…development of new, more adaptive habits and functional routines in adults with ADHD demands a certain degree of unambiguous emphasis and repetition so that desired behaviors (e.g., checking a planner every day) ultimately become automatic and no longer dependent on the individual's active executive or decision-making functions.” [22
] (p. 4). This type of repetition of adaptive skills to become habits is also at the core of our approach [33
] which also has a similar effect size to the Solanto intervention. These preliminary findings raise the hypothesis that the “active ingredient” in successful CBT for adult ADHD is the introduction and, most importantly, repetition and reinforcement of compensatory skills that target core symptoms versus covering too broad a range of topics in a treatment at the sacrifice of enough repetition and practice of newly acquired, core skills. After a more solid base of efficacy trials to attain a designation of CBT as an empirically supported treatment for ADHD, comparative effectiveness studies may be able to more definitively test this hypothesis in the future,
Examination of elements common to the most effective treatments also reveal some important themes. Based on total ADHD symptoms, three treatments showed effect sizes more than ½ of one standard deviation above the others () and a closer examination indicates common features that may contribute to treatment effectiveness. They involve teaching of specific skills and strategies and emphasis on practice of those skills outside of session. These are highly structured programs, elements of which include, 1) short-term work, averaging about 10 sessions 2) manualized content, and, 3) inclusion of client handouts or a workbook to guide work outside of session. With respect to the content, the three treatments with the largest effect sizes, 1) focus mostly on learning of compensatory skills to ameliorate ADHD-related difficulties, 2) focus on organization and planning skills, and 3) consider skills to deal with difficulties in motivation. To varying degrees, all three programs also target the role of internal processes (thoughts, feelings) in increasing or decreasing the likelihood of appropriate skill use. For example, Metacognitive Therapy [22
] teaches positive and negative visualization of long-term consequences while Safren's program [33
] includes a module on adaptive thinking skills. Overall, then, these programs can be described as primarily behavioral, incorporating cognitive elements to the degree that these processes block adaptive behavior and skill use. These preliminary findings suggest that psychoeducation alone—even when it covers the range of topics that might be relevant to adults with ADHD—is not sufficient to have a significant impact on ADHD symptoms and that learning and practice of specific skills is critical.