Filetype PDF | Posted on 26 Sep 2022 | 4 years ago
Authentication
digital resources
285x Filetype PDF File size 1.08 MB Source: liu.diva-portal.org
539998JADXXX10.1177/1087054714539998Journal of Attention DisordersPettersson et al.
research-article2014
From Research to Practice
Journal of Attention Disorders
2017, Vol. 21(6) 508 –521
Internet-Based Cognitive Behavioral © The Author(s) 2014
Reprints and permissions:
Therapy for Adults With ADHD in sagepub.com/journalsPermissions.nav
DOI: 10.1177/1087054714539998
Outpatient Psychiatric Care: A journals.sagepub.com/home/jad
Randomized Trial
1 1 1
Richard Pettersson , Staffan Söderström , Kerstin Edlund-Söderström ,
1,2
and Kent W. Nilsson
Abstract
Objective: The purpose of the study was to evaluate an Internet-based cognitive behavioral therapy (iCBT) program
targeting difficulties and impairments associated with adult ADHD. Method: Forty-five adults diagnosed with ADHD were
randomized to either self-help (iCBT self-help format [iCBT-S]), self-help with weekly group sessions (iCBT group-therapy
format [iCBT-G]), or a waiting-list control group. Treatment efficacy was measured at pre- and posttreatment and at
6-month follow-up. Results: Intention-to-treat (ITT) analysis showed a significant reduction in ADHD symptoms for the
iCBT-S group in comparison with the waiting-list controls at posttreatment, with a between-group effect size of d = 1.07.
The result was maintained at 6-month follow-up. No significant difference was found at posttreatment or 6-month follow-
up between the iCBT-S and iCBT-G groups. Conclusion: The findings show that a CBT treatment program administered
through the Internet can be a promising treatment for adult ADHD. Limitations of the study design and directions for
future research are discussed. (J. of Att. Dis. 2017; 21(6) 508-521)
Keywords
ADHD, adult, cognitive behavioral therapy, randomized controlled trial
Introduction Because there is such a broad range of functional impair-
There is growing evidence that a substantial number of peo- ments in adults with ADHD, the majority of national and
ple diagnosed with ADHD in childhood continue to be international guidelines on ADHD recommend some form
affected into adolescence and adulthood (Barkley, Fischer, of psychosocial intervention as a complement to stimulant
Smallish, & Fletcher, 2002; Davidson, 2008; Fischer & medication, which is still considered to be the first-line
Barkley, 2007). Prevalence studies indicate that about 2% to treatment for adult ADHD (Seixas, Weiss, & Muller, 2012).
7% of the adult population meet the Diagnostic and Even if stimulant medication can ameliorate the core symp-
Statistical Manual of Mental Disorders (4th ed., text rev.; toms, it does not provide the patient with coping skills to
DSM-IV-TR; American Psychiatric Association [APA], manage functional and quality of life impairments. In addi-
2000) criteria for the diagnosis (Barkley et al., 2002; tion, many adults with ADHD prefer not to take medication
Davidson, 2008; Fayyad et al., 2007; Fischer & Barkley, and many continue to experience significant residual symp-
2007; Kessler et al., 2006; Simon, Czobor, Balint, Meszaros, toms (Safren, Sprich, Cooper-Vince, Knouse, & Lerner,
& Bitter, 2009). 2010).
The core symptoms in ADHD, as described in the DSM- Research on psychosocial treatment for adults with
IV-TR, are in the domains of hyperactivity, impulsivity, and ADHD is still in its infancy and the most studied treatment
inattentiveness, and cause difficulties in coping with every- is cognitive behavioral therapy (CBT) oriented approaches
day life. They often lead to psychosocial problems, relation- (Knouse & Safren, 2010; Philipsen, 2012). Two different
ship problems, substance abuse, and problems with work 1County Hospital, Västerås, Sweden
performance and maintaining employment (Barkley, 2002). 2Uppsala University, Västerås, Sweden
Adults with ADHD also show high lifetime comorbidity Corresponding Author:
with other psychiatric diagnoses such as antisocial person- Richard Pettersson, Neuropsykologiska mottagningen, Karlsgatan 17A,
ality disorder, and mood, anxiety, and substance-related dis- 722 14, Västerås, Sweden.
