Abstract
Background:
Substance use disorders are associated with lower cognitive functioning, and this impairment is associated with poorer outcomes. The Therapeutic Education System (TES) is an internet based psychosocial intervention for substance use disorders that may provide enhanced treatment for individuals with cognitive deficits. This secondary analysis investigates the association between cognitive functioning and treatment outcomes in a large (N=507) randomized controlled effectiveness trial of TES compared to treatment-as-usual conducted within outpatient programs in the National Drug Abuse Treatment Clinical Trials Network.
Methods:
All participants completed a computer-based cognitive assessment (Microcog™ short version) at baseline. Scores on subtests of attention, reasoning, and spatial perception were tested as moderators of the treatment effect on abstinence and retention at the end of the 12-week treatment phase using mixed effects logistic regression.
Results:
Cognitive functioning was not found to be a moderator of treatment on abstinence or retention. Post-hoc analysis of the main effect of cognitive functioning on retention and abstinence found impaired reasoning and cognitive flexibility were associated with lower retention. There were no other main effects of cognitive functioning on retention or abstinence.
Conclusions:
The benefit of internet delivered treatment over standard care was unchanged across a range of cognitive functioning. Consistent with previous research, mild to moderate impairment in reasoning and cognitive flexibility were associated with lower retention across both treatment arms.
Scientific Significance:
An internet-delivered cognitive behavioral intervention for substance use disorders, TES, is equally effective across a spectrum of cognitive functioning among diverse patients.
Mesh Terms: Substance Use Disorders, Cognitive Disorders, Cognitive Functioning
Introduction
Individuals with substance use disorders have significantly lower scores on cognitive testing when compared to the general population1–5. Further, cognitive impairment is associated with poorer outcomes suggesting that interventions which are compatible with a broad range of neuropsychological function are necessary6.
One promising approach is to leverage technology to deliver treatment in a manner that is appropriate for individuals with cognitive dysfunction. Several computer-assisted, internet delivered interventions for substance use disorders have been developed and tested7–10. These interventions provide quality controlled evidence-based psychosocial interventions which are potentially scalable and may also provide a benefit to individuals with cognitive impairment by allowing for individualized learning, greater flexibility, and ongoing assessment of content mastery.
The “Therapeutic Education System” (TES), approved by the FDA in September 2017 as the first mobile medical application for the treatment of SUD11, is an Internet-delivered psychosocial intervention grounded in the Community Reinforcement Approach (CRA) and often used in conjunction with automated contingency management8. The TES program uses “fluency-based” Computer-Assisted Instruction (CAI), a way of teaching information and skills that promotes retention of new information12. This instruction requires patients to achieve mastery of material to progress. Mastery is evidenced by accuracy and speed of response on exercises, promoting active participation by the patient8. Experiential learning environments are included via interactive videos of actors modeling important skills (e.g., progressive muscle relaxation, drug refusal skills) to further engage the patient in thinking through real-world scenarios.
TES has been shown to be superior to standard care among methadone patients (N=160) with impaired attention/mental control, memory, information processing accuracy or general cognitive functioning13. In those assigned to standard care the cognitively impaired subgroup had lower rates of abstinence compared to those without cognitive impairment, but no such differential response was found in the group receiving TES.
This improved response among those with cognitive impairment was not found in a trial (N=77) in the general substance use population of a second technology based intervention, CBT 4 CBT, using other cognitive testing of intelligence, attention, processing speed and impulsivity. That CBT 4 CBT study did find that greater impulsivity as measured by a virtual test of impulse control correlated with poorer retention and increased drug use only in the intervention arm14.
