The efficacy of neurofeedback in the management of children with attention deficit/hyperactivity disorder

Abstract

Attention deficit/hyperactivity disorder (ADHD) is a behavioural disorder characterized by an inappropriate level of inattention with or without impulsivity or overactivity. The estimated prevalence of ADHD is 7% to 10% in boys and 3% in girls aged four to 11 years. The higher prevalence in boys is believed to be the result of a referral bias because boys with ADHD are more disruptive and aggressive, and, therefore, more likely to be referred to specialty clinics. ADHD is caused by a combination of biological – often genetically determined neurochemical disturbances – and environmental disadvantages that are associated with learning difficulties, behavioural problems and social rejection. The identification and treatment of children with ADHD are essential in preventing or at least minimizing the serious complications associated with this disorder. Stimulant medications are the most effective means of symptomatic control of ADHD symptoms, and the safety and efficacy of these medications is well established in the literature. Despite the known efficacy of stimulant medications, alternatives are often sought by parents of children with ADHD. A number of alternative and controversial treatments for ADHD are available, including dietary management, nutritional supplementation, vision therapy, hypnotherapy, guided imagery, relaxation training and electroencephalogram (EEG) neurofeedback. Published well controlled scientific studies either to support or refute the effectiveness of EEG neurofeedback for children with ADHD are not available. At the present time, EEG neurofeedback needs to be considered as an experimental treatment, the validity of which has not yet been determined.

Attention deficit/hyperactivity disorder (ADHD) is a behavioural disorder that is characterized by an inappropriate level of inattention, with or without impulsivity or overactivity. The estimated prevalence of ADHD is 7% to 10% in boys and 3% in girls aged four to 11 years (1). The higher prevalence in boys is thought to be due to a referral bias because boys with ADHD are more disruptive and aggressive, and, therefore, more likely to be referred to specialty clinics (2). ADHD is caused by a combination of biological, often genetically determined neurochemical disturbances and environmental disadvantages (3). It is associated with learning difficulties, behavioural problems and social rejection (4). These complications are more likely to occur in the second half of elementary school as a result of academic failure, and inappropriate and disruptive behaviour, which is often tolerated during the early elementary years.

Inattentive behaviour frequently occurs in children with depression and anxiety disorder; both conditions must be ruled out before a diagnosis of ADHD can be confirmed (5). Twenty per cent to 37% of children initially diagnosed with depression develop a manic outcome (6,7). Milder expressions of mania are associated with high energy, decreased sleep, over talkativeness, and thrill seeking and reckless activities that can mimic ADHD, especially in adolescence. Symptoms of hyperactivity or inattention may also occur in children with anxiety disorders. Strauss et al (8) found that 34.6% of preadolescent children with anxiety disorders met the criteria for ADHD (8). It can be difficult to determine whether inattentive symptoms are a result of the child’s anxiety or whether the anxiety symptoms are secondary to the problems that the child is having because of his or her inattentive behaviour. Issues regarding the overlap among symptoms of ADHD, depression and anxiety are complex, and little direct research on this problem is available (9). A thorough evaluation of all children with inattentive symptoms is required to ensure that an accurate diagnosis is made.

The diagnosis of ADHD requires that all of the following five diagnostic criteria be met (10):

• the presence of either six symptoms of inattention or six symptoms of hyperactivity-impulsivity, which have persisted for at least six months;

• the presence of some symptoms that caused impairment before the age of seven years;

• the presence of some impairment from symptoms in two or more settings (eg, school and home);

• evidence of clinically significant impairment in social or academic functioning; and

• symptoms that do not occur exclusively during the course of a pervasive developmental disorder (autism), schizophrenia or other psychotic disorder, and that are not better accounted for by another mental disorder (eg, mood or anxiety disorder).

