The Efficacy Document: A Celebration and a Critique

by Siegfried Othmer | June 23rd, 2004

The AAPB has just published the document “Evidence-Based Practice in Biofeedback and Neurofeedback,” by Carolyn Yucha and Christopher Gilbert. It has been more than ten years since the AAPB has issued an official statement of conditions where efficacy of biofeedback is recognized by the organization. Since that earlier document, we have witnessed the continuing thrust toward evidence-based medicine. As a professional community, there was no choice but to respond in a fashion such as this.

So let’s look at the good news: First of all, the document puts together under the same covers the case for neurofeedback and for peripheral, somatic neurofeedback. The document makes the case for “self-regulation” as the active ingredient (page 2), irrespective of how it may be achieved with biofeedback. The document acknowledges that evidence-based practice must take into account not only efficacy in controlled studies but effectiveness in actual clinical practice. It forthrightly acknowledges the methodological problems confronting biofeedback when it is placed in random controlled studies under blinded conditions.

The document then proceeds to review the scientific evidence, narrowly constructed, for as many as forty conditions, assigning each to one of five categories of claimed efficacy. The first category is “Not Empirically Supported.” This terminology is suggestive of some kind of evidence that militates against the claim of biofeedback efficacy, an unfortunate implied bias. What is really meant here is that the evidence is insufficient at this point to make a stronger claim, not that there is adverse evidence.

The conditions listed under category 1 are as follows:
Level 1: Not Empirically Supported
Eating Disorders
Multiple Sclerosis
Spinal Cord Injury

Remarkably, the most stunning long-term success in all of biofeedback, namely neuromuscular re-education for spinal cord injury, is here reduced to the lowliest category. A person with no other hope of remediation, and contemplating visiting Bernard Brucker’s clinic in Miami, is therefore up against the collective judgment of our professional community to the effect that his ground-breaking work, building on excellent prior research of John Basmajian, is not empirically supported. One must wonder about a process that leads to such an outcome.

The inappropriateness of this listing highlights the methodological flaw that is perniciously at work in this construction of the efficacy hierarchy. Proof of efficacy is by itself insufficient to also prove clinical effectiveness, as the document makes clear (p. iv). But the converse is not true. Clinical effectiveness is quite sufficient to prove efficacy. The cumulative record of Bernard Brucker leaves clinical efficacy of his approach of EMG training in no doubt whatsoever. None. To imply doubt where none exists is to be needlessly pedantic. If one’s assumptions lead to such an absurd conclusion, one must question the assumptions. (This is elegantly done, by the way, in an article by Thomas Rossiter in the current [June 2004] issue of the AAPB Journal, where the case is made for efficiency studies along with efficacy research.)

Curiously, Brucker may have become a victim of his own success. Effectiveness of his technique is so obvious on inspection that there is simply no further interest in mounting a controlled study. This has also happened in medicine, where for one particular kind of rare cancer a potential remedy was uncovered accidentally in one study. The benefit was so obvious that no placebo group could ethically be recruited subsequently, and no controlled study has ever been done. No one questions efficacy.

Level 2 of the efficacy hierarchy is Possibly Efficacious. This term serves as yet another implicit indictment of the conditions listed under Level 1, which for one reason or another don’t even rise to the level of “possibly efficacious.” No doubt they should be given wide berth, and one should save one’s money. Imagine approaching an insurance company about supporting a treatment that the practitioners themselves aren’t willing to list as “possibly efficacious.”

Level 2: Possibly Efficacious
Cancer and HIV
Cerebral Palsy
Chronic Obstructive Pulmonary Disease
Cystic Fibrosis
Depressive Disorders
Diabetes Mellitus
Foot Ulcers
Hand Dystonia
Myocardial Infarction
Irritable Bowel Syndrome
Mechanical Ventilation
Motion Sickness
Post-Traumatic Stress Disorder
Raynaud’s Disease
Repetitive Strain Injury
Urinary Incontinence in Children

Fibromyalgia is a case in point of a condition that has been successfully treated with a combination of peripheral and central (EEG) biofeedback. The same is true of incontinence, although the research record does not yet reflect this. Asthma has also shown improvement with traditional biofeedback, heart rate variability training, and EEG neurofeedback. Each of these conditions has the benefit of several independent strands of evidence for the central proposition of responsiveness to a self-regulation-based intervention. More generally, each of these conditions, and many others with them, indicate the benefit of a unitary conception of the self-regulation strategy, one that includes all of the various modalities, including in particular neurofeedback.