In light of the above results showing possible added benefit of internet delivered treatment for individuals with low cognitive functioning and substance use disorders, the current study set out to examine cognitive functioning as a moderator of treatment outcome in a large randomized controlled effectiveness trial. The primary outcome of this trial found that TES was superior to treatment as usual on abstinence and retention outcomes and this finding was especially prominent among those who were not abstinent at baseline assessment. The current exploratory secondary analysis of cognitive functioning as a moderator of treatment response adds to previous work in this area in that the trial included a diverse community sample recruited from 10 outpatient treatment locations throughout the United States and was not restricted to individuals with a specific alcohol or drug use disorder. The trial did exclude individuals who were currently prescribed opioid maintenance therapy. We set out to replicate the finding of a significant moderation effect between TES and cognitive impairment, whereby individuals with greater cognitive impairment would have better treatment outcomes in the TES condition compared to standard care only.
Methods
Initial Study Design
Details of the methods of this study and primary outcomes have previously been published7. Briefly, the study was a randomized controlled trial for patients (N=507) seeking treatment for drug or alcohol problems at 10 community-based, outpatient treatment programs across the United States. Patients were eligible if they were age 18 or older, reported illicit substance use in the 30 days before study entry and were within the first 30 days of their current treatment episode. Patients were excluded if they were currently prescribed opioid maintenance therapy (i.e., buprenorphine or methadone). Patients were randomized in a 1:1 ratio to receive 12 weeks of either: (1) treatment-as-usual in their outpatient treatment program; or (2) treatment-as-usual plus the Therapeutic Education System (TES) and automated contingency management. The TES intervention was a substitute for approximately 2 hours per week of clinician-delivered treatment-as-usual (the equivalent of 4 TES modules or topics per week).
Intervention
The Therapeutic Education System (TES) includes 62 interactive, multi-media modules grounded in the Community Reinforcement Approach (CRA), plus contingency management. The CRA consists of education and skills training to increase drug-free activities and to improve interpersonal interactions15. TES modules include training on basic cognitive behavioral relapse prevention skills (e.g. drug refusal, managing thoughts about using, conducting functional analyses), additional skills aimed at improving psychosocial functioning (e.g. communication, mood management, family/social relations, time management), as well as prevention of HIV, hepatitis and other sexually transmitted infections. The modules also included video clips showing actors modeling the skills being taught and short quizzes that assess grasp of the material to ensure content mastery before moving to the next module. The pace and level of repetition of material is adjusted to maximize individual mastery of the skills and information being taught. Patients accessed modules on computers provided at the treatment program or offsite at a location of their choice.
TES includes a flexible system for delivering contingency management using a prize-based incentive system that has been demonstrated to be effective in improving outcomes16. Incentives take the form of opportunities to draw vouchers from a virtual “fish bowl”. Vouchers yield congratulatory messages (e.g. “good job”) about half the time, or prizes of mostly modest value (usually around $1, occasionally around $20, rarely $80-$100). In the present study, draws were awarded for abstinence, measured by negative urine or alcohol breathalyzer screens, and for completion of modules (up to the recommended 4 per week although there was no cap on the number of modules that could be completed).
Drug/Alcohol Abstinence Assessment
Participants self-reported their primary substance of use (i.e., the substance for which they were seeking treatment or considered the most problematic). Abstinence from drugs was assessed via urine collected and screened for 10 drugs with standard lateral flow chromatographic immunoassays (QuickTox® dip card). Abstinence from alcohol was assessed via breathalyzer and self-report using the Timeline Follow-back method17. The outcome measure abstinence was defined as abstinence from all drugs and alcohol over the final four weeks of the study, allowing for up to two (of eight) half-weeks of missing data per subject. Subjects with more than two half-weeks of missing data were considered non-abstinent.
Neuropsychological Assessment
The MicroCog™ computerized assessment of cognitive functioning is an age and education normalized and standardized cognitive assessment for a nationally representative sample of adults18. Based on our previous work with neurocognitive testing and treatment outcomes, a shortened version of the MicroCog™ (20–25 minutes in length; 8 sub-tests) was used. The subtests measured three cognitive domains: attention/mental control, reasoning, and spatial processing, and were chosen because they had been shown in previous work to be more sensitive in predicting poor treatment outcomes19.