MANAGEMENT OF ADHD

The identification and treatment of children with ADHD is essential in preventing or, at least, minimizing the serious complications associated with this disorder. Stimulant medications are the most effective means of symptomatic control of ADHD symptoms, and the safety and efficacy of these medications is well established in the literature (11). Stimulant medications are especially effective when used as a part of a multimodal treatment plan that includes academic and behavioural interventions and support. As a result of their effectiveness, relative safety and enormous frequency of use, stimulant medications have become the gold standard against which any treatment for ADHD is measured (12). Despite their known efficacy, alternatives to stimulant medication are often sought by parents of children with ADHD. These parents are concerned about the side effects of medication and often prefer, at least initially, to try a more natural or less invasive approach to therapy. A number of alternative and controversial treatments for ADHD are available, including dietary management, nutritional supplementation, vision therapy, hypnotherapy, guided imagery, relaxation training and electroenchephalography (EEG) neurofeedback (13–18).

EVALUATING THE VALIDITY OF ALTERNATIVE TREATMENTS

It is important to evaluate the validity of all treatments in children with disease or disabilities, especially those children who are difficult to treat or have potentially serious complications. Questions to consider when reviewing articles about new and alternative therapies have been summarized by the Evidence-Based Medicine Working Group (19).

If it is necessary to know how well an intervention prevents or improves disease or disability, the best evidence is from double-blind, randomized studies of groups of patients. The studies should include a control group with patients with the same set of symptoms who receive a different form of treatment from the one being tested or no treatment at all. For example, in a study of children receiving EEG neurofeedback, the control group would include similar children with ADHD of the same severity and who attend therapy sessions, but do not get actual neurofeedback training. It may be that some other less expensive component of therapy may result in improvement rather than the neurofeedback itself. Confounding factors, placebo effects, and selection and information biases can all contribute to significant methodological flaws in reported results, and need to be evaluated before an alternative treatment can be considered to be valid (20).

NEUROFEEDBACK

EEG studies have shown that children with ADHD have increased theta (slow wave) activity and decreased beta (fast wave) activity compared with normal controls (17,21–24). Children with ADHD who receive EEG neurofeedback training are able to decrease theta activity, which is associated with daydreaming or distraction, and increase beta activity, which is associated with sustained attention (21,26–35). This change in desirable brain wave activity is thought to be the direct result of EEG neurofeedback training and is the basis for neurofeedback programs. It is unclear, however, how much of the improvement is due to EEG neurofeedback itself and how much is due to other nonspecific factors.

Numerous studies on EEG neurofeedback have been published. A review of all of the scientific articles using the key words ‘attention deficit disorder’, ‘electroencephalography’, ‘biofeedback’ and ‘neurofeedback’ from MEDLINE from 1966 to May 2000 and PsycINFO from 1984 to April 2000 found that none of the studies could be considered valid based on the five guidelines provided by the Evidence-Based Medicine Working Group (36). Only one study (37) included a control group. In this study, 18 children were randomly assigned to either a treatment group, which included 40 45-min sessions of EEG training, or a control group (those on the waiting list) who received no treatment. Study personnel were blind to treatment; however, parents, who also completed behavioural rating scales, were not, and outcomes may have been affected by parental expectations for improvement. In addition, EEG data were not evaluated before and after treatment, and improvements may have been secondary to unidentified cointerventions. In the remaining studies, the assignment of patients to treatment was not random, and any reported results are, therefore, subject to methodological flaws, including confounding factors, placebo effects and bias (38,39).

Confounding factors

Confounding is caused by a mixing of effects. The effect of the exposure of interest may be inaccurate if it is mixed with the effect of an extraneous factor. For example, in children who receive EEG neurofeedback training, the desired outcome is a change in brainwave activity, and the exposure is EEG neurofeedback. The extraneous factors are those factors that are present at the same time as the exposure and that may contribute to the outcome. EEG neurofeedback is rarely performed in isolation. To engage the child in activities required for neurofeedback to be successful, an experienced practitioner must establish and maintain a therapeutic alliance with the child. The therapeutic alliance itself may be the factor that results in a change in brainwave activity. Other extraneous factors include relaxation, self-control training, homework and behaviour modification, with rewards provided for less active behaviour; any of these factors may coexist with neurofeedback training. In children who receive EEG neurofeedback, it is unclear what is causing the improvement in brainwave activity – EEG neurofeedback itself or one of the extraneous factors (40).