This highlights a shortcoming of efficacy criteria in which each constituent study must “stand on its own feet,” so to speak. In the abstract, one would argue that greater credence should be given to entirely independent strands of evidence. This holds true even if each of the strands could not stand entirely on its own by efficacy criteria. Such complementary evidence should be admitted particularly when one is talking about weak claims, i.e. possibly or probably efficacious.

The third category is Probably Efficacious:

Level 3: Probably Efficacious
Alcoholism/Substance Abuse
Chronic Pain
Fecal Elimination Disorders
Headache –Pediatric Migraine
Traumatic Brain Injury
Vulvar Vestibulitis

This list should probably be a lot longer, but for the moment we shall try to be happy with what we have been given. It is a mystery, however, why epilepsy is not ranked as efficacious at the least. No condition has received more quality research support in neurofeedback than epilepsy, and this efficacy has been confirmed in clinical work for more than 25 years. Now there is additional evidence of responsiveness to GSR training as well.

Also, with regard to pediatric migraine: The evidence is that children are even more responsive to biofeedback than adults. Why is adult migraine ranked higher? If efficacy for adult migraine is rigorously established, and a single study shows pediatric migraine to be even more responsive to biofeedback, why is this not sufficient? Why must pediatric migraine stand entirely on its own? This is simply perverse.

Level 4 is Efficacious. The surprise may be the brevity of the list. It is certainly shorter than the list of conditions for which efficacy was claimed back in 1992 by Schellenberger et al. At that time, neurofeedback efficacy was recognized for both epilepsy and ADHD. The down-grading of epilepsy by our own community must be regarded as a significant step.

Level 4: Efficacious
Attention Deficit Hyperactivity Disorder
Headache –Adult
Temporomandibular Disorders
Urinary Incontinence in Males

I suspect that most biofeedback and neurofeedback clinicians will find little overlap between their current practice and this list. Only one category, anxiety, fits comfortably within a conventional mental health practice. Given such an extreme disconnect, it is not likely that the official efficacy document will circumscribe and govern clinical practice in the manner of the DSM-IV, for example. Rather, it will simply mean that clinicians will increasingly go their own way, as they already have done.

Finally, Level 5 refers to Efficacious and Specific. The criterion is as follows: “The investigational treatment has been shown to be statistically superior to credible sham therapy, pill, or alternative bona fide treatment in at least two independent research settings.” Curiously, this criterion is not meaningfully distinguishable from Level 4. For example, Level 4 stipulates “the investigational treatment is shown to be statistically significantly superior to the control condition or ….equivalent to a treatment of established efficacy…”(4a). It also stipulates “superiority or equivalence…has been shown in at least two independent research settings” (4f). Thus any condition meeting Level 4 criteria with respect to sham treatment also meets Level 5. In particular, there is no criterion offered by which specificity might be demonstrated beyond Level 4.

What then are the conditions for which biofeedback of any stripe is both efficacious and specific? Here is the list:

Level 5: Efficacious and Specific
Urinary Incontinence in Females

The only condition for which evidence is deemed to be of the first rank is urinary incontinence in females. Nothing else measures up. Given the fact that criteria for Level 5 are essentially indistinguishable from Level 4, there is apparently some criterion off the page that has led to this truncation. I take this as further evidence of an undisclosed agenda by the authors. We are left with something that most mental health professionals cannot get excited about, and have no particular interest in.