Attention and mental control refers to the ability to attend to new information and to store and mentally manipulate this information in the short-term. This area of function was measured using the Numbers Forward and Numbers Reversed subtests. These tests asked participants to recall digit spans, either forward or reversed. The Wordlist 1 and Wordlist 2 subtests were used to test incidental verbal learning, or memory for words presented without initially being prompted to recall them.
The reasoning domain was measured via the Analogies subtest, a test of the individuals’ ability to connect new ideas with information they have already learned. Also included in this domain were the Object Match A and Object Match B, measures of cognitive and conceptual flexibility and abstraction; that is, an individual’s ability to shift their thinking to patterns or ideas.
Finally, spatial processing was evaluated using the Clocks subtest, a test showing analog clocks hands, asking subjects to transform the time to digital representation. This measures spatial processing and executive functioning abilities20.
Subtest raw scores based on total correct response were transformed to population based scaled scores (mean = 10.0, sd = 3.0). Clinically significant impairment was defined as being more than one standard deviation below the mean. Proficiency scores were also calculated for six of eight subtests using a combination of measured accuracy and speed, with greater weight given to accuracy.
Statistical Analysis
Mixed effects logistic regression models were used to examine the moderating effect of the eight cognitive assessment subtests on treatment effect (treatment-as-usual vs. TES) by examining the interaction between the subtest scores and treatment arm on two outcomes: abstinence (yes or no in the last 4 weeks of treatment) and treatment retention at the end of the 12-week treatment phase (yes or no). Cognitive subtest scores were examined as continuous variables and as dichotomized variables whereby individuals were categorized as impaired (cognitive scores more than one standard deviation below the population based mean) or not impaired. The model controlled for baseline abstinence (yes or no based on urine drug screen and alcohol breath screen) and site was treated as a random effect. Separate analyses were performed for each cognitive subscale measure for both total and proficiency scores and separately for both abstinence and treatment retention.
When the interaction term was not significant, a logistic regression analysis was conducted post-hoc to explore the main effect of cognitive functioning. All analyses were conducted using SAS 9.4.
Results
The baseline characteristics of the two groups, TAU and TAU+TES, are displayed in Table 1. Further demographic details are provided elsewhere21. In brief, the sample was 38% females and a majority identified as White (56%) or Black/African American (23%)and approximately 11% identified as Hispanic. Subjects included relatively equal numbers of individuals seeking treatment for alcohol, cocaine, cannabis, and heroin, with slightly fewer individuals seeking treatment for stimulants. Sixty percent of the sample had graduated high school, fifteen percent completed some post high school education and forty percent were employed.
Table 1.
Baseline Demographic and Clinical Characteristics
| Treatment as Usual (TAU) N=252 |
Therapeutic Education System (TES) N=255 |
|
|---|---|---|
| n (%) or Mean (SD) | ||
| Age (years) | 34.2 (11.1) | 35.6 (10.7) |
| Female | 101 (40.1) | 91 (35.7) |
| Race | ||
| White | 148 (58.7) | 136 (53.3) |
| Black/African American | 47 (18.7) | 69 (27.1) |
| American Indian or Alaska Native | 1 (0.4) | 2 (0.8) |
| Asian | 7 (2.8) | 6 (2.4) |
