To: Lucia Williams, Director of Children’s Services
City of
From: Marvin H. Berman, Ph.D. Executive Director, Quietmind Foundation
600
Linda H. Lasner, M.S.,BCIAC Project Administrator
Date: 11/28/2001
A pilot study of sixty children with Attention Deficit Hyperactivity Disorder (ADHD) from within the public mental health outpatient and residential care systems was designed to assess the efficacy of two protocols of brainwave biofeedback or neurofeedback (NFB) as a means for reducing the complex of symptoms associated with this disorder. Thirty-nine subjects (32 treatment and 7 control) completed in the assessment and training protocol. The treatment consisted of forty, thirty-minute twice weekly sessions of brainwave biofeedback training. Based on studies indicating anomalous brainwave frequency profiles of ADHD as compared to normal children. The objective of this study was either to reduce the overall amplitude of brainwave activity across the full brainwave spectrum (.5-30hz) or to selectively increase one frequency band (beta 13-22hz) while decreasing another (theta 4-7hz) in order to create a frequency profile similar to children without ADHD symptoms. Pre and post measures of nonverbal intelligence, attention, and distractibility were gathered to assess the impact of the training. No adverse events ocurred during this study related to brainwave biofeedback training. Significant improvement in non-verbal intelligence was noted for the treatment group and not the controls. Positive trends were noted based on neuropsychological assessment data indicating improved executive functioning and for reduced hyperactivity for the treatment group and not controls that were supported by parent observation measures of ADHD symptoms. Finally, post hoc review of mental health service utilization by participants in this study showed a reduction in the need for other more expensive treatment modalities including individual and family therapy.
This study began in June of 1999
and was completed in July of 2001, during which time the venue for the study
changed twice as did the demographic population from which the study sample was
drawn. The original sample was chosen based on meetings with senior staff at
the Office of Mental Health/Mental Retardation, Northwest Human Services, Inc.
and Quietmind Neurofeedback Services. The initial location was the after-school
partial program at
At this time, we solicited the
participation of other service providers in the area with more stable
populations and programs. The
Treatment with this second group
was completed in June, 2001. Data analysis was completed in late November by
Wayne Whitehouse, Ph.D., Associate Professor in the Department of Psychology at
Treatment of AD/HD has traditionally involved use of psychostimulants and/or behavioral interventions. Among the psychostimulants, methylphenidate, dextroamphetimine, and pemoline are the most commonly used medications. Between 70-80% of children with AD/HD appear to respond favorably to psychostimulants as compared to over 35% that improve with placebos (Barkley, 1990). The primary areas of improvement include attention span, impulse control and motor activity. However, psychostimulants are not without their drawbacks. "To date research studies have not found any single treatment which provides for any long-lasting improvement in ADHD children, particularly once treatment is terminated (Barkley, 1992, p. 8). Perhaps the most serious shortcoming of psychostimulants in treating ADHD is that the benefits are temporary unless the subject is willing to take the medication indefinitely (Barkley, 1990). In addition, side effects including decreased appetite, insomnia, anxiety, irritability, stomach aches and headaches occur in 20-50% of children treated with psychostimulants (S. Goldstein & M. Goldstein, 1990). In most cases, these side effects are mild and short term (Barkley, 1990). A potentially more serious, but infrequent, side effect involves the possible development or increase in tics produced by psychostimulants (Denckla, Bemporad, & MacKay, 1976). Finally, a recent meta analysis of studies assessing the efficacy of treating ADHD with stimulant medication and suggests that, “the extension of this placebo-controlled effect beyond 4 weeks of treatment has not been demonstrated and that exact knowledge of the extent and definition of the short-term behavioral usefulness of methylphenidate is questioned.” (Schachter, 2001)
EEG biofeedback protocols are showing considerable promise as effective treatment for AD/HD and may even be the treatment of choice in cases where medication is ineffective, only partially effective, has unacceptable side effects, or where compliance with taking medication is low (Whalen, Henker and Hinshaw (1985). In addition, since 60-70% of children with AD/HD continue to have symptoms of the disorder into their adolescent and adult years (Weiss & Hechtman, 1994) and psychostimulants do not result in any lasting reduction of AD/HD symptoms, their use must be continued indefinitely if the symptoms are to be controlled. Many adolescents resist taking psychostimulants whether or not they had responded favorably in the past. For this reason, there is a substantial population of AD/HD adolescents and young adults for whom medication is not an acceptable treatment option. EEG biofeedback training programs provide an alternative for this population.
