Few things generate more interest than new treatments, particularly for impairing disorders, like attention deficit hyperactivity disorder (ADHD), where considerable treatment data already exist. Gevensleben and colleagues evaluate a novel, non-pharmaceutical treatment for ADHD, in an article that raises two broad sets of questions: How, exactly, do new treatments emerge? How do we integrate data on these new treatments with the data on established ones?

In some ways, our field has enjoyed success over the past 30 years: applications of standard nosologies have elucidated the impact on public health from pediatric mental illness. This, in turn, has stimulated research on treatment, which has brought rigor to evidence-based practice; advances in allied disciplines, such as neuroscience and genetics, raise hopes that further treatment advances are rapidly approaching. In other ways, however, our successes remain modest: most treatments emerge through serendipitous discovery; other biomedical fields generate novel treatments through a stronger interplay between serendipity and theory. Failure to develop highly-effective, theory-driven treatments for mental disorders has led some leaders to view progress from a relatively sobering and critical perspective (Insel, 2009). Moreover, far more resources are devoted to development of pharmaceutical as opposed to other treatments.

Against this backdrop, Gevensleben et al. test a novel non-pharmaceutical treatment. Relative to available treatments, this new one emerges more directly from theory, in this case theory linking ADHD to perturbed regulation of cortical excitability. The new treatment utilizes computer-based training to teach children to regulate cortical excitability, modeling other approaches where patients use feedback from physiological signals to learn how to control the processes that generate the signals. Whereas the general term “biofeedback” refers to treatments targeting various signals, the term “neuro-feedback” applies to treatment targeting neural signals.

This work generates a first set of key questions: how, exactly, do new treatments emerge? such treatments emerge through an orderly progression that begins with systematic but uncontrolled studies. These initial studies typically extend serendipitous discoveries by astute clinicians, but they also can involve, as in the case of neuro-feedback, more direct applications of a theory. In this latter instance, the treatment targets a specific process that the theory links to an illness. Whether the initial stimulus is clinical observation or theory, however, conclusions on efficacy ultimately rest on results from the same tried-and-true process: a series of rigorously conducted randomized controlled trials (RCTs). In these RCTs, two groups are randomly assigned either to the new, “active” treatment or to a controlled treatment that does not contain the “active” ingredient but otherwise resembles the active treatment in as many ways as is possible. Unfortunately, despite the availability of this well-worn path to success, the path is often ignored: strong enthusiasm for new treatments often outpaces results from RCTs. Such a disconnect between research and practice applies even in the current instance, for neuro-feedback, where enthusiasm for clinical utility in ADHD emerged well before publication of the current work, the first reasonably-designed test of therapeutic efficacy in ADHD.

So where does this leave neuro-feedback as a treatment for ADHD? As suggested by Gevensleben and colleagues, prior research in this area does not include important features that characterize the most rigorous RCTs, and the current article usefully fills this gap. The new study uses random assignment to “active” or credible controlled treatments with a reasonably sized sample; comprehensive, multi-informant evaluations are conducted with appropriate blinding; and similar expectancies of improvement manifest in the two groups. Thus, the methods are strong and appear decidedly better than in prior work on the efficacy of neuro-feedback in ADHD. This extends a notable theme in research on novel treatments, both in ADHD and in pediatric anxiety disorders, where extensions of theory are gaining prominence (Klingberg et al., 2005; Pine, Helfinstein, Bar-Haim, Nelson, & Fox, 2009) Moreover, the overall trend in the current study was for neuro-feedback to outperform the control treatment. That said, none of the effect sizes for between-group comparisons extended beyond the range of “medium”. Therefore, the effect cannot be considered robust and of unequivocal “clinical significance” (Kraemer et al., 2003). Thus, we can view research on neuro-feedback as potentially promising, but Gevensleben and colleagues provide only the first rigorous test of efficacy and do not generate a large clinical effect. Until effects are convincingly replicated, we are not ready as a field to recommend neuro-feedback as a therapy for ADHD.

This conclusion, in turn, leads to a second set of questions: How do we integrate data on newly emerging treatments with the data on established ones? Because so few promising new treatment have emerged in recent years (Insel, 2009), our field has had relatively little experience with such questions. Nevertheless, because we know as much about treating ADHD as any other condition, these questions can be framed particularly precisely. Most experts view stimulant medication as the single most efficacious first-line treatment for ADHD, findings that frame questions emerging from Gevensleben and colleagues. Hence, we first can begin to compare the magnitude of the effect in stimulants versus neuro-feedback. In this area, the therapeutic effect for stimulants is clearly better replicated and appears to be larger in magnitude than for neuro-feedback, though definitive conclusions must await the results of further research.