| Native Hawaiian/Pacific Islander | 5 (2.0) | 7 (2.7) |
| Multi-racial | 31 (12.3) | 23 (9.0) |
| Other | 13 (5.2) | 13 (3.9) |
| Hispanic/Latino |
29 (11.5) | 26 (10.2) |
| Education | ||
| < High School | 58 (23.0) | 60 (23.5) |
| High School/GED | 149 (59.1) | 161 (63.1) |
| > High School | 45 (17.9) | 34 (13.3) |
| Self-reported Primary Substance | ||
| Alcohol | 46 (18.3) | 58 (22.7) |
| Cocaine | 49 (19.4) | 53 (20.8) |
| Other Stimulants | 36 (14.3) | 33 (12.9) |
| Opiates | 60 (23.8) | 54 (21.2) |
| Marijuana | 59 (23.4) | 49 (19.2) |
| Other | 2 (0.8) | 8 (3.1) |
| Cognitive Subtest Total Scores | ||
| (M (SD), % Impaired) | ||
| Attention and Mental Control | ||
| Numbers Forward | 8.5 (2.4), 18.3% | 8.7 (2.7), 20.0% |
| Numbers Reversed | 8.1 (2.6), 25.4% | 8.2 (2.7), 27.5% |
| Wordlist 1 | 7.9 (4.5), 34.3% | 7.1 (4.5), 41.2% |
| Wordlist 2 | 9.7 (3.1), 12.7% | 9.6 (3.3), 13.7% |
| Reasoning | ||
| Analogies | 6.7 (2.8), 49.2% | 6.7 (3.0), 49.8% |
| Object Match A | 8.2 (4.0), 20.8% | 8.4 (3.8), 17.4% |
| Object Match B | 8.8 (2.9), 25.7% | 8.4 (3.3), 33.2% |
| Spatial processing | ||
| Clocks | 11.0 (1.9), 3.2% | 11.0 (1.8), 2.7% |
| Cognitive Subtest Proficiency Scores | ||
| (M (SD), % impaired) | ||
| Attention and Mental Control | ||
| Numbers Forward | 8.4 (2.3), 20.6% | 8.5 (2.4), 22.4% |
| Numbers Reversed | 7.6 (2.2), 29.4% | 7.5 (2.1), 28.6% |
| Reasoning | ||
| Analogies | 7.9 (2.2), 25.4% | 7.7 (2.4), 34.9% |
| Object Match A | 8.5 (2.9), 24.8% | 8.5 (3.0), 23.7% |
| Object Match B | 9.4 (3.2), 17.3% | 8.9 (3.5), 27.3% |
| Spatial processing | ||
| Clocks | 9.7 (1.3), 2.0% | 9.7 (1.3), 1.2% |
Cognitive Impairment as a Moderator of Treatment
In the mixed effect logistic regression models controlling for baseline abstinence, no interaction terms (treatment arm by cognitive assessment subtest) were significant on either retention or abstinence (See Table 2). That is, none of the cognitive subtests (total score or proficiency scores) moderated the treatment (treatment-as-usual vs TES) effect on the outcome of abstinence in last four weeks of treatment or retention in treatment at the end of the 12-week treatment phase.
Table 2:
Interaction effects between Baseline cognitive scores (continuous) and treatment on Retention and Abstinence (N=507)
| Cognitive Subtests | Retention | Abstinence |
|---|---|---|
| F, p-value | ||
| Total | ||
| Attention and Mental Control | ||
| Numbers Forward | F(1, 493)=0.68, P=.4092 | F(1, 493)=1.51, P=.2202 |
| Numbers Reversed | F(1, 493)=0.54, P=.4626 | F(1, 493)=0.31, P=.5767 |
| Wordlist 1 | F(1, 492)=0.49, P=.4835 | F(1, 492)=0.80, P=.3714 |
| Wordlist 2 | F(1, 492)=0.02, P=.8787 | F(1, 492)=0.10, P=.7525 |
| Reasoning | ||
| Analogies | F(1, 493)=0.06, P=.8019 | F(1, 493)=0.68, P=.4091 |
| Object Match A | F(1, 489)=0.02, P=.8838 | F(1, 489)=0.89, P=.3451 |
| Object Match B | F(1, 488)=1.13, P=.2878 | F(1, 488)=0.16, P=.6881 |
| Spatial processing | ||
| Clocks | F(1, 493)=0.59, P=.4442 | F(1, 493)=3.61, P=.0579 |
| Proficiency | ||
| Attention and Mental Control | ||
| Numbers Forward | F(1, 493)=0.00, P=.9870 | F(1, 493)=0.01, P=.9093 |
| Numbers Reversed | F(1, 493)=1.91, P=.1674 | F(1, 493)=0.03, P=.8691 |
| Reasoning | ||
| Analogies | F(1, 493)=0.22, P=.6407 | F(1, 493)=0.56, P=.4563 |
| Object Match A | F(1, 489)=1.44, P=.2301 | F(1, 498)=0.45, P=.5008 |
| Object Match B | F(1, 488)=0.07, P=.7941 | F(1, 488)=1.19, P=.2756 |
| Spatial processing | ||
| Clocks | F(1, 493)=0.32, P=.5700 | F(1, 493)=0.02, P=.8961 |
We also explored the moderation effect of dichotomous cognitive subtest scores (impaired vs not impaired). Cognitive impairment was not a significant moderator of the treatment effect on either outcome.