RESEARCH DESIGN
Subjects were solicited to
participate in a study of two approaches to employing brainwave biofeedback to
treat ADHD. Training sessions were conducted at the
offices of Quietmind Neurofeedback Services, 600 E. Germantown Pike,
SUBJECT SELECTION PROCESS/CRITERIA
Therapists
at each of the study sites were asked to suggest children from their caseload
whom they thought met the study’s inclusion criteria. Client chart reviews were
undertaken by clinical staff and the principal investigator to confirm
appropriateness of referrals. After obtaining appropriate consent for
participation, subjects, parents or legal guardians had the opportunity to meet
with the principal investigator, medical consultant and biofeedback therapists
to review the study protocols, assessment procedures and to ask questions. Two
meetings were scheduled for this purpose during which a number of parents
discussed the project and signed consent forms. Informed consent forms were
mailed or taken to the subject’s home for authorization. Ultimately, it was
necessary in more than half of the cases to go to the subjects’ homes to obtain
written consent.
Subjects
were included if they:
1. were actively
enrolled in the program at
2. were between
the ages of 7-12.
3. had an Axis I Dx:
Attention Deficit Disorder with or without hyperactivity
4. demonstrated conduct control expected to be sufficient to
participate in the study protocol without destroying equipment, attacking
staff, etc.
Subjects
were excluded from the study if they:
1. had any other Axis I Dx except Learning Disorders, e.g., Mental Retardation, Pervasive Developmental Disorder, Major
Affective Disorder, psychosis, etc.
2. were taking
neuroleptic medication for treatment of a psychiatric disorder.
3. had a history
of head trauma with loss of consciousness.
4. had a history
of seizures.
5. had a history of known maternal substance abuse during
pregnancy, e.g., Fetal Alcohol Syndrome.
6. had
significant loss or trauma within the past year which warranted V code Dx.
7. had a current
medical condition requiring treatment with oral pain medication, radiation therapy, or surgery. NSAIDS were
not considered as exclusionary medications.
PRE/POST-TREATMENT ASSESSMENT PROTOCOL
All subjects were given the pre- and post-treatment assessment protocol which included the, TONI (Test of Nonverbal Intelligence), 3 core domain subtests of the NEPSY (Neuropsychological Investigation for Children), the Stroop Color and Word Test, the CPT (Continuous Performance Test, Multihealth Systems Inc.) and the Connor’s Behavioral Rating Scale for Parents and Teachers. The two Connor’s behavioral measures were included to balance the ecological validity of data between laboratory and real-world measures (DuPaul, et al., 1992).
TREATMENT PROTOCOLS
This study sought to compare two approaches to EEG neurofeedback training. The two methods being assessed included: Theta/SMR or Theta/Beta brainwave training (JOMH) and wide-band suppression brainwave training protocols (DOMH). The primary distinction between the two methods is in the way in which the feedback signal is defined, i.e., either two signals representing the peak amplitude of the theta and beta wavelengths or a single average peak-to-peak amplitude which accounts for all activity from 1-30Hz. Treatment approaches vary in the selection of sites on the head to do training. The goal in most brainwave biofeedback training however is to reduce the percentage of slow Theta (4-7hz) activity\ and increase the amount of faster Beta (13-22hz) waves. (Lubar,
Each subject’s EEG was recorded at the standard nineteen sites using a specially designed cap with sensors attached which was connected to a 2-channel EEG recording device (NRS-2D Lexicor Medical Technologies). Impedance readings of less than 5kohms were required to begin recording data. The electrical activity reaching the scalp surface was recorded at each of the sites for a period of two minutes while the subjects sat looking at a specific spot on the wall in front of them. Peak amplitude and theta/beta ratio data were recorded for each of the nineteen sites. The six sites evidencing the highest readings were selected for each child as the training locations for the forty biofeedback sessions.
