Cri-Help Drug Addiction Treatment Neurofeedback Study: Effects of an EEG Biofeedback Protocol on a Mixed Substance Abusing Population

William C. Scott, B.S.W., C.C.D.P.,1 David Kaiser, Ph.D.,2
Siegfried Othmer, Ph.D.,3 and Stephen I. Sideroff, Ph.D.4

1Neuropsychiatric Institute, University of California, Los Angeles, California, USA and Restorative Technologies, Santa Monica, California, USA
2Department of Psychology, Rochester Institute of Technology, Rochester, New York, USA
3EEG Spectrum International Inc., Encino, California, USA
4Department of Psychiatry and BioBehavioral Sciences, University of California, Los Angeles, California, USA and Moonview Sanctuary, Santa Monica, California, USA


This study examined whether an EEG biofeedback protocol could
improve outcome measures for a mixed substance abusing inpatient population.
Method. One hundred twenty-one volunteers undergoing an inpatient substance abuse
program were randomly assigned to the EEG biofeedback or control group. EEG
biofeedback included training in Beta and SMR to address attentional variables,
followed by an alpha-theta protocol. Subjects received a total of 40 to 50 biofeedback
sessions. The control group received additional time in treatment equivalent to
experimental procedure time. The Test of Variables of Attention (TOVA), and MMPI,
were administered with both tester and subject blind as to group placement to obtain
unbiased baseline data. Treatment retention and abstinence rates as well as
psychometric and cognitive measures were compared.


Experimental subjects remained in treatment significantly longer than the control group.

(p<0.005). Of the experimental subjects completing the protocol, 77% were abstinent
at 12 months, compared to 44% for the controls. Experimental subjects demonstrated
significant improvement on the TOVA (p<.005) after an average of 13 beta- SMR sessions.

Following alpha-theta training, significant differences were noted on
5 of the 10 MMPI-2 scales at the p<.005 level. Conclusions. This protocol enhanced
treatment retention, variables of attention, and abstinence rates one year following treatment.


Alcohol and drug abuse is an ongoing societal and treatment problem (1, 2).
While major resources have been employed to study and treat addiction,
there has been little significant improvement in the success rate of treatment.
Relapse rates remain high, typically over 70% (3–5). Gossop et al.
(6) reported 60% of heroine addicts relapsed one year following addiction

Peniston and associates have demonstrated significantly higher abstinence
rates with alcoholics when they incorporated EEG biofeedback into the
treatment protocol (7–10). Eighty percent of subjects in these experiments
were abstinent one-year posttreatment.

EEG biofeedback training is an operant conditioning technique used to
reinforce or inhibit specific forms of EEG activity. In the alpha-theta protocol
employed by the Peniston studies, low frequency EEG activity was
reinforced. The alpha theta protocol was first demonstrated to be effective
with posttraumatic stress disorder (11).

The efficacy of alpha-theta EEG biofeedback may lie in its ability to
allow participants to better tolerate stress, anxiety, and anxiety-eliciting
situations, which are particularly evident during the initial phases of
recovery. This protocol was shown to significantly lower 13 of the scales
of the Millon Clinical Multiaxial Inventory (MCMI), including anxiety,
whereas traditional treatment produced decreases in only two of these scales
(7). There have been, however, questions raised in the literature regarding the
sample size, sample independence, and methodology in the Peniston et al.
studies (12). Furthermore, there have been no controlled studies reported that
extend these findings to other substances of abuse.

In addition to the psychological problems that substance abusers face
in remaining abstinent, they also experience comorbid conditions that
affect cognitive and attentional deficits. These deficits may be acquired
from prolonged substance abuse (13–17), but the evidence also points to
deficits that predate the abusing behavior (18).

For example, in onestudy, approximately 35% of treatment-seeking cocaine abusers met the
DSM-IV criteria for childhood attention deficit hyperactivity disorder
(19). Also, adult alcoholics report more residual-type attention deficit
disorder (ADD) than controls (20). Low cognitive ability also has been
shown to predict relapse after treatment at an alcohol treatment facility

EEG biofeedback also has been used successfully to improve
attentional, cognitive, and psychosocial functioning, including reductions
in impulsivity (22–24). These and other studies have employed a protocol
in which beta and SMR frequencies (15–18 Hz and 13–15 Hz, respectively)
were operantly conditioned, while inhibiting theta frequencies,
in remediating attentional and cognitive deficits in children and adults
with ADD (25–28, 35). Given the relationship between cognitive/attentional
impairment and addiction it would strengthen a treatment model to
address these deficits.