Main effect of Cognitive Impairment
Analysis of the main effects of continuous cognitive functioning subtest scores on abstinence and retention, controlling for baseline abstinence, showed that subtests of attention, mental control, and spatial processing were not significantly associated with abstinence nor retention at end of treatment (see Table 3).
Table 3:
Models of each Cognitive Subtest as a Main Effect adjusted by baseline abstinence on Retention and abstinence (N=507)
| Cognitive Subtests | Retention | Abstinence |
|---|---|---|
| Odds Ratio (95%CI), p-value | ||
| Total Score | ||
| Attention and Mental Control | ||
| Numbers Forward | 1.02 (0.95–1.10), P=.5257 | 0.98 (0.90–1.06), P=.5942 |
| Numbers Reversed | 0.99 (0.92–1.06), P=.7524 | 0.96 (0.88–1.04), P=.3055 |
| Wordlist 1 | 1.01 (0.97–1.05), P=.7299 | 0.99 (0.94–1.04), P=.6347 |
| Wordlist 2 | 0.98 (0.93–1.04), P=.4996 | 0.95 (0.89–1.02), P=.1273 |
| Reasoning | ||
| Analogies | 1.09 (1.02–1.17), P=.0081 | 1.03 (0.95–1.11), P=.4820 |
| Object Match A | 1.06 (1.01–1.11), P=.0231 | 1.05 (0.99–1.11), P=.0882 |
| Object Match B | 1.07 (1.01–1.14), P=.0197 | 0.99 (0.92–1.06), P=.7249 |
| Spatial processing | ||
| Clocks | 1.04 (0.94–1.15), P=0.4694 | 1.02 (0.90–1.14), P=.8070 |
| Proficiency Score | ||
| Attention and Mental Control | ||
| Numbers Forward | 1.04 (0.96–1.12), P=.3353 | 1.01 (0.93–1.11), P=.7789 |
| Numbers Reversed | 0.97 (0.89–1.06), P=.4943 | 1.02 (0.92–1.13), P=.6697 |
| Reasoning | ||
| Analogies | 1.11 (1.03–1.21), P=.0108 | 1.05 (0.95–1.16), P=.3176 |
| Object Match A | 1.05 (0.98–1.11), P=.1706 | 1.06 (0.99–1.15), P=.1145 |
| Object Match B | 1.04 (0.99–1.10), P=.1405 | 1.01 (0.95–1.08), P=.7180 |
| Spatial processing | ||
| Clocks | 1.04 (0.91–1.20), P=.5548 | 1.14 (0.96–1.35), P=.1467 |
However, several subtest scores in the domain of reasoning were significantly associated with retention (Table 3). When adjusted by baseline abstinence, analogies total score (P=.008; OR=1.09, 95%CI:1.02–1.17) and proficiency score (P=.011; OR=1.11, 95%CI:1.03–1.21) was significantly associated with retention, as was object match A (P=.023; OR=1.06, 95%CI:1.01–1.11) and object match B (P=.02; OR=1.07, 95%CI:1.01–1.14). For all significant main effects of the subtest scores, higher cognitive functioning (higher scores) were associated with greater treatment retention. Odds ratios refer to the increased odds of retention for each additional point on the cognitive subtest. None of the reasoning subtests were associated with the abstinence outcome.
Discussion
In this exploratory secondary analysis of a large randomized controlled trial of an Internet-delivered psychosocial treatment, cognitive functioning did not moderate the treatment effect on the outcomes of abstinence at the end of treatment or on retention. Thus, the hypothesis that TES might be especially effective among patients with lower cognitive functioning was not confirmed. Instead, the findings suggest no evidence that TES differs in its effectiveness across levels of cognitive function, and is therefore an appropriate treatment for patients with cognitive impairment.