EEG NEUROFEEDBACK PROTOCOLS
The most commonly used EEG biofeedback training protocol employs a single active electrode, a reference electrode on the left ear, and a ground electrode on the right ear. The active electrode(s) is placed midway between CZ and FZ and midway between CZ and PZ using the 10-20 International System. Skin preparation was conducted according to recommendations by the equipment manufacturer. Skin impedance during training sessions was less than 7K ohms protocols. The Lexicor NRS-2D digitizing EEG system (Lexicor Medical Technology, Boulder, CO) was used to provide EEG biofeedback. The traditional protocol is based on Lubar and Lubar (1984). The Lubar protocols emphasize suppressing activity in the theta range (4-7 Hz) with children and adolescents through the age of fourteen, increasing beta (16-20Hz) or sensorimotor rhythm (SMR). (Arbarbanel & Evans, 1999)
In both brainwave biofeedback protocols subjects were randomly assigned to one of two treatment groups. The subjects’ experience was similar in both conditions. A typical neurofeedback session consisted of the following steps: (a) the subject has a sensor placed onto his/her scalp and both ears., (b) the subject is told what he/she would hear from the computer if they were achieving the desired outcome. (c) a child would spend 5 minutes trying to make the computer’s auditory and visual displays match the desired pattern; this was repeated for a total of three five-minute trials during each session. (d) the sensors were removed from the scalp and ears.
In one treatment condition (DOMH) the feedback to the subject indicated a single signal that represented the average peak-to-peak amplitude at the site where the sensor was placed for the frequency range of 0.5-30hz. The training goal was to reduce amplitudes across the entire spectrum. The other condition (JOMH) used the traditionally employed protocol of rewarding decreasing theta (4-8hz) activity while increasing Beta (13-18hz) wave activity.
The personnel providing this
training were made up volunteers and paid staff with varied backgrounds and
levels of experience with this technology. Staff who were
skilled neurofeedback therapists with extensive training and certification in
this technique were assigned equally to the two training protocols along with
graduate students and volunteers with advanced degrees in behavioral sciences
and peripheral biofeedback who were given forty hours of training in the
specific neurofeedback protocols being used in this study. These technicians
were also equally distributed to deliver training over the course of the study.
To encourage participation and reduce frustration in
the learning process, we elected to set the training thresholds so that
subjects would obtain performance scores of at least 75%. When a subject could
achieve a score of 80% or higher for 3 trials, the amplitude thresholds were
lowered or raised 1hz based on the protocol being
employed.
Other considerations are, of
course, sufficient reliability and parsimony to yield statistically meaningful
results with a limited pool of subjects. Given the budgetary constraints within
which this study was conducted, and thus the necessarily limited sample sizes,
this is best considered a pilot study. Consideration given to the design of
future studies must include the ability to retain access to the subjects and
their caregivers so that effective posttest evaluations are possible. For
example, we had two subjects leave the area with no forwarding address, almost
all the subjects phones were disconnected or numbers
changed during the course of the study. In one circumstance, a child was placed
into a residential facility that refused to allow the investigators to
interview the child and collect post treatment data. DHS workers were
cooperative in some cases but also presented numerous administrative and legal
obstacles that prevented the investigators from being able to contact, or in
several cases follow up with control group subjects. In total 18 subjects or a
third of the overall sample were lost due to these problems.
Subjects
were given token rewards of small items, e.g., small toys, pencils, stickers,
etc. after each session in which they completed all 3 training sessions.