Consider what is not mentioned here at all. Myofascial Pain Syndrome, on which a White Paper has previously been issued by the AAPB, only appears as a subcategory of chronic pain (which in turn is listed only as “probably efficacious”). Torticollis is not listed. Bruxism only appears as a subcategory of Temporomandibular Disorder. Hypoglycemia is not mentioned, although it might be considered under diabetes. Among the sleep disorders there is only insomnia. No mention is made of nocturnal myoclonus, restless leg syndrome, narcolepsy, sleep apnea, night terrors, nightmares, night sweats, snoring, nocturnal bruxism, and somnambulism. Parkinsonism is not listed. Nor is Asperger’s, or dementia. Relative to ADHD no mention is made of Oppositional-Defiant Disorder or Conduct Disorder or Episodic Explosive Disorder. If truth be told, it is the behavioral disorders rather than inattention or mere impulsivity that volatilize little boys out of the classroom and get them into our offices. ODD and CD is what we get to treat more than garden-variety ADHD. How can this be overlooked? Anxiety is treated, but not Panic Disorder. Bipolar Disorder is missing. So is Reactive Attachment Disorder, as well as Dissociative Identity Disorder. Specific learning disabilities are excluded.

It may be quite true that most of the above-listed conditions don’t have enough published evidence to merit listing by established efficacy criteria. But the result is a gross distortion of the day-to-day realities of this field. At some point, one must start to question assumptions that give rise to such manifest distortions.

So what is going on here? This field has played out one of the grand battles within the field of psychology, that between the research psychologists and the clinical community. At every juncture, the research mentality has attempted to re-establish hegemony over a field that has experienced a huge diaspora into clinical practice despite a very hostile climate. The movement toward evidence-based medicine provides the official sanction by which research criteria may once again be installed as the arbiters of what is acceptable in clinical practice. In addition to our playing a role in this grand battle within psychology, our community also has its particular pathologies and familial dysfunctions.

If one takes a mental health perspective on our own field, one might say that ours is a traumatized community, and we have taken to self-mutilation. We are dying the death of a thousand cuts, only to find that most are self-inflicted. Trauma disregulates; our community is perhaps the most disregulated among mental health communities. To quote a recent attendee at one of our professional conferences: “For a community oriented toward mental health, I don’t see a lot of health.” The organization may now be composed largely of people who have been intimidated by the success of pharmacology, and who fundamentally do not understand or buy into the Disregulation Model or the Self-Regulation Remedy. We may as a community have succumbed to a collective inferiority complex. We hope for crumbs from the table of our betters. If our claims are sufficiently modest, perhaps they will finally pass muster.

I suspect that most biofeedback professionals would not see the efficacy hierarchy as representing an accurate picture of what they know to be true from their own experience. If this is the best we can go on to make our case to third-party payers and to fellow health professionals, this field could be doomed. Fortunately, we are not dependent on either reimbursement or the good will of other health communities in order to make progress and establish ourselves. Ours is already a force within the health care field. Neurofeedback alone accounts for $200M per year in revenue in my estimation. Peripheral biofeedback, with all of the incontinence work included, is presumably still bigger.

This progress has occurred largely without the sanction of the leaders of the field, whether on the political or the research side. So in charting our progress into the future, we would do well to be guided by what has gotten us this far. As a first order of business, we should resist the obvious message implied by the efficacy document, which is to intimidate the clinical community into compliance. To help stiffen our spine, we can also benefit from the history of prior scientific controversies.

Several examples come to mind. The most immediate one is the science of climatology. Here we have a situation where almost all of the evidence favors the global warming hypothesis, yet controversy persists because the theory inveighs against entrenched economic interests. Somehow, in the face of all the evidence, a stance of scientific doubt is maintained, with the active collusion of captive scientists. (Note 1) Skepticism, always the highest of scientific virtues, is elevated to the point where it takes on a life of its own. Individual research papers are disputed or assailed, thus sustaining a posture of fundamental ambiguity. Yet this sense of ambiguity is not shared by the vast majority of climatologists. The remedy for this state of affairs was a consensus statement by the Intergovernmental Panel on Climate Change. Here was a chance for a consensus statement to emerge which made clear that disagreements among most climatologists with respect to global warming were at the margins and not at the core. Our professional community could coalesce around a similar consensus statement to great mutual benefit.