orders (Biederman et al., 2012; Sobanski et al., 2007). Email: richard.pettersson@ltv.se
Pettersson et al. 509
reviews of CBT for adult ADHD show that the results are One form of therapy that provides an opportunity for
generally promising, but there have been few randomized more self-directive treatment is Internet-based CBT (iCBT)
controlled trials (Knouse & Safren, 2010; Mongia & or computer-based CBT (cCBT). This form of treatment has
Hechtman, 2012). been studied intensely in the last decade, and there are now
Three of the most recent randomized controlled trials several reviews and meta-analyses that indicate that iCBT
used attention-matched comparisons; their results supported and cCBT are effective for a range of psychological disor-
the efficacy of CBT in the treatment of adult ADHD. Solanto ders (e.g., depression, panic disorder, social phobia, and
and colleagues evaluated a 12-week manualized meta-cog- generalized anxiety disorder) and other health problems
nitive therapy intervention that targeted time management, (e.g., insomnia, chronic back pain, headache, and tinnitus;
organization, and planning (Solanto et al., 2010). Eighty- Andersson & Cuijpers, 2009; Andrews, Cuijpers, Craske,
eight adults with ADHD, stratified by use of ADHD medica- McEvoy, & Titov, 2010; Cheng & Dizon, 2012; Cuijpers,
tion, were randomly assigned to either meta-cognitive van Straten, & Andersson, 2008). Even though most studies
therapy or supportive therapy in a group modality. The meta- on iCBT have been conducted with minimal therapist con-
cognitive therapy group showed significantly greater reduc- tact, it has been shown that, at least for depression and anxi-
tion in independent evaluator ratings, collateral report, and ety, the amount of support or contact with clients is
self-report of inattention symptoms at posttreatment, than correlated with the treatment effect; effect sizes are larger
the supportive therapy group. Another study that investi- when more support is provided (Andersson & Cuijpers,
gated the efficacy of group-therapy treatment, in this case 2009; Berger, Hammerli, Gubser, Andersson, & Caspar,
Dialectical Behavioral Therapy (DBT)-based skills training, 2011; Spek et al., 2007).
was conducted by Hirvikoski et al. (2011). Fifty-one adults Our aim was to investigate the efficacy of two iCBT pro-
with ADHD were randomly assigned to 14 sessions of DBT gram formats, self-help alone (iCBT self-help format
or a loosely structured discussion group. A per-protocol [iCBT-S]) and self-help with weekly group-therapy ses-
analysis, excluding patients who were unstable on medica- sions (iCBT group-therapy format [iCBT-G]), for adults
tion or did not complete the treatment, showed significantly with ADHD in outpatient psychiatric care. Our hypothesis
greater reduction in self-reported ADHD symptoms at post- was that both treatment groups would show significant
treatment for the DBT group than from the discussion group. reductions in self-rated ADHD symptoms compared with a
However, the difference was not significant when an inten- waiting-list control group. We also wanted to explore
tion-to-treat (ITT) analysis was conducted. Safren, Sprich, whether the iCBT-G group would benefit more from the
Mimiaga, et al. (2010) assessed the efficacy of 12 sessions of program than the iCBT-S group, as earlier studies on iCBT
individual CBT for 86 medication-treated adults with ADHD have indicated that the degree of support is related to out-
and persistent symptoms. Patients were randomized to either come. Finally, we wanted to explore whether the two treat-
CBT or relaxation with educational support. Patients in the ment groups would show reductions in comorbid symptoms
CBT group had lower scores on both assessor- and self-rated compared with a waiting-list control group.
ADHD symptoms at posttreatment compared with patients
in the control condition. The result was maintained at 6 and Method
12 months follow-up.