These data also cast in relatively bold relief other related questions. Namely, is it reasonable, given the magnitude of the effect seen with stimulants, to search for new treatments that are better or even comparable in magnitude in terms of clinical benefit? Moreover, if we accept the search for treatments of lesser benefit as important, how small of an effect is worth pursuing? Each of these questions pertain to issues of clinical significance, an area where consensus continues to evolve (Kraemer et al., 2003), and to concerns about the use of medication with children.

Finally, if the effect of stimulants is so strong and well-replicated, why do we need new treatments? Probably only this final question is readily answerable in light of current knowledge: we need new treatments because many children still do not respond sufficiently well and because many parents prefer non-pharmaceutical treatments. This is especially salient in light of recent evidence that ADHD can be identified in preschoolers and that stimulants are used in this age group. Moreover, even for those children who do respond, we still do not understand with sufficient clarity how, exactly, it is that stimulants change brain function in children with ADHD. Such an understanding would allow us, from a stronger scientific position than is currently possible, to more fully explain stimulants’ beneficial effects and to more completely address parents’ concerns about stimulants’ potential adverse consequences. Clearly, evidence establishes the favorable risk-benefit ratio in ADHD. However, some parents and clinicians may prefer treatments of lesser efficacy, in exchange for fewer concerns about adverse effects.

So how can we compare the relative merits of stimulants and new treatments, such as neuro-feedback? Answers about questions on comparative efficacy arise from head-to-head RCTs, studies which might follow from Gevensleben and colleagues’ work. Until results from comparative efficacy trials are forthcoming, however, stimulants should remain the first-line treatment for ADHD. Moreover, even after we do obtain such results, history teaches us that answer to questions addressed by comparative efficacy studies do not come quickly or easily. Indeed, considerable prior research already compares stimulants and other non-medication therapies. Even in the face of such data, experts still can disagree (Klein & Jensen, 2001).

In closing, it is a treat to see RCTs on novel treatments, like the one from Gevensleben and colleagues, since this provides readers with a novel opportunity to take stock of how we come to evaluate efficacy data. When considering where the field has come and where it is going, one cannot help but discern a dual message in recent research. The field has generated an array of treatments for ADHD, including stimulants, which produce large, clinically-meaningful therapeutic effects. Nevertheless, we do not fully understand available treatments’ mechanisms of action or their full range of potential adverse effects; we have yet to develop novel treatments, which target disease mechanisms and produce clinical effects as large as those that manifest with stimulants for ADHD. Gevensleben and colleagues provide a welcome foray into this arena, by showing benefits for a theoretically-driven experimental therapy in ADHD. Critical evaluation might force us, as a field, to recognize the considerable distance we have to travel before any novel treatment can emerge as a first-line therapy for pediatric mental illnesses. Given the importance of the task and the exciting science it embraces, results from studies such as the one from Gevensleben and colleagues can move us towards this challenge with renewed vigor.

The opinions and assertions contained in this paper are the private views of the author and are not construed as official or as reflecting the views of the NIMH or the Department of Health and Human Services.


  1. Top of page
  2. References
  • Insel, T.R. (2009). Translating scientific opportunity into public health impact: A strategic plan for research on mental illness. Archives of General Psychiatry, 66, 128133.
  • Klein, R.G., & Jensen, P.S. (2001). MTA findings fail to consider methodological issues. Archives of General Psychiatry, 58, 11841185.
  • Klingberg, T., Fernell, E., Olesen, P.J., Johnson, M., Gustafsson, P., Dahlström, K., Gillberg, C.G., Forssberg, H., & Westerberg, H. (2005). Computerized training of working memory in children with ADHD--a randomized, controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 44, 177186.
  • Kraemer, H.C., Morgan, G.A., Leech, N.L., Gliner, J.A., Vaske, J.J., & Harmon, R.J. (2003). Measures of clinical significance. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 15241529.
  • Pine, D.S., Helfinstein, S.M., Bar-Haim, Y., Nelson, E., & Fox, N.A. (2009). Challenges in developing novel treatments for childhood disorders: Lessons from research on anxiety. Neuropsychopharmacology, 34, 213228.