In the present study, a beta/SMR EEG biofeedback training regimen was
combined with an alpha-theta protocol in the treatment of a mixed substance
abusing population. One expected objective was the enhanced ability of the
subjects to focus on the treatment program, reduce impulsivity, and, thereby,
increase program retention.

In order to extend the positive EEG biofeedback findings in the alcoholic
population, an addict population was selected that included patients addicted
to the following primary drugs: heroin, crack/cocaine, and methamphetamine,
as well as alcohol.



One hundred twenty-one volunteers from the Cri-Help, Inc. residential
treatment program in the Los Angeles area participated in this study. There
were 49 females and 72 males. They were 19 to 53 years of age, with a mean
age of 32.4. The primary drug of choice reported at admission was 31%
heroin, 28% crack cocaine, 26% methamphetamine, 6% alcohol, and 9%
other controlled substances; 94% were multiple-drug users.

Subjects determined to have a diagnosed psychotic or personality disorder
(based on DSM-IV criteria), or a seizure disorder, were excluded. Subjects were
randomly assigned to the EEG biofeedback plus conventional treatment group
(60 experimental subjects) or the conventional treatment-only group (61
control subjects).

EEG Biofeedback Protocol 457

Subjects were provided informed consent before participating in this
experiment, approved by the UCLA Human Subjects Protection Committee.

All subjects received treatment based on the Minnesota Model 12-step
oriented program described by Stinchfield and Owen (29) supported by
group, family, and individual counseling. In addition, the experimental group
received 40–50 EEG biofeedback sessions. The control group received
additional treatment time equivalent to the biofeedback sessions.

Experimental subjects underwent two sessions of EEG biofeedback
training (45 minutes per session) five days a week for four to five weeks.
EEG biofeedback was performed on a Neurocybernetics 2- Channel EEG
biofeedback system.

In Phase I, experimental subjects underwent 10–20 sessions of Beta-
SMR EEG biofeedback in which operant conditioning was used to augment
either 15–18 Hz (beta) or 12–15 Hz (SMR) EEG activity. At the same time,
training attenuated elevated activity in the 2–7 Hz (theta) and 22–30 Hz
(high beta) ranges. Active bipolar electrode placement was at C3-FPZ for
beta and at C4-PZ for SMR, based on the international 10–20 system of
electrode placement (30).

The starting protocol consisted of beta training 50% of the time and SMR
training 50% of the time. These percentages would be altered based on
changing symptomatology and TOVA results (31) with inattentive or
impulsive profiles resulting in increased beta or SMR training, respectively.

After 10 Beta-SMR EEG biofeedback sessions, participants were
reassessed with the TOVA. If a participant scored within the normal range
(i.e., scores of 85 or above), he or she began alpha-theta training. If the
TOVA remained abnormal after the initial 10 Beta-SMR sessions, 5 or 10
additional Phase 1 treatments were administered. It took a median of 10 Beta-
SMR sessions with a mean of 13 sessions for the TOVA to normalize for the
experimental subjects.

In Phase II, subjects underwent 30 sessions of alpha-theta training. The
frequency range for alpha was 8–11 Hz and for theta it was 5–8 Hz. The
initial sessions were used to train down alpha levels that were above 12 mV
(peak to peak), while augmenting theta, until there was ‘‘crossover.’’ This
was defined as the point at which the alpha amplitude drops below the level
of theta. Subsequent to the first achievement of crossover, both alpha and
theta frequencies were augmented.

Before initial crossover was achieved, excess EEG activity in the range
of 15–30 Hz was inhibited. This was intended to reduce muscle tension and
to quiet the mind. After crossover was achieved, the 2–5 Hz frequency range
also was inhibited. This was intended to discourage the sleep transition
during low-arousal states.

Each alpha-theta session began with the subject sitting in a chair with
eyes closed. The active electrode was placed at Pz with a left-ear reference
(A1). The right earlobe was connected to circuit ground. Two distinct tones
were employed for alpha and theta reinforcement, with the higher pitched
sound used to index the higher-frequency alpha band.