Most of the evaluated treatment programs have offered a Study Design and Recruitment of Patients
high degree of interpersonal support to patients during
treatment. We found only one study that investigated the The study used a three-group randomized controlled trial.
effect of a more self-directed treatment. In a randomized Patients were randomized to either iCBT in a self-help for-
controlled trial, Stevenson and colleagues examined the mat (iCBT-S), iCBT with weekly group-therapy sessions
effect of a self-directed psychosocial intervention with min- (iCBT-G), or a waiting-list control group. After posttreat-
imal therapist contact (Stevenson, Stevenson, & Whitmont, ment measures, the waiting-list controls were randomized
2003). The treatment consisted of a self-help book with to either iCBT-S or iCBT-G.
education about ADHD and content and exercises covering Adult patients with a diagnosis of ADHD were recruited
strategies to cope with ADHD-related problems. Thirty-five from psychiatric clinics within the county of Västmanland
participants were randomized to either a treatment group or in Sweden or from those referred for ADHD assessment at
a waiting-list control group. At the end of the 8-week inter- the Neuropsychological Clinic (NPC), County Hospital,
vention, the treatment group reported a significant reduc- Västerås, Sweden.
tion in ADHD symptoms, improved organizational skills Because resources were limited, the study was only
and self-esteem and better anger control. This study gives powered to detect a difference in the primary outcome mea-
an indication that more self-directed psychosocial programs sure (self-rated ADHD symptoms) at posttreatment between
with a minimum of interpersonal support can also be a each treatment group and the waiting-list control group. A
promising treatment alternative for adults with ADHD. power analysis was performed with the computer program
510 Journal of Attention Disorders 21(6)
G*Power v. 3.1.9 (Faul, Erdfelder, Buchner, & Lang, 2009). occupational therapist. Fifteen (26%) of the referred patients
The analysis was based on two pairs of one-way ANOVAs had been diagnosed with ADHD outside the NPC. They
of posttreatment scores; iCBT-G versus waiting-list con- were screened with a shorter procedure to confirm diagnosis
trols and iCBT-S versus waiting-list controls. The expected during the introductory assessment at the NPC.
effect size (Cohen’s d) was set to 1.2 based on a mean effect
size for the psychosocial treatments presented in the review Evaluations and Measures
by Knouse & Safren (2010). A power analysis based on an
alpha level of .05 showed that a total sample size of 54 Independent evaluators, blinded to group assignment,
patients would be needed for a power of 80%, with 18 administered the self-report measures and conducted the
patients randomized to each condition. semi-structured interview listed below at pretreatment,
A randomization protocol was created by an independent posttreatment, and 6-month follow-up.
statistician where a series of 54 patients were randomized in
blocks to one of the study conditions over a period of four Primary outcome measure. The Current Symptoms Scale—
semesters (spring 2009 to autumn 2010). The results were Self-Report Form (CSS; Barkley & Murphy, 2006) was
kept in sealed envelopes, each coupled to the number in the used as the primary outcome measure. The scale includes
consecutive series of patients referred to the study and who items based on DSM-IV Criterion A symptoms of adult
met the inclusion criteria. ADHD. Each item is estimated on a scale from 0 to 3 (not at
Unfortunately, the study had to be adjusted before the all or rarely, sometimes, often, very often). A total score—
planned sample of 54 patients had been recruited because of the sum of all scores in the scale across all items—was used
the referral of fewer patients than expected, as well as lim- in the analysis.