The finding that cognitive functioning did not moderate the effect of an internet delivered treatment differs from a prior trial of TES which, unlike this trial, was performed in a methadone maintenance treatment setting. That trial found a significant interaction between Microcog scores and treatment (TES vs TAU) on percentage of total weeks with continuous abstinence and percentage of tested weeks with continuous abstinence. In that trial Microcog scores predicted abstinence in treatment as usual group but did not show similar predictions in the TES arm13. Differential findings may be due to variation in the study samples, as the previous trial only included individuals in methadone maintenance programs while the current study only included patients in general outpatient addiction treatment programs. Individuals in methadone maintenance programs differ significantly from the general population included in the current study in that all have opioid use disorder, generally this disorder is severe, and are all maintained on medication assisted treatment
Our findings are consistent with previous research by Carroll and colleagues who found that cognitive impairment did not moderate the effect of a computer-assisted cognitive behavioral therapy for substance use disorders on abstinence and retention10. Both studies were similar in including individuals seeking outpatient treatment not on medication assisted opioid maintenance treatment.
The implication of these findings are important in that it appears that individuals with clinically significant cognitive impairment derive similar benefit from internet-based interventions compared to those with normal cognition, even among a broad sample of individuals presenting to community-based addiction treatment.
The finding that cognitive impairment does not preclude use of technology-based interventions has been supported in other clinical areas as well. Internet delivered treatments have been found to provide equal benefit regardless of education history in the areas of depression and transdiagnostic symptom severity after hospital discharge22,23.
More research is needed to determine if there are ways to leverage technology-based interventions to be more effective for those with lower levels of education or greater cognitive impairment. For example, future work might explore ways to more effectively use technology-based interventions for individuals with cognitive impairment. These interventions are by design completed independently with minimal oversight by clinical staff, something that might be problematic for individuals with cognitive impairment. It may be difficult for these individuals to focus and engage in treatment without the presence of a provider. Verbal and reading deficits may also create difficulty in understanding and absorbing the material. Future interventions might track eye movements or add continuous verbal response requirements throughout the intervention to monitor and maintain patient focus. In addition, future research is needed to identify thresholds by which technology-based interventions may not be indicated based on cognitive impairment.
This study partly replicated prior findings showing that cognitive functioning is associated with treatment retention as a main effect. Our group was among the first to identify an adverse impact of cognitive functioning on outcomes in individuals treated with CBT based relapse prevention therapy. Impairment in the domains of attention, memory, spatial ability, speed, accuracy, global cognitive functioning and global proficiency were associated with lower treatment retention19. In contrast to this previous study, the present study showed no association between cognitive impairments and decreased retention, except for the cognitive domain of reasoning. This variation was consistent with the findings of a recent systematic review on this topic that identified 46 studies addressing the issue of cognition and treatment response6. The review noted a high degree of variability between studies in testing measures, population, and outcomes. Despite this variability in outcomes, cognitive impairments of various kinds are prevalent in populations with substance use disorders and a significant number of studies have found aspects of cognitive functioning are associated with treatment outcome. Further research focused on optimizing interventions for patients with substance use disorder and cognitive impairments therefore seem warranted.
Strength of the current study includes the large sample size, real world clinical settings, and inclusion of patients with a variety of primary substance use disorders. An important strength for this study was that screening occurred in community settings and without stringent eligibility criteria, suggesting a representative sample of the treatment community.
Despite these strengths, several limitations should be noted. First, the study used a shorter version of the Microcog assessment, and global scores of cognition could not be calculated. Second, this was an exploratory study, therefore corrections for multiple statistical tests were not utilized, which can increase the possibility of Type I error. This limitation is attenuated given lack of significant findings of moderation across both total and proficiency scores. The significant association between impairment in some neurocognitive domains and decreased retention should be interpreted with caution because of the multiple tests performed. Finally, the parent study was not specifically powered to detect an interaction between cognitive functioning and abstinence or retention outcomes. The absence of positive interaction might therefore have been due to lack of sufficient power.