Yet another example is evolutionary theory. Here is another instance of a science under siege because of forces from outside the discipline. First of all, the attack on evolutionary science was based on acknowledged uncertainties about mechanisms, which were elevated to indicate a fundamental ambiguity about evolution itself. Because of issues that lie outside of the realm of science, controversy has dogged the field far beyond any fundamental remaining scientific ambiguity. Conversely, it is likely that the pressure under which evolutionary science had to operate was not conducive to the best science being done, as paleontologists themselves acknowledge. Actual uncertainties within the field and unresolved issues were probably submerged somewhat in the face of the critical onslaught. The same is true in biofeedback. The best science cannot be done under siege. The message to us is that the existence of an adversarial climate should not divert us from what we believe to be good science and sound practice.

A third example is environmental science. In the early days, “scientific criteria” were to be relied upon to determine just how much environmental cleanup needed to be done in order to reach certain goals with respect to our water and air quality. Once the battle had been moved to the scientific turf, policy was effectively check-mated. Industry could always raise ambiguities that postponed policy decisions. The scientific process, when conducted in an adversarial context, could never be moved along sufficiently to resolve the important issues. The disagreements went on for many years. It was only when implementation of the “best available technologies” was mandated that the dam broke and environmental cleanup started to happen. Similarly, in biofeedback we must simply declare in which areas we consider the self-regulation remedy to be the “best available technology,” and then move forward to have that implemented as social policy. In this enterprise, we must give appropriate weighting to efficiency studies or other outcome data, along with formal efficacy studies.

A fourth example is the risk of smoking tobacco. Back in 1958 a long-term, large-scale prospective study on the hazards of smoking was published in the American Scientist. This prospective study, covering perhaps twenty years, established solid scientific evidence indicting smoking tobacco in a variety of health concerns. The statistical evidence was overwhelming, and convincing down to the least detail. Comparisons had been made with non-smoking populations such as Seventh-Day Adventists. Urban populations were compared with rural to quantify cumulative effects. The beneficial effect of quitting smoking was studied in detail as a function of time. Pipes and cigars were compared to cigarettes. The study left no ambiguity: Cigarette smoking contributes to lung cancer risk. Yet despite this incontrovertible evidence, the tobacco companies managed to maintain a sense of uncertainty for decades on this issue (and the Japanese Government does so to this day). They promoted their cause with scientific prostitutes. Commercial interests trumped good science. The field of biofeedback finds itself in a similar situation. The science is not really our zone of weakness, and scientific arguments will not win the day. We can solve significant societal problems, and our appeal should be to those whose problems we can solve. Our appeal must be to the pragmatists in our society, not to the scientists.

Another example is the current controversy about the potential hazards of vaccines in general, and of the MMR vaccine in particular. For as long as I can remember, the official position has been that vaccines are generally safe, but that some small percentage of children can be severely affected. These hazards were always forthrightly acknowledged. With parents raising the issue of autism as a possible consequence, “science” is invoked by officialdom to exonerate the vaccines in general, and the mercury-containing thimerosal preservative in particular. With damning evidence gathering, officialdom retreats to the least defensible and truculent position, namely that the vaccines are categorically safe. How could that happen? When did they suddenly become safer than they were before? Or one acknowledged that bad things can happen, but autism is not among them? Unstated is the fact that this conclusion can be drawn only on the simplest of models, namely the assumption that the MMR vaccine directly causes autism. Only then is the epidemiology tractable.(Note 2) Ruling out that simple connection, however, does not exonerate the vaccine or thimerosal. There could simply be a more complex causal chain in which the vaccine is merely one link. Evidence from mice is coming along on just that point. One genotype may be vulnerable, for example, while others are not. (Note 3) Here we have another instance in which science, when placed in a political context, suffers a loss of integrity and becomes a mockery. The real issue is a political one, with the scientific issue just serving as a political football.