At the start of each session, the technician spent 3–5 minutes reading a
script of guided imagery to the experimental subject that dealt with identified
essential elements of maintaining abstinence. These included ongoing regular
attendance at 12-step meetings; weekly meetings with a sponsor, expanding
the individuals identified comfort zones, and mental exercises dealing with
cue extinction and relapse rejection.

After the guided imagery, it was made clear to the subject that the
objective of the training did not involve explicit rehearsal of the script during
the EEG biofeedback. Subjects reporting previous meditative practices were
asked not to use them during the training, since meditation has been observed
to override alpha-theta reinforcement effects. Following the alpha-theta
training, clients were given the opportunity to process their experience.

When it appeared that sleep might be occurring during training, subjects
were told prior to their next session to move a limb if they heard the
technician say either, ‘‘Right foot, left foot, right hand, or left hand.’’ At
points where the subject’s delta activity (2–5 Hz EEG) started to elevate, as
well as at their highest amplitudes (indications of sleep onset), the limb
commands were given to determine responsiveness. The delta amplitude
value at which the subject transitioned to nonresponsiveness was documented.
Subsequently, during sessions where delta was elevating toward
nonresponsiveness levels, the feedback sounds were inhibited in order to
discourage the sleep transition.


Tests were administered prior to training, after Beta-SMR training (Phase 1)
and after alpha-theta training (Phase 2) for experimental subjects and at
commensurate points in time for the control group (typically 1, 16, and
46 days into the research program). All subjects had acclimated to the
institutional setting for a minimum of 7 days prior to testing. The initial
testing was accomplished with subjects and experimenters blind to
group placement.

The TOVA was administered to assess attentional and cognitive
functions (31–33). The Minnesota Multiphasic Personality Inventory
(MMPI-2) was administered at the start of the study and again at 46 days.

EEG Biofeedback Protocol 459

Patient abstinence was determined by collateral contacts in addition to
self-report. Follow-up interviews for this purpose took place at 3-month
intervals over a 12-month period. Research subjects gave permission to
contact individuals who were intimately involved in their recovery. These
individuals were their 12-step sponsors, family members, and those people
referring the subject into the program. Subjects who used substances beyond
one 4-week window were considered to have fully relapsed. Those whose
relapse duration was within a single 4-week window were categorized as a
brief relapse (Please see appendix for procedures flow chart.).


Days in Treatment

Length of stay in treatment averaged 138 days for experimental subjects and
101 days for controls.
This difference was significant t(119)=3.07,
p<0.005. Median length of stay was 147 days for experimental and 103
days for control subjects. Figure 1 shows retention in the program over the
first 12 weeks of the program. As can be seen, at the end of this period, 46%
of control subjects had dropped out of treatment, compared to only 24% of
those who received EEG biofeedback. A chi-square analysis demonstrated a
significant difference in drop-out rate between experimental and control
groups over the 12-week period [X2 (n = 121) = 6.29, p<.05.] There was no

Figure 1. Effect of the EEG biofeedback protocol on patient retention for control
(n = 61) and experimental (n = 60) subjects.
460 W. C. Scott et al.


significant interaction between drug type used (stimulant vs. sedating drugs)
and days remaining in treatment [F(1,118) = .004, ns].
Abstinence Rate
Figure 2 presents the data for the 103 subjects who had reached their 12-
month poststudy status. This includes 55 experimental and 48 control
subjects. Of these subjects, there were 7 experimental and 17 control subjects
who dropped out of treatment prior to completing the study (the initial
45 days), while there were 4 control subjects and 1 experimental subject who
could not be contacted at the 12-month interval.
Of the remaining experimental subjects who completed the study and
were assessed at 12 months, 36 of 47 (77%) were abstinent. This included
8 subjects who had one brief relapse period of less than 30 days during the
year. Of the control subjects who completed the study, there were 12 of 27
subjects (44%) who were abstinent. This included 1 subject who had one
brief relapse period of less than 30 days. A chi-square analysis demonstrated
a significant difference between one year abstinence rates of the experimental
group versus the control group (X2 2(74) = 7.78 p<0.01). There was no
significant interaction between drug type used (stimulant versus depressant)
and abstinence rate [F(1,113) = .844, p>.05].
MMPI-2 Data
Figure 3 presents pre and posttraining MMPI data, including the 10 clinical
scales and 3 validity scales, for the experimental and control groups. Subjects
with Lie scores greater than 70 on either pre or posttraining tests were
excluded from analysis (n = 3, 2 experimental and 1 control). A univariate
Figure 2. Twelve-month follow-up abstinence data for experimental (n = 55) and
control (n = 48) groups.
EEG Biofeedback Protocol 461