In sum, it appears that an internet-delivered cognitive behavioral intervention for substance use disorders, the Therapeutic Education Systems, is equally effective across a spectrum of cognitive functioning among diverse patients seeking outpatient substance use treatment. Further study is needed to determine mechanisms by which cognitive functioning influences retention in treatment and to optimize technology-based interventions for those with cognitive impairment.
Acknowledgement and Sources of Funding:
The National Institute on Drug Abuse (NIDA) contributed to the development of study design and initial protocol. The Emmes Corporation, a subcontract vendor of NIDA, contributed to data management and quality assurance. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper. Analysis, interpretation, manuscript preparation, and decision to submit the manuscript for publication were the sole responsibility of the authors. The study in this manuscript was funded by the NIDA grant U10 DA013035 (PI: Nunes). Dr. Aharonovich was funded by the NIDA grant R01 DA024606 (PI: Aharonovich). Dr. Shulman was funded by the NIDA grant T32 DA007294. Dr. Nunes was funded by the NIDA grant K24 DA022412 (PI: Nunes).
References
- 1.Tapert SF, Baratta MV, Abrantes AM, Brown SA. Attention dysfunction predicts substance involvement in community youths. J Am Acad Child Adolesc Psychiatry 2002;41(6):680–686. DOI: 10.1503/jpn.130052 [DOI] [PubMed] [Google Scholar]
- 2.Luijten M, Machielsen MWJ, Veltman DJ, Hester R, de Haan L, Franken IHA. Systematic review of ERP and fMRI studies investigating inhibitory control and error processing in people with substance dependence and behavioural addictions. J Psychiatry Neurosci 2014;39(3):149–169. PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Davis PE, Liddiard H, McMillan TM Neuropsychological deficits and opiate abuse. Drug Alcohol Depend 2002;67(1):105–108. PMID: [DOI] [PubMed] [Google Scholar]
- 4.Bechara A, Damasio H Decision-making and addiction (part 1): Impaired activation of somatic states in substance dependent individuals when pondering decisions with negative future consequences. Neuropsychologia 2002;40(10):1675–1689. PMID: [DOI] [PubMed] [Google Scholar]
- 5.Aharonovich E, Shmulewitz D, Wall MM, Grant BF, Hasin, DS. Self‐reported cognitive scales in a US national survey: reliability, validity, and preliminary evidence for associations with alcohol and drug use. Addiction 2017. PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Domínguez-Salas S, Díaz-Batanero C, MartinLozano-Rojas O, Verdejo-García A Impact of general cognition and executive function deficits on addiction treatment outcomes: systematic review and discussion of neurocognitive pathways. Neurosci Biobehav Rev 2016;71:772–801. PMID: [DOI] [PubMed] [Google Scholar]
- 7.Campbell AN, Nunes EV, Matthews AG, Stitzer M, Miele GM, Polsky D, Turrigiano E, Walters S, McClure EA, Kyle TL, Wahle A, Van Veldhuisen P, Goldman B, Babcock D, Stabile PQ, Winhusen T, Ghitza UE. Internet-delivered treatment for substance abuse: a multisite randomized controlled trial. Am J Psychiatry 2014;171(6):683–690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bickel WK, Marsch LA, Buchhalter AR, Badger GJ. Computerized behavior therapy for opioid-dependent outpatients: a randomized controlled trial. Exp Clin Psychopharmacol 2008. April;16(2):132–43. PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gustafson DH, McTavish FM, Chih MY, Atwood AK, Johnson RA, Boyle MG, Levy MS, Driscoll H, Chisholm SM, Dillenburg L, Isham A, Shah D. A smartphone application to support recovery from alcoholism: a randomized clinical trial. JAMA Psychiatry 2014;71(5):566–572. PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Carroll KM, Ball SA, Martino S, Nich C, Babuscio TA, Nuro KF, Rounsaville BJ. Computer-assisted delivery of cognitive-behavioral therapy for addiction: a randomized trial of CBT4CBT. Am J Psychiatry 2008;165(7):881–888. PMID . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Dwyer JB, Beyer C, Wilkinson ST. FDA. The Brown University Child & Adolescent Psychopharmacology Update. 2017. doi: 10.1002/cpu.30251. [DOI]