Finally, and closer to home, there is the issue of the use of anti-depressants in children. The fact is that nearly all studies of anti-depressant use in children have found no benefit over placebo. For many years there was not a single study that demonstrated such benefit. This did not constrain medical practice one iota. It was simply not discussed. Now that a single study has found efficacy for Prozac, it is pounced upon as proof of efficacy. There is no reference to “the preponderance of evidence” in this case. Conveniently, only this study counts. Where is scientific integrity lodged here?

One anecdote from the past neatly draws the threads together: Years ago Barry Garfinkel, prominent psychiatrist at the University of Minnesota, was one of the leading lights in the field of ADHD whom we wished to persuade of the efficacy of neurofeedback. He and I exchanged a number of letters at the time. Shortly thereafter he disappeared from the scene. He had been convicted of research fraud in a quarter million dollar study of anti-depressants! We might have wasted a lot more effort to persuade someone who was in fact unpersuadable as an acolyte of Big Pharma. I suspect an equivalent lack of integrity on the part of many of our adversaries. Unfortunately, most are too slick to be exposed. Practically the only way is to catch them in a contradiction. Russell Barkley once said about neurofeedback: “It doesn’t work, and besides that, it doesn’t last!” Clearly it has to have “worked” initially in order to fail subsequently. The contradiction? Draw your own conclusion. A posture of feigned skepticism can hide intellectual dishonesty as effectively as a priest’s garb can hide wayward personal predilections.

In fact, where in medical practice is anyone losing sleep over the fact that such techniques as ECT were not put in place subsequent to suitably controlled studies. If placebo controls are so important, why were they not also part of the $6M Multi-Site Study on ADHD? Who is complaining about the lack of studies prior to the administration of the latest atypical anti-psychotics to 5-year-old children? Once something like clonidine is well established in clinical practice, the lack of controlled studies is no longer determinative for events. Eventually the sheer bulk of clinical practice confers a kind of usage validity, and that should hold as well for the technique of biofeedback. We can gain a lot of elbowroom as a profession by simply declaring ourselves in that regard. In any event, we should reject being the only modality currently being held to such unrealistic standards as a precondition for getting any kind of hearing.

The lesson I draw from the above examples is that whenever significant political or economic factors are involved, the ostensible scientific issue is not allowed to be settled on scientific grounds. Science becomes a way of sustaining controversy rather than of resolving it. At that point, it no longer matters how good the science is. For us to be diverted to deal with the science when nobody is going to allow the process to work is a distraction. As David Kaiser once said, “We are not operating on a level playing field.”

Secondly, when it comes to the science, we should not allow the adversaries to set the terms of debate. A new paradigm can be established in new ways that are appropriate to it. This is most obvious in mathematics. No one tells the mathematician how his theorem is to be proved. When a math paper is being reviewed for publication, other mathematicians have no choice but to follow the lines of the proof through from beginning to end, no matter how far it strays from conventional patterns of thought.

Jean Achterberg and Frank Lawlis said in 1992 about randomized and controlled trials: These were “designed chiefly for agricultural and horticultural studies…. Randomized control group designs have not and simply cannot yield satisfactory answers in complex behavioral and psychological studies with human beings…To let a research methodology dictate the design and nature of the research question (instead of the other way around) is an aberration of the scientific method. The tail, quite plainly, is wagging the dog.” (From “Remarkable Recovery” by Carlisle Hirshberg & Marc Ian Barash. G. P. Putman’s Sons, 1995, p. 363)

We must have the liberty to prove our concepts any way that they can be proved. We do not need to yield to convention. In our case, we already have proved them. The scientific controversies about neurofeedback and biofeedback are largely resolved. What remains is mere tactics and implementation. It is important to distinguish between genuine scientific controversy and professional ignorance. We are of course largely dealing with the latter. Our working harder on the science does not constitute a remedy for those who are not on the same page with us in any event.