mixed-design analysis of variance (ANOVA) was used to evaluate the effects
of the experimental protocol compared to controls on the 10 clinical scales.
As shown in Figure 3, the experimental group’s changes exhibited
significant improvement compared with the changes in the control subjects
(p < 0.005), on the Hs (Hypochondriasis), F(1, 81)=14.087; D (Depression),
F(1, 81)=48.129; Hy (Conversion Hysteria), F(1, 81)=32.682; Sc (Schizophrenia),
F(1, 81)=15.241; and Si (Social Introversion) scales, F(1, 81)=
24.647, p<.005. The experimental group also improved on the Pt
Figure 3. Change in 10 MMPI clinical scales and 3 validity scales for the
experimental group (n = 50) and the controls (n = 33) (+ p<.05, * p < .005).
462 W. C. Scott et al.


(Psychasthenia) scale, although the difference between groups on this scale
was not significant F(1, 81)=1.727, p > .05. Both groups improved on the
Pd (Psychopathic Deviate) scale, F(1, 81)=29.016; F(1, 81)=12.832, p <
.05, respectively.
Mean TOVA standard scores are presented for both groups in Figure 4 (42
experimental, 28 controls). More participants were tested but only those who
provided scores from all three test periods (baseline, post-SMR, post-alphatheta)
were analyzed. There was no significant difference between groups in
initial baseline TOVA scores [F(1,303) = 1.333, p > .05]. A univariate, mixeddesign
ANOVA was used to compare the two groups on four dependent
measures of the TOVA: inattention (percent omission), impulsivity (percent
commission), response time, and response variability. Low scores were
truncated at four standard deviations below normal.
As can be seen in Figure 4, the experimental group exhibited significant
improvement in impulsivity and variability measures in response to Beta-
SMR training F(1, 68)=18.749; p < .005 whereas no comparable change was
found for the control group F(1, 68)=19.405; p > 0.05. Experimental subjects
also demonstrated significant improvement in inattention; however, the score
Figure 4. TOVA standard scores for experimental and control groups for pretraining,
post-SMR, and post-alpha-theta assessments (+ p<.05, * p<.005).
EEG Biofeedback Protocol 463

only marginally differed from that of the control group F(1, 68)=5.549
(p < .05). TOVA scores were not further enhanced by either the alpha-theta
training nor 30 additional days of treatment.


The results of this study support the efficacy of EEG biofeedback training in an
inpatient drug treatment program. Success was determined by length of time in
treatment, or treatment retention, as well as by abstinence rates one year after
termination of treatment. Results were further supported by positive changes
in attentional variables, and positive changes on the MMPI 2. These findings
extend the previous research findings employing alpha-theta EEG biofeedback
with an alcoholic population, to other substances of abuse.

The present study employed a Beta-SMR protocol prior to the alpha-theta
procedure previously used in addiction studies (7–10). Beta-SMR training
previously had been shown to be effective in remediating attentional and
cognitive deficits. Results of baseline performance testing using the TOVA
did not demonstrate that this population had significantly below average
attentional indices. However, testing following the Beta-SMR protocol
showed that this procedure improved these test measures for the experimental
subjects, particularly impulsivity and variability. This result may partly
account for the improved treatment retention of this group.

It has been shown that time in treatment is one of the best
predictors of remaining abstinent (34). In the present study, the experimental
subjects averaged 136 days in treatment. This compared to 98
days for the control population. In addition, treating therapists reported
that they noticed experimental subjects appearing more cooperative and
more attentive as EEG biofeedback progressed.
This subjective observation
should be a focus in future studies with a more systematic observation of
subjects’ behavior.

There were 8 experimental subjects who used briefly (less than 30 days)
but were abstinent at the 12-month follow-up, and there was 1 subject from
the control group who had this experience. It has been noted in the previous
alpha-theta treatment studies that patients report dysphoria when they used
a substance following the EEG biofeedback protocol (8). Some of the experimental
subjects in this study had similar experiences. This may indicate
that a more fundamental neurophysiological change had taken place as a result
of the treatment.
Peniston and Kulkosky (7) for example, noted that
experimental subjects receiving EEG biofeedback did not show increased
circulating beta-endorphin levels, an index of stress, which was found in the
control group.

