Anxiety Disorders
Conditions
Keywords
Child, Adolescent, Pediatric, Youth, Anxiety, Separation anxiety, Social phobia, Social anxiety, Generalized anxiety, Interpretation bias, Cognitive bias, Cognitive bias modification, Interpretation bias modification, Randomized controlled trial, Threat bias, Threat interpretation, Intervention, Randomized, Cognitive training, National Institute of Mental Health, Personalized
Brief summary
Anxiety is the most common mental health problem in children and adolescents. This two-phased study will test the effects of an experimental computerized intervention aimed at reducing threat-based thinking (i.e., interpretation bias) in anxious youth. Participants in both the R61 (N=46) and R33 (N=72) trials will be youth ages 10 to 17 with a primary anxiety disorder (Separation, Social, Generalized). In the R61 trial, youth will be randomly assigned to receive 16 sessions over 4 weeks of either a personalized cognitive bias modification program for interpretation bias (CBM-I) or a computerized control condition (ICC). If CBM-I reduces interpretation bias significantly more than the ICC, the R33 trial will commence. In the R33, youth will be randomly assigned to either CBM-I or an equal amount of time in a cognitive restructuring intervention, which also aims to reduce threat-based thinking in anxiety. Please note that only the R61 phase of the trial has been completed and currently this record summary only reflects the R61 phase.
Detailed description
Anxiety is the most common mental health problem in youth, affecting one in four children and adolescents. Unfortunately, evidence-based treatments (pharmacotherapy, cognitive-behavioral therapy) are costly, not widely available, and ineffective for a substantial proportion of youth. In response, experts have called for novel treatments that directly target mechanisms underlying youth anxiety while simultaneously addressing barriers to care (i.e., cost, accessibility). One such promising mechanism is interpretation bias - the inaccurate interpretation of threat from ambiguity. The investigators have previously demonstrated that interpretation bias occurs in over 90% of anxious youth, is predictive of anxiety severity in clinical samples of youth, and differentiates between anxious and non-anxious youth. These data indicate that interpretation bias may be a ubiquitous phenomenon underlying anxiety expression in children and adolescents and therefore may be an ideal intervention target. Cognitive bias modification for interpretation bias (CBM-I) is a computerized intervention that attempts to reduce anxiety by directly modifying interpretation bias. CBM-I has demonstrated preliminary efficacy for reducing anxiety symptoms in adults. Yet extant CBM-I data in anxious youth are sparse, with little work addressing whether CBM-I significantly reduces interpretation bias, and whether this in turn reduces anxiety symptoms, as well as the dose necessary to reduce both bias and anxiety. This two-phased study tests personalized CBM-I in youth ages 10 to 17 who meet diagnostic criteria for a primary anxiety disorder (Separation, Social, Generalized). In the R61 Phase (N=46), a randomized clinical trial (RCT) examines whether CBM-I personalized to youth anxiety symptoms significantly reduces interpretation bias compared to a computerized interpretation control condition (ICC). The interpretation target will be measured at multiple time points (4, 8, 12, 16 sessions) to identify the optimal dose for reduction in interpretation bias. If the R61 trial results indicate that CBM-I outperforms ICC on interpretation bias reduction, the R33 phase will commence. In the R33 Phase, an RCT (N=72) will validate whether CBM-I significantly reduces interpretation bias, and conducts a mechanism test (i.e., does bias reduction precede and predict anxiety reduction?), by comparing CBM-I to cognitive restructuring, a clinically relevant psychosocial intervention that also targets anxious cognition. Please note that only the R61 phase of the trial has been completed and currently this record summary only reflects the R61 phase. If the R33 phase is funded, it will be registered as a separate NCT.
Interventions
Computerized intervention in which youth see word-sentence pairs personalized to their anxiety symptoms, and indicate whether these are related. Youth receive feedback aimed to reduce interpretation bias.
OTHERInterpretation Control Condition (ICC)
Computerized control condition in which youth see word-sentence pairs personalized to their anxiety symptoms. Youth indicate whether these are related, but are not provided with feedback that aims to train a reduction in interpretation bias.
Sponsors
National Institute of Mental Health (NIMH)
University of California, Los Angeles
University of Denver
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
DOUBLE (Subject, Outcomes Assessor)
Masking description
Both participants and outcome assessors will be blind to condition. Unblinding occurs following all post-intervention assessment (i.e., after 16 CBM-I or ICC sessions).
Intervention model description
Eligible youth will be randomly assigned to receive either 16 sessions of a personalized cognitive bias modification for interpretation bias (CBM-I) intervention or a computerized interpretation control condition (ICC). Regardless of randomization, participants will complete 16 sessions of their assigned condition within 4 weeks (1 in-lab training per week, and 3 at-home trainings per week).