ited financial resources and access to personnel. A total of
45 patients had been randomized to the study at the time of Secondary outcome measures. Comorbid symptoms of
adjustment, resulting in uneven numbers of patients in the depression and anxiety were measured by the Beck Depres-
three different conditions: 13 in iCBT-S, 14 in iCBT-G, and sion Inventory (BDI-II; Beck, Steer, Brown, & Lindfors,
18 in the waiting-list control group (Figure 1). 2006) and the Beck Anxiety Inventory (BAI; Beck, Steer, &
Criteria for inclusion were as follows: (a) being at least Järvå, 2012), respectively. Quality of life was measured by
18 years old, (b) having ADHD as the primary diagnosis, the ADHD Impact Module–Adult (AIM-A), a self-report
(c) having access to a computer and the Internet, and questionnaire designed to capture different dimensions of
(d) being able to set aside one afternoon a week for group quality of life of adults with ADHD. It is comprised of four
meetings. Exclusion criteria were as follows: (a) borderline global quality of life items, five economic impact items, and
or antisocial personality disorder, (b) bipolar disorder, five multi-item scales. Two multi-item scales (“Living with
(c) ongoing substance abuse, (d) suicidal ideation, (e) dys- ADHD” and “Work, Home and School Performance and
lexia, (f) mental retardation, and (g) ongoing psychother- Daily Functioning”) were chosen a priori to be included in
apy. Patients who were taking prescribed ADHD medication the study because the primary interest was to investigate
had to be stable on the medication during the whole study whether it was possible to alter the patients’ diagnosis-asso-
time. Approval of the study was obtained from the Regional ciated problems and their ability to cope with daily life
Ethical Review Board in Uppsala, Sweden, and all patients using the techniques of the treatment regimen. The original
gave informed written consent. article describing the development of the AIM-A and an
evaluation of its psychometric properties was based on a
Diagnostic Assessment sample of 317 adult patients with ADHD. It showed that the
scale had good internal consistency, Cronbach’s alpha
Forty-three (74%) of the 58 patients assessed for eligibility between .83 and .91, satisfying convergent and discriminant
were assessed for ADHD at the NPC. The basic ADHD validity and indicated that it was sensitive to change (Land-
assessment consists of self-rating scales concerning current graf, 2007). To capture the patient’s perspective about his or
symptoms, a clinical interview with the patient and a signifi- her occupational performance, the Canadian Occupational
cant other, intelligence testing, cognitive screening and a Performance Measure (COPM) was used. The COPM is a
general psychiatric assessment (Axis I and II). If this basic semi-structured interview where the therapist asks the
investigation fails to reach a diagnostic conclusion accord- patient to identify and rate issues in areas of self-care, pro-
ing to the criteria for ADHD in DSM-IV-TR, further assess- ductivity, and leisure, using a scale from 1 to 10. Two
ment is initiated. This consists of a more in-depth scores, performance and satisfaction with performance, are
investigation that includes hereditary aspects and early obtained. In a review of articles examining the psychomet-
childhood symptoms, specific testing of cognitive functions, ric properties, research outcomes, and practice related to
extended personality assessment, and a semi-structured COPM, the authors found 19 articles examining the psycho-
interview regarding self-care, productivity, and leisure by an metric properties of COPM (Carswell et al., 2004). Three
Pettersson et al. 511
Enrollment
114 referrals
56 patients declined/not contactable
58 assessed for eligibility at NPC
13 ineligible (did not meet inclusion criteria)
Ongoing psychological treatment (n = 1)
Did not meet diagnostic criteria (n = 9)
Bipolar disorder (n = 1)
Dyslexia (n = 1)
Unstable medication (n = 1)
45 eligible and randomized
Allocation
iCBT-G iCBT-S Waiting list
(n =14) (n = 13) (n = 18)
Follow-up
Completed Completed Completed
posttreatment posttreatment posttreatment
measures (n = 13) measures (n = 11) measures (n = 15)
Completed Completed
6-month follow-up 6-month follow-up
measures (n = 11) measures (n = 9)
Analysis
Analyzed n = 14 Analyzed n = 13 Analyzed n = 18
Figure 1. Flow chart of the recruiting process.
Note. NPC = Neuropsychological Clinic; iCBT-G = Internet-based cognitive behavior therapy group-therapy format; iCBT-S = Internet-based cognitive
behavior therapy self-help format.
studies that had examined the reliability of the instrument in different settings and for various populations. Generally,
showed strong test–retest reliability, ranging between .84 the studies supported the COPM as a valid measure of occu-
and .92. Eleven studies had evaluated the validity of COPM pational performance. Sensitivity to change was evaluated
no reviews yet
Please Login to review.