It can be noted that once the EEG biofeedback was concluded, at week
five, the subsequent attrition rates became indistinguishable between the two
groups. It may be useful in future studies to extend the length of the
biofeedback training to see if it has further impact on experimental results.

One of the more striking findings of the present study and similar to
the Peniston (8) results, is the positive change noted in the MMPI. The
experimental subjects showed significant improvement in five of the
clinical scales: Hypochondriasis, Depression, Hysteria, Schizophrenia, and
Social Introversion. These changes indicate a lowered level of general
distress or discomfort. More specifically there may be a reduced sense of
alienation and depression, as well as defensiveness.
These are vital factors
in recovery.

The present study did not demonstrate differential effectiveness of the
EEG biofeedback protocol for sedative or stimulant drug abusers. This should
be a focus of future research in which larger numbers of subjects are
employed. Both groups of subjects appeared to benefit from this protocol. If
the lack of dependency on drug type is confirmed, the case can be made that
alpha/theta training addresses core issues in addiction rather than drugspecific
aspects of dependency.

In the present study, one-year abstinence was determined by collateral
contacts in addition to self-report. These individuals were reliable sources
who were intimately connected to the recovery process, including their 12-
step sponsors, family members, and those people referring the subject into the
program. Future research results should be supported by incorporating urine
testing as a further corroboration of abstinence.

Since EEG-based reinforcement was such a prominent constituent of the
experimental program, the question arises as to whether the benefits of
training could also be documented through observable EEG changes. The
present study was not designed to analyze the appropriate artifact free data.
Future research should incorporate methodology to record and analyze the
appropriate quantitative EEG data.

It is important to place the results of this study in the context of the longstanding
difficulty in achieving successful abstinence with the drug-abusing
population. EEG biofeedback appears to promote and support positive
change in the level of neurophysiological and psychosocial functioning in the
addict, as well as enhancing treatment retention.
It therefore constitutes a
promising approach that now requires additional study for further validation
as well as to elucidate operative mechanisms to optimize the procedures, and
to facilitate integration into standard treatment programs.


The present study supports the efficacy of an EEG biofeedback protocol as
adjunctive therapy in an in-patient drug treatment program. This protocol
appears to be beneficial for both sedative as well as stimulant substances of
abuse. Success was determined by length of time in treatment as well as by
abstinence one year after termination of treatment. Supportive data were
provided through attentional/cognitive and psychological assessments. These
findings extend the research employing alpha-theta EEG biofeedback with an
alcoholic population to other drugs of abuse.



We wish to thank Marcus Sola (CRI-Help, Chairman of the Board), Jack
Bernstein (CRI-Help, CEO), and Marlene Nadel (CRI-Help, Clinical
abuse. Success was determined by length of time in treatment as well as by
abstinence one year after termination of treatment. Supportive data were
provided through attentional/cognitive and psychological assessments. These
findings extend the research employing alpha-theta EEG biofeedback with an
alcoholic population to other drugs of abuse.
We wish to thank Marcus Sola (CRI-Help, Chairman of the Board), Jack
Bernstein (CRI-Help, CEO), and Marlene Nadel (CRI-Help, Clinica