Somewhat larger rewards were distributed upon completing twenty sessions. The
children from the Northwest sample were also made aware that on completion of
their fortieth session their names would be entered into a raffle for a new
iMac computer system. Each treatment group had its own computer raffle so the
chances of winning were 1:16. The
ASSESSMENT INSTRUMENTS
We used a combination of cutting
edge and firmly established instruments in the field of clinical child
neuropsychological assessment. We considered the following factors: validity
and reliability for measuring the clinical symptoms and sensitivity to clinical
change, i.e., changes in symptom patterns that would be recognized by parents
and teachers. We were looking to include objective, both clinically
administered measures and natural setting behavioral observations. This is
consistent with literature that supports a combined approach. Sensitivity to practice effects, time and expertise
required for administration and amount and quality of previous research using
these instruments with similar populations were added considerations in
selecting instruments. The available clinical and research suggests that the
most effective assessment of ADHD symptomatology
includes tasks in different modalities, e.g., visual, auditory processing and
different but interrelated functional domains including vigilance,
concentration, executive functioning (problem solving, mental flexibility,
organization and planning) and impulsivity. In addition, some baseline data
regarding overall cognitive ability is important. Data is needed regarding
subjects’ intelligence in order to distinguish between deficits which are
symptomatic of ADHD and other deficits in overall cognitive functioning.
The Test of Nonverbal Intelligence (TONI-3) is a nonverbal, language-free, forty five-item test of abstract reasoning and problem solving. Abstract reasoning and problem solving are the most prominent and most frequently mentioned components of intelligence within many of the major theories of intelligence including Thorndike, Binet, Terman, Spearman, Wechsler, Cattell, Goddard, Piaget, Cronbach, and Sternberg (Brown,1997). This test is culturally reduced and has minimal motor requirements. It estimates intellectual competence by evaluating an individual’s skill at solving novel abstract/figural problems.
The Neuropsychological Investigation for Children (NEPSY) is a forty five-minute standardized neuropsychological assessment that consists of a series of neuropsychological subtests that can be used in various combinations to assess neuropsychological development in children ages 3-12 years. This test covers five functional domains: Attention and Executive Functions, Language and Communication, Sensorimotor Functions, Visuospatial Functions, Learning and Memory. NEPSY is designed for use in a variety of cultural and ethnic groups. In addition to conducting empirical analyses of potential item bias, all items are reviewed by a panel of recognized bias experts. The normed samples include equal samples of rural and urban children and are census balanced for ethnicity. African American and Hispanic children are oversampled so that specific norms will be derived for the target population specifically for this study. The NEPSY is validated for use with children diagnosed with learning disabilities, attention-deficit/hyperactivity disorder, autistic disorders, and speech and language impairment. The reliability statistics include interrater and interscorer agreement, subtest internal consistency, and test-retest stability. In this study, three subtests from the core domain of attention and executive function were given. These include the Tower, the auditory visual attention and response and the visual attention subtests.
The Connor’s Continuous Performance Test (CPT) provides a timed-response assessment for individuals aged 4-19+ years suspected of having attention problems. It is a vigilance, or attention test, which is presented in a game-like format. The CPT Computer Program displays the test stimuli on a monitor, and the individual responds using the PC keyboard or a mouse. The new version of this test contains norms which are automatically compared to the general population norms as well as a large database for comparison to ADHD norms. Administration normally takes 20 minutes and the results can be accessed immediately. The report includes the total number of stimuli, the number correct, omission errors, commission errors, and various reaction times. It is a commonly used method for assessing distractibility, concentration and attention.
This test is ‘based on findings that it takes longer to call out the color names of colored patches than to read words, and even longer to read color names when the print ink is in a color different than the name of the color word.’ The pattern of slowed naming response when a color word is printed in ink of a different color has been explained in terms of response conflict, failures of response inhibition, and/or selective attention. Performance is found to be poorer in people with conditions that include difficulty concentrating and/or difficulty in warding off distractions. (Lezak, 1999, p.374)
The Test of Nonverbal Intelligence
(TONI-3) Quotient improved significantly for the combined treatment groups but
not for the control group. A dependent t-test applied to the mean pre-post
difference scores was reliable, t(27) = 10.30, p
=.042, 1-tailed. The raw means and (SDs) were:
Pre Score SD Post Score SD
Treatment Group 94.47 (3.61) 102.79 (2.29)
Control Group 95.10 (4.97) 94.50 (3.16)
A trend was found toward improvement in executive
functioning for children in both treatment groups as measured by the
Conner's Parent ratings of
hyperactivity showed improvement that approached statistical significance for
the treatment groups, but not for control subjects. A Dependent t-test of pre-post difference
scores found t(35) = 2.80, p =.10, 1-tailed. The raw means and (SDs)
were:
Pre Score SD Post
Score SD
Treatment Group 75.72 (2.35) 70.60 (2.72)
Control Group 73.83 (3.39) 74.42
(3.92)
Performance on the continuous performance test did not change significantly for subjects in either treatment group or setting. We did however notice that several subjects in both settings were able to sit and complete the 14-minute assessment without receiving the almost constant attention and support from test administrators, which had been previously necessary. In many instances, subjects in the pretest condition attempted to stop the test or asked to quit after a few minutes. These same subjects were able to sit and complete the entire posttest assessment process including the CPT with minimal involvement or coaching by the administrator. Subjects reported awareness of the difference in their ability to stay with the task and attributed this to their biofeedback experience.