History can be brought to bear on this point. The new book “Prematurity in Scientific Discovery: On Resistance and Neglect,” by Ernest Hook shows how devastatingly the scientific community deals with ideas that appear before their time, that is to say before they fit naturally and incrementally into the progression of scientific thought. Neurofeedback suffers from nothing so much as having been discovered thirty years too early. By now trenches have been dug and positions taken. More science will not get people out of their trenches. They would rather die in their trenches. (But then, as we know, science advances one funeral at a time.) Rather, advances in the cognitive neurosciences will come to the point where what we are saying will simply become obvious. We need not wait for this happy circumstance. Implementation can proceed apace. (Note 4)

This leads me to the final example from the history of science: Plate tectonics. The “empirical evidence” in support of continental drift was quite complete by the early 1920’s. That was not the problem. The real issue was an absence of a viable mechanism by which continental drift might have come about. Absent such a model, the data were rejected for more than forty years, until the mid-Atlantic ridge was discovered, along with the reversal of magnetic domains on the ocean floor. It took geologic history being written out on the ocean floor before a dead theory could finally be revived! Neurofeedback is essentially in the same boat. It was discovered before people understood brain plasticity, before functional imaging, before network models, before people had much of a clue about the EEG, and before chaos theory was elaborated. In retrospect we realize that it was a scientific blunder to neglect the geological data. But as long as there was no theory to work with, what was “science” to do? It is probably prudent for the enterprise of science to set aside—even neglect– “premature” discoveries, but such neglect should not be taken as more than it was. It is similarly a scientific blunder to ignore the voluminous data accumulated by clinicians doing biofeedback. The opposition to our work is historically rooted, just as it was with plate tectonics. It was not the result of wise or competent judgment, not then and not now. It should simply be treated as the side show that it is. It cannot be allowed to sideline or even slow down our work.

It is important at this juncture to understand fully the elements involved in bringing innovation to the society. In his Theory of Economic Development, Joseph Schumpeter distinguishes between invention and innovation, the latter referring to the successful insertion of an invention to the point where it becomes economically successful. It is the insertion of biofeedback into the marketplace that is now the issue. This is largely a clinical burden, and continuing scientific developments are only an accompaniment to this larger task.

Given the history of the staggering failure by the leading organization to move this field forward at a time of a major thrust toward Complementary and Alternative Medicine (CAM) within the society at large, and of a groundswell toward Integrative Medicine within the health professions, the leaders of our field no longer have standing. Not only have they failed to succeed in moving the field forward, but in the process they sometimes actively impeded the natural, organic progression of our discipline. Narrow perspectives kept new initiatives from taking root. The leaders of the field cultivated an adversarial climate that was hostile to innovation. The efficacy document is a final testimonial to the thorough-going disparagement of clinical findings over the more than thirty-year history of the field by hegemonists in exalted places.

A Path Forward

Since the efficacy document so thoroughly distorts what we know to be true about our field, it urgently needs to be complemented with a “Consensus Declaration” from our professional community. In keeping with the fact that this has become a clinically driven discipline, this document should be prepared largely by clinicians who have made a day-to-day living with the practice of biofeedback. After the document is offered up for revisions and alterations, other clinicians and researchers can then sign on to the document. Then this declaration can take its place alongside the efficacy document to represent what we stand for by the two operative sets of criteria—one by formal research criteria, and the other by efficiency and effect-size criteria–by what actually makes a difference in the real world, and by what we actually do.

Such a document will not be bereft of scientific validity. It will simply be based on a different kind of epistemology, one more suited to the nature of what we have discovered, and by the nature of how we work day-to-day and share our findings.