10. Saxby E, Peniston EG. Alpha-theta brainwave EEG biofeedback
training: an effective treatment for male and female alcoholics with
depressive symptoms. J Clin Psychol 1995; 51:685–693.
11. Peniston EG, Kulkosky PJ. Alpha-theta brainwave neurofeedback
therapy for Vietnam veterans with combat-related posttraumatic stress
disorder. Med Psychother: Int J 1991; 4:47–60.
12. Graap K, Freides D. Regarding the database for the Peniston alpha-theta
EEG biofeedback protocol. Appl Psychophysiol Biofeedback 1998;
13. Ardila A, Rosselli M, Strumwasser S. Neuropsychological deficits in
chronic cocaine abusers. Int J Neurosci 1991; 5:73–79.
14. Arria AM, Tarter RE, VanThiel DH. Liver-brain relations in alcoholics.
Alcohol Health Res World 1990; 14:112–117.
15. Hambidge DM. Intellectual impairment in male alcoholics. Alcohol
Alcohol 1990; 25:555–559.
16. Marchesi C, de Risio C, Campanini G, Maggini C. TRH test in
alcoholics: relationship of the endocrine results with neuroradiological
and neuropsychological findings. Alcohol Alcohol 1992; 27:531–
17. O’Mahony JF, Doherty B. Intellectual impairment among recently
abstinent alcohol abusers. Br J Clin Psychol 1996; 35:77–83.
18. Pogge DL, Stokes J, Harvey PD. Psychometric vs. attentional correlates
of early onset alcohol and substance abuse. J Abnorm Child Psychol
1992; 20:151–162.
19. Carroll KM, Rounsaville BJ. History and significance and childhood
attention deficit disorder in treatment-seeking cocaine abusers. Compr
Psychiatry 1993; 34:75–82.
20. Nixon SJ, Tivis R, Parsons OA. Behavioral dysfunction and cognitive
efficiency in male and female alcoholics. Alcohol: Clin Exp Res 1995;
21. Miller L. Predicting relapse and recovery in alcoholism and addiction:
neuropsychology, personality, and cognitive style. J Cocaine Abuse
Treat 1991; 8:277–291.
22. Lubar JF, Swartwood MO, Swartwood JN, O’Donnell PH. Evaluation
of the effectiveness of EEG biofeedback training for ADHD in a
clinical setting as measured by changes in T.O.V.A. scores, behavioral
ratings, and WISC–R performance. Biofeedback Self-Regul 1995;
23. Lantz D, Sterman MB. Neuropsychological assessment of subjects with
uncontrolled epilepsy: effects of EEG feedback training. Epilepsia
1988; 29(2):163–171.
24. Othmer S, Othmer SF, Kaiser DA. EEG Biofeedback: training for AD/
HD and related disruptive behavior disorders. In: Incorvaia JA, Mark-
Goldstein BS, Tessmer D, eds. Understanding, Diagnosing, and
Treating AD/HD in Children and Adolescents, an Integrative Approach.
New York: Aronson Press, 1999:235–318.
25. Cartozzo HA, Jacobs D, Gevirtz RN. EEG Biofeedback and the remediation
of ADHD symptomatology: a controlled treatment outcome
study. Presented at the Annual Conference of the Association for
Applied Psychophysiology and Biofeedback, Cincinnati, Ohio, March
26. Linden M, Habib T, Radojevic V. A controlled study of the effects of
EEG biofeedback on cognition and behavior of children with attention
deficit disorders and learning disabilities. Biofeedback Self-Regul
1996; 21:35–50.
27. Rossiter TR, La Vaque TJ. A comparison of EEG biofeedback and
psychostimulants in treating attention deficit/hyperactivity disorder. J
Neurother 1995; 1:48–59.
28. Thompson L, Thompson M. Neurofeedback combined with training in
metacognitive strategies: effectiveness in students with ADD. Appl
Psychophysiol Biofeedback 1998; 23(4):243–263.
29. Stinchfield R, Owen P. Hazelden’s model of treatment and its outcome.
Addict Behav 1998; 23(5):669–683.
30. Jasper HH. The 10–20 system of the international federation. Electroencephalogr
Clin Neurophysiol 1958; 10:371–375.
31. Dupuy TR, Greenberg LM. TOVA Manual. Los Alamitos, CA:
Universal Attention Disorders, 1989.
32. Greenberg LM, Waldman ID. Developmental normative data on the
Test of Variables of Attention (T.O.V.A.). J Child Psychol Psychiatry
1993; 34:1019–1030.
33. Kane RL, Kay GG. Computerized assessment in neuropsychology: a
review of tests and test batteries. Neuropsychol Rev 1992; 3:1–117.
34. Ciraulo DA, Piechniczek-Buczek J, Iscan EN. Outcome predictors in
substance use disorders. Psychiatr Clin North Am 2003; 26(2):381–
35. Egner T, Gruzelier JH. Learned self-regulation of EEG frequency
components affects attention and event-related brain potentials in
humans. Neuroreport 2001; 12:4155–4159.
EEG Biofeedback Protocol 469

Address correspondence to Stephen I. Sideroff, Ph.D., Department of Psychiatry
and Biobehavioral Sciences, University of California, Los Angeles, 1245 16th St.,
Suite 210, Santa Monica, CA 90404, USA; E-mail:
The American Journal of Drug and Alcohol Abuse, 31:455–469, 2005
Copyright D Taylor & Francis Inc.
ISSN: 0095-2990 print / 1097-9891 online
DOI: 10.1081/ADA-200056807
Order reprints of this article at