Placebo effects

A placebo effect is a physical or emotional change that occurs after a treatment has been given, and is not the result of any special property of that treatment. Parents and children are provided with impressive literature, and exposed to sophisticated equipment that they are told will improve a child’s ability to focus attention and stay on task. These families are then asked to make a significant financial investment for a therapy that they want to ‘believe’ will be effective. These impressions and beliefs make it difficult for the family to be objective in their reports on the effectiveness of the treatment that their child has received. Experts who have studied the subject for years, and have both a scientific and financial interest in the outcome are also less able to provide an objective measure of the results. Improvements in the behaviour of children with ADHD who receive EEG neurofeedback may occur as a result of placebo effects.

Bias

Bias is a distortion of the measured effect as a result of the method by which information was obtained or subjects were selected. Information bias occurs if there is an error in the classification of the subjects enrolled in a study (40). It is unclear from most of the research on EEG neurofeedback how the diagnosis of ADHD was made and whether other causes of inattentive behaviour, including learning disabilities, depression and anxiety, were excluded. For research to be valid, it must include an accurate diagnosis or classification of the children being tested.

Selection bias results from the method of selecting subjects that leads to an outcome different from the estimated outcome in the entire population with ADHD. As a result of the financial investment and time commitment (four to six months) required for a program of EEG neurofeedback to be undertaken, the majority of children who receive therapy are most likely from stable, upper socioeconomic households. Several large prospective studies of children with ADHD have found that socially disadvantaged backgrounds, including poverty, low socioeconomic status and parental psychopathology, are associated with persistent symptoms and a poorer outcome in children with ADHD. Children who receive EEG neurofeedback are a select group of children who may have simply improved over time because of their advantaged position. This illustrates the importance of a matched control group in any study that addresses the efficacy of EEG neurofeedback.

Despite the fact that there are no well controlled scientific studies to support the efficacy of EEG neurofeedback, several reasons why the procedure has been accepted as an alternative to medication by some members of the community are as follows.

Crossover study design:

A crossover design is often used in studies of applied behaviour analyses, using children as their own controls. Children who have been studied using this design, receive EEG neurofeedback and any improvement in their brainwave activity and attentive behaviour is documented. These studies have shown that after neurofeedback is discontinued (crossover), any improvements return to baseline (41). When the neurofeedback is resumed, the improvements can again be documented. Unfortunately, studies using this design have included small numbers of children and have not controlled for confounding factors, placebo effects or bias, all of which significantly limit the reliability of the results.

Reproducibility of results:

There have been numerous uncontrolled studies by authors from various institutions that have documented the positive effects of EEG neurofeedback (34,42–57). However, because these studies are uncontrolled, the results are not valid. In addition, there have been several case reports and numerous testimonials from parents who believe that the treatments are effective. Negative results from these types of studies are less likely to be reported (publication bias) and the impression of the general public, therefore, remains that EEG neurofeedback must be effective.

Medication side effects:

Stimulant medications are associated with a number of side effects, including appetite suppression, sleep disturbance, mood changes and tics. Parents are naturally concerned about the well-being of their child and are willing to try new approaches to treatment that do not involve long term medication.

SUMMARY

Well controlled scientific studies published by any individual, institution or organization are not available either to support or refute the effectiveness of EEG neurofeedback for children with ADHD. Until additional research has been completed, the evidence will remain inconclusive. Randomized, double-blind, placebo controlled trials evaluated by those researchers without conflicting financial agendas are required. If EEG neurofeedback is effective, it is crucial that its efficacy be properly documented and treatment be readily available at low cost to all children at risk. If EEG neurofeedback is not effective, that component of therapy that results in an improvement in brainwave activity needs to be identified so that parents are not paying for an expensive component of therapy that is not effective or required. At the present time, EEG neurofeedback needs to be considered as an experimental treatment, the validity of which has not yet been determined.