- 12.Kuhn M, Stahl SA. Fluency: a review of developmental and remedial practices. J Edu Psychol 2003;95(1):3. doi: 10.1037/0022-0663.95.1.3 [DOI] [Google Scholar]
- 13.Acosta MC, Marsch LA, Xie H, Guarino H, Aponte-Melendez Y. A web-based behavior therapy program influences the association between cognitive functioning and retention and abstinence in clients receiving methadone maintenance treatment. J Dual Diag 2012;8(4):283–293. doi: 10.1080/15504263.2012.723317 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Carroll KM, Kiluk BD, Nich C, Babuscio TA, Brewer JA, Potenza MN, Ball SA, Martino S, Rounsaville BJ, Lejuez CW. Cognitive function and treatment response in a randomized clinical trial of computer-based training in cognitive-behavioral therapy. Subst Use Misuse 2011; 46(1):23–34. doi: 10.3109/10826084.2011.521069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Meyers RJ, Miller WR (2001). A Community Reinforcement Approach to Addiction Treatment. New York, Ny: Cambridge University Press; 2001. ISBN:0521771072 [Google Scholar]
- 16.Petry NM, Peirce JM, Stitzer ML, Blaine J, Roll JM, Cohen A, Obert J, Killeen T, Saladin ME, Cowell M, Kirby KC, Sterling R, Royer-Malvestuto C, Hamilton J, Booth RE, Macdonald M, Liebert M, Rader L, Burns R, DiMaria J, Copersino M, Stabile PQ, Kolodner K, Li R. Effect of prize-based incentives on outcomes in stimulant abusers in outpatient psychosocial treatment programs: a national drug abuse treatment clinical trials network study. Arch Gen Psychiatry 2005;62(10):1148–1156. PMID . [DOI] [PubMed] [Google Scholar]
- 17.Sobell L, Maisto S, Sobell MB, Cooper AM. Reliability of alcohol abusers’ self-reports of drinking behavior. Behav Res Ther 1979;17(2):157–160. PMID: [DOI] [PubMed] [Google Scholar]
- 18.Elwood RW. MicroCog: assessment of cognitive functioning. Neuropsychol Rev. 2001;11(2): 89–100. PMID: [DOI] [PubMed] [Google Scholar]
- 19.Aharonovich E, Hasin DS, Brooks AC, Liu X, Bisaga A, Nunes EV. Cognitive deficits predict low treatment retention in cocaine dependent patients. Drug Alcohol Depend. 2006;81(3):313–322. PMID: [DOI] [PubMed] [Google Scholar]
- 20.Strauss E, Sherman EMS, Spreen O. A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary. 3rd edition New York, NY: Oxford University Press; 2006. [Google Scholar]
- 21.Campbell AN, Nunes EV, McClure EA, Hu MC, Turrigiano E, Goldman B, Stabile PQ. Characteristics of an outpatient treatment sample by primary substance of abuse. J Addict Med. 2013;7(5):363–371. PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ebert DD, Gollwitzer M, Riper H, Cuijpers P, Baumeister H, Berking M For whom does it work? moderators of outcome on the effect of a transdiagnostic internet-based maintenance treatment after inpatient psychotherapy: randomized controlled trial. J Med Internet Res. 2013;15(10):e191 PMID: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pots WTM, Trompetter HR, Schreurs KMG, Bohlmeijer ET. How and for whom does web-based acceptance and commitment therapy work? Mediation and moderation analyses of web-based ACT for depressive symptoms. BMC Psychiatry. 2016;16:158 doi 10.1186/s12888-016-0841-6 [DOI] [PMC free article] [PubMed] [Google Scholar]