One anecdotal note regarding the
students’ behavior at the
No significant differences were identified between the two treatment modalities. Given the limited sample sizes of the study population no conclusions can be drawn as to the relative efficacy of these two approaches. It is suggested that reduction of slow wave (1-7hz) activity may be the mechanism of action involved in the therapeutic use of neurofeedback, but this hypothesis will require confirmation by subsequent investigations.
Follow-up interviews were
conducted with twelve families of subjects from
Pre/Post
Treatment Service Utilization and Family Interview Data
|
|
Increase Utilization |
Decreased Utilization |
No Change |
|
Outpatient Services |
1 |
5 |
|
|
Family Therapy |
1 |
3 |
2 |
|
TSS support |
|
|
6 |
For the
six outpatients, we looked at service usage according to
Parents of an additional three children in the afterschool program reported lasting effects from biofeedback training. These included improved attention and compliance with directions, reduced fighting and complaining, improved school performance, and improved sleep. This information was gathered through telephone interviews two months following completion of the training sessions. It was extremely difficult to obtain additional data from parents in some cases as many had moved without forwarding addresses or had phone numbers that had been disconnected or were never answered.
Of the parents we were able to
interview, one parent said that her daughter's behavior and performance at home
and school could be described as both positive and negative using the six
months criteria. The child's psychiatrist supported her desire to discontinue
aftercare (partial hospitalization) reporting that it was no longer needed.
This continued through the six months post-treatment phase. Two families reported
positive changes in listening behavior. One child showed more willingness to
listen to instructions given by the parents, and remembered to follow these
instructions at a later time. This positively influenced the child’s
relationship with his TSS worker. In school, he also seemed better able to
think about the impact of his actions before behaving aggressively. Another
child was described as improved regarding listening to and respecting his
parents' instructions. In addition, this child's habit of taking food up to his
room for safekeeping was reduced. Mother said that he was "less
compulsive" about storing food.
Another two families reported
improved focus and ability to complete homework. Both of these boys also seemed
better able to resist disagreements with other children in the classroom. One
who previously could only sit for fifteen minutes to do homework is now able to
work for a full hour.
One subject aged 13 took the
Kaufman Test of Educational Achievement KTEA/NU in the winter of 1999-2000 while
in the 7th grade. This test assesses decoding and comprehension. The subject
achieved a score of 5.1 equivalent to a fifth grade
performance level. In the Fall of 2000, having
completed the 40-session protocol, he was tested again using the Gates-MacGinitie Reading Test which yielded a score of 7.4. These
scores are standardized scores and are therefore comparable to each other.
These results were so remarkable to the school counselor that he telephoned the
parents, asking what their son had done over the summer to account for this
dramatic change. The boy's mother credits neurofeedback with the two grade-year
improvement in her son’s reading. There
were no other educational enrichment activities of any kind provided during
that period.
The parents of two children
reported increased difficulties in behavior and performance at home and school.
The first boy's behavior was described by his mother as maladaptive before
treatment, improved during treatment, and increasingly more problematic after
neurofeedback sessions were terminated. At six months post treatment, he became
more agitated and his psychiatrist changed his medication, which helped improve
the situation. During his psychotherapy sessions during this period he told his
psychologist that he didn't want his neurofeedback sessions to end. He said ,"If I could use my brain to control feedback and
feelings, I could learn to control my anger."