We claim the following: A Self-regulation Strategy is believed to be efficacious for improving any bodily function which is subject to dynamic regulation by the CNS, and about which information can be made available to the trainee. We know of no exception to this general rule, and none has appeared in the literature of this field. All observations of clinically effective biofeedback, including individual cases and research studies, can be adduced in support of this general proposition. The placebo hypothesis cannot be recruited to negate this proposition. The commonplace observation of spontaneous self-recovery in medicine and psychology only supports the view that robust self-recovery techniques are available to us, if we but come to understand regulatory systems, their failure modes and their recovery mechanisms. Through various biofeedback modalities, we have arrived at empirically validated approaches of eliciting the self-healing response whenever deficits in self-regulation are at issue.

A subsidiary thesis is that much of psychopathology is grounded in, or accompanied by, various Disorders of Disregulation. These disregulations affect fundamental regulatory systems. They do not line up with canonical diagnostic boundaries. Consequently, the remedies are directed toward these fundamental disregulations, and are no respecters of diagnostic categorizations, whether with respect to their qualitative distinctions or with regard to levels of severity, i.e. diagnostic thresholds. Even if the condition at issue is not grounded in disregulations, it may be accompanied and exacerbated by them. In these cases, self-regulation strategies may still be supportive of the recovery process, by analogy to psychodynamic psychotherapy. In many other conditions, however, disregulation is believed to be the core issue, and a self-regulation strategy may in such cases represent a categorical remedy. That having been said, the efficacy, effectiveness, and clinical utility of self-regulation strategies do not rest upon this proposition. They hold up even if self-regulation strategies are deemed to play mainly an adjunctive role. Whatever mental illness one is heir to is best endured with a stable, well-regulated physiology.

As brain function is increasingly understood, it will come to be recognized that functional models of mental disorders have more explanatory power than structural models. The revolution in Biological Psychiatry has taken place over the last forty years. It has concerned itself exclusively with steady-state neurochemical behavior. The rise of functional medicine, with its emphasis on bioelectrical mechanisms, and on timing, frequency-based analysis, and transient event analysis, will expand the scope of our understanding. The operational arm of this development will be found to be the self-regulation technologies on the psychology side, in conjunction with such techniques as rTMS, vagal stimulators, and deep brain implants on the medical side. This development will occupy us over the next twenty years. The increasing utilization of such techniques promises to lead to a golden age of mental health in which even the most intractable mental health concerns yield to our interventions.

Biofeedback is a general means of impinging on brain function to effect improved regulatory control. Among the primary regulatory functions that can be normalized with biofeedback we can list autonomic arousal and autonomic balance, central arousal, attentional networks, affect regulation and problems of attachment, the anxiety-depression spectrum, and the problem of addiction. There is broad efficacy for recovery from the functional deficits that attend physical brain trauma, and for improved function in the face of degenerative conditions. Biofeedback enhances the stability of brain function wherever that is found to be deficient, as in seizures, migraines, vertigo, rage, hot flashes, etc. Finally, biofeedback offers comprehensive pathways of recovery from deep emotional trauma, as well as from the characterological disorders that issue from unremediated early childhood trauma. Biofeedback can effect improvement in motor control for a variety of movement disorders. It can be invoked to improve cognitive function and working memory, as well as to remediate sensory hyperexcitability. Last, but perhaps not least, biofeedback can help to normalize other regulatory functions such as appetite, blood glucose level, as well as endocrine system and immune system activation.

The World Health Organization, in its study of the Global Burden of Disease, projects depression to rise to the top of the list in terms of cumulative health impacts over the next several years, above cardiovascular disease, cancer, and early childhood afflictions. Efficacy of biofeedback strategies for depression means that self-regulation in general may be the single most important agent of world-wide health recovery that is currently available to policy-makers. A second major category of adverse health impact is obesity and diabetes. Self-regulation strategies could be the most important additional resource currently available to policy-makers in addressing both of these issues. A third category is addictions, where no remedy has shown as much promise as neurofeedback. A fourth category is minor traumatic brain injury, for which neurofeedback represents a substantial, most promising remedy. A fifth category, largely unrecognized, is that of psychological trauma, for which neurofeedback promises a thorough-going, comprehensive remedy. Finally, the impending burgeoning of populations of the elderly portends major mental health concerns. Self-regulation based technologies may be the most important additive modalities available to policy-makers for these issues at the present time.