The last interview was with a
family whose child had a fistfight with an older child post-treatment. This child
had the poorest compliance record with the training protocol in that he
required an additional 4 weeks in which to complete the required 120 training
trials. The court appointed a child advocate for six months, twenty five-hours
a week. No such assistance was in place prior to neurofeedback. The mother
describes her son as enjoying NFB during treatment. After treatment this child
showed behaviors that were calm and more helpful around the house.
This pilot study’s results suggest
that brainwave biofeedback may be a cost-effective treatment modality for
children in a variety of public healthcare settings who are diagnosed with
attention deficit hyperactivity disorder (ADHD). Given the small sample size
and the amount of variance introduced by the changes in site, staff, and
population, obtaining any significant results is highly encouraging. Further
study based on this pilot program should be undertaken to include a larger and
more stable sample, multi-site trials using independent research teams. The
study should include equal populations of children with ADHD from different
ethnic subgroups and take place in settings that support regular biofeedback
training sessions, e.g., in school, afterschool or
residential programs. Data should be collected from the same teacher at pre and
post conditions, and should include grades, standardized test measures,
disciplinary actions along with specific measures of attention, concentration
and hyperactivity. Participants should receive a Quantitative EEG (QEEG) at the
pre- and post-training conditions in order to compare their brainwave activity
against population norms for children with ADHD. Specific neurofeedback
training protocols should be determined to the QEEG assessments and sessions should
be extended to forty-five minutes with thirty minutes.
References:
Barkley, R. A. (1990). Attention
deficit hyperactivity disorder: A handbook for diagnosis and treatment.
Barkley, R. A. (1992). Is EEG biofeedback treatment effective for ADHD children? CHADD Box, 5-11.
Brown, L, Sherbenou, RJ, Johnsen, S.K., (1997) Test of Nonverbal Intelligence: A
Language-Free Measure of Cognitive Ability. Third Edition.
Pro-Ed,
Connors, C.K., et. al. Revision and Restandardization of the Connors Teacher and Parent Rating Scales: Factor Structure, Reliability and Criterion Validity. J. of Abnormal Child Psychology, Vol. 26:4, 1998, p. 279-291 and p.257-268.
DuPaul, G.J., Anastopoulos, A.D.,
Golden, C.J. (1978) Stroop Color and Word Test: A Manual for Clinical and
Experimental Use, Stoelting Co.
Goldstein, S.,
& Goldstein, M. (1990). Managing attention disorders in children: A
guide for practitioners.
Korkman,
M, Kirk, U., Kemp S., NEPSY: A Developmental Neuropsychological Assessment, The Psychological Corporation,
Lezak,
M.D.,(1995) Neuropsychological Assessment 3rd
Edition.
Lubar, J.F., Lubar,
J.O., (1999) Neurofeedback Assessment and Treatment for Attention Deficit/Hyperactivity
Disorders. In Arbarbanel, A., Evans, J.R.,
Introduction to Quantitative EEG and Neurofeedback, Academic Press,
Lubar, J. 0. & Lubar, J. F. (1984). Electroencephalographic biofeedback of SMR and beta for treatment of attention deficit disorders in a clinical setting. Biofeedback and Self-Regulation, 9, 1-25.
Rossiter, Thomas R. & La Vaque ,Theodore J., A (1995) Comparison of EEG Biofeedback and Psychostimulants in Treating Attention Deficit/Hyperactivity Disorders. Journal of Neurotherapy, Summer.
Schachter, H.M., Pham B., King, J., Langford, S., Moher, D., How efficacious and safe is short-acting methylphenidate for the treatment of attention-deficit disorder in children and adolescents? A meta-analysis. Canadian Medical Association Journal, 165(11): 1475-88.
Weiss, G.,
& Hechtnian, L. T. (1993). Hyperactive
children grown up: ADHD in children, adolescents, and adults (2nd edition).
Whalen, C., Henker, B., & Hinshaw, S. (1985). Cognitive-behavioral therapies for hyperactive children: Premises, problems, and prospects. Journal of Abnormal Child Psychology, 13(3), 391-410.
Whitehouse,
W.G., & Echenhofer, F.G. (1997). EEG normalization
therapy for mild head trauma. Research grant abstract to the Office of
Alternative Medicine, National Institutes of Health,