Clinical research studies have already demonstrated the utility of self-regulation strategies in significantly reducing criminal recidivism, treatment-resistant alcoholism and substance dependency, academic failure, and the consequences of physical or emotional trauma. Just these specific applications of self-regulation strategies could reduce the “lost costs” paid by the society for the maintenance or incarceration of dysfunctional members of our society. One may project a high cost-benefit ratio of an intervention based on self-regulation technologies. It should be the priority of health authorities to move self-regulation strategies to the center of health care rather than to the margins. As part of a health maintenance and optimum health strategy, biofeedback holds promise of keeping many people at risk functional. There is, dare I say it, finally also the moral cost to ourselves of a failure to act in the face of such certain knowledge.


In addition to formulating and issuing a Consensus Declaration, the deeply flawed Efficacy Document must be ushered into a process whereby it is continually updated, so that it can keep up with the changing research picture. Two steps in particular are recommended that would enhance the utility of the document for our purposes:

1) The Level 1 category of “Not empirically supported” should simply be dropped. It is misleading. Positive evidence of efficacy, after all, qualifies a condition for admission to this category. Yet the title is pejorative. In no instance among the conditions listed under Level 1 has there been a single adverse outcome study published. At a minimum, if the category is to be retained, the title should be changed to read “Insufficiently supported.”

2) The Level 5 category of “Efficacious and Specific” should be dropped. There is nothing in the criteria which allows for the determination of specificity. Moreover, specificity is contrary to the realities of our intervention. Self-regulation strategies are by their very nature non-specific with respect to diagnostic categories. Specificity is therefore a claim to which we do not aspire, and a standard by which we do not want our techniques to be judged.

Finally, the authors should give more than lip service to their own statement to the effect that “Evidence-based practice must take into account both efficacy in controlled research settings and effectiveness in the real world of clinical practice.” The document fails to take into account clear demonstrations of clinical effectiveness over the last twenty-five years. It treats effectiveness research as a subset of efficacy research, when in fact effectiveness can be independently established. For example, it is a distortion of major proportions and inexcusable dimension not to acknowledge the work of Bernard Brucker with regard to spinal cord injury, of Margaret Ayers with respect to traumatic brain injury, and of the Menninger group and others with respect to alcoholism and substance abuse. Clinical effectiveness for these conditions at a minimum is simply no longer in question. To claim less it to foist a grotesquery upon the world. Surely the august group that assembled to prepare this document is burdened with the obligation to render their best judgment that the document reflects the reality as they know it. If it does not, then assumptions must be re-evaluated. If the necessary correctives cannot be encompassed within an efficacy document that is constrained by its own rules, then it must be done independently through a Consensus Declaration. Until that is done, the story is not only incomplete but even erroneous.

Note 1. If there should be any doubt about this, it is resolved by Republican strategist Frank Luntz: “Should the public come to believe that the scientific issues are settled, their views about global warming will change accordingly. Therefore, you need to continue to make the lack of scientific certainty a primary issue.” (Italics mine.) Scientific certainty, as we know, is a largely elusive goal whatever the issue. Hence, perpetual controversy is assured.

Note 2. We have known since Thomas Hobbes that science has a logical difficulty in proving cause and effect relationships. The converse problem, showing that a cause and effect relationship does not exist, is even more difficult. It is simply fraudulent for a scientist to suggest that such relationships can be straight-forwardly ruled out.

Note 3. This work is to appear in the Journal Molecular Psychiatry.

Note 4. The case is also made that the choice of ignoring hypotheses that don’t lie close
To mainstream preoccupations is appropriate for the scientific community, lest it be overwhelmed by vast numbers of false leads. Conversely, the early neglect of hypotheses should not be given too much weight. Hence, we should not attach too much significance to the rejection of Sterman’s and Lubar’s early work. It’s not like scientists gave that work much thought. They were just not interested.

Leave a Reply