Eligibility
Sex/Gender
ALL
Age
10 Years to 17 Years
Healthy volunteers
No
Inclusion criteria
* youth aged 10 to 17 * diagnosed at study baseline with a primary big three anxiety disorder (Separation, Social, Generalized) * standard score greater than or equal to 85 on the Wide Range Achievement Test - Word Reading Subtest (to ensure ability to read stimuli during interpretation bias assessment and CBM-I/ICC), and estimated IQ standard score of at least 80 on the Wechsler Abbreviated Scale of Intelligence * youth and consenting parent/legal guardian speak sufficient English to complete consent/assent and study procedures * no concurrent psychosocial services during study participation to reduce likelihood that other interventions are responsible for change in primary or secondary outcomes * no psychotropic medications with no plans to start medications during study, or six weeks stable on SSRI or psychostimulant medication and dose with no plans to change medication/dose during study
Exclusion criteria
* severe anxiety indicting that youth requires higher level of care (e.g., intensive treatment such as psychiatric hospitalization), significant diagnostic comorbidity (e.g., presence of psychosis or significant mood disorder), or another primary diagnosis that warrants alternate intervention * significant uncorrected vision impairment (e.g., uncorrected blindness) that precludes participation in CBM-I/ICC * safety concerns due to recent or acute suicidality with plan, intent, and/or attempt that warrants alternate intervention
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in Linguistic Interpretation Bias as Assessed by the Word-sentence Association Paradigm for Youth (WSAP-Y) | 6 weeks; post-intervention time point | The WSAP-Y is a computerized assessment of interpretation bias in which youth indicate whether word-sentence pairings are related. This measure provides information about the degree to which youth evidence interpretation bias, as well as a behavioral (reaction time) index for bias endorsement. For the primary outcome of percent threat interpretations endorsed, the possible range is 0 (zero) to 100%. Higher percentage indicates more threat interpretation bias (i.e., the youth endorsed a higher proportion of threat words as related to amiguous sentences); lower scores indicate lower threat interpretation bias. Greater decrease in the percent of threat interpretations endorsed and lower absolute scores indicate better outcome (i.e., more reduction in threat interpretation bias; less interpretation bias overall). |
| Change in Visual Interpretation Bias as Assessed by the Ambiguous Faces Task | 6 weeks; post-intervention time point | Youth view faces portraying neutral expressions or subtle emotional expressions (i.e., morphed faces ranging in intensity of emotional valence). Youth categorization of faces as neutral or threatening provides their sensitivity and bias for reporting presence of threat. The criterion mean value in outcomes below reflects a score from -1 to 1, with higher negative numbers reflecting greater bias toward angry faces, and higher positive values approaching one reflecting a bias toward happy faces. Theoretically, higher positive values indicate better outcome such that youth have more positive bias toward facial stimuli. H |
| Change in Self-reported Interpretation Bias as Measured by the Children's Automatic Thoughts Scale (CATS) | 6 weeks; post-intervention time point | The Children's Automatic Thoughts Scale is a youth self-report questionnaire which assesses presence and frequency of a variety of anxious thoughts from domains of: physical threat, social threat, personal failure, and anxious hostility. Total scores range from 0 to 160, with higher scores indicating more threat interpretation bias. Higher scores indicate worse outcome, or that youth self-report that they have more threatening thoughts and/or threatening thoughts at higher frequency; lower scores indicate better outcome or that youth self-report fewer threatening thoughts and/or threatening thoughts at lower frequency. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Number of Trainings Completed of 16 Intended Sessions | 6 weeks; post-intervention time point | This outcome is related to the feasibility of CBM-I aim. The number of CBM-I/ICC trainings completed of 16 intended sessions. Theoretically, completion of more training sessions is better and indicates that youth received a higher dose of training. |
| Participant/Parent Acceptability Questionnaire (PAQ) | 6 weeks | The Participant Acceptability Questionnaire is a 10-item youth and parent report questionnaire, accompanied by an exit interview, that assesses burden (travel, boredom), credibility of computerized intervention techniques, and youth comprehension of the intervention. Parents and youth completed the PAQ separately. |
Countries
United States
Participant flow
Pre-assignment details
Please note that youth were enrolled in the study and their caregiver completed questionnaires and diagnostic interviews about the youth (not about themselves as caregivers/adults). As such, youth were the only true participants in the study (i.e., this was not dyad recruitment and caregivers did not participate as study participants but as additional reporters on the youths' symptoms as is typical in clinical anxiety trials with child and adolescent participants).
Participants by arm
| Arm | Count |
|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) Computerized 16-session intervention aimed at reducing interpretation bias. In this study, CBM-I is personalized to youth anxiety symptoms. During CBM-I sessions, youth indicate whether word-sentence pairs are related, and are provided with feedback aimed to reduce bias.
Cognitive Bias Modification for Interpretations (CBM-I): Computerized intervention in which youth see word-sentence pairs personalized to their anxiety symptoms, and indicate whether these are related. Youth receive feedback aimed to reduce interpretation bias. | 25 |
| Interpretation Control Condition (ICC) Computerized 16-session intervention that is not believed to significantly modify bias. In this study, youth see stimuli personalized to their anxiety symptoms. During ICC sessions, youth see word-sentence pairs and are required to indicate whether word and sentence are related, but are not provided with feedback that aims to train a reduction in interpretation bias.
Interpretation Control Condition (ICC): Computerized control condition in which youth see word-sentence pairs personalized to their anxiety symptoms. Youth indicate whether these are related, but are not provided with feedback that aims to train a reduction in interpretation bias. | 25 |
| Total | 50 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Adverse Event | 2 | 0 |
| Overall Study | ineligible participant unknowingly enrolled | 0 | 1 |
Baseline characteristics
| Characteristic | Cognitive Bias Modification for Interpretations (CBM-I) | Total | Interpretation Control Condition (ICC) |
|---|---|---|---|
| Age, Continuous | 12.76 years STANDARD_DEVIATION 1.64 | 12.48 years STANDARD_DEVIATION 1.73 | 12.20 years STANDARD_DEVIATION 1.8 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 2 Participants | 7 Participants | 5 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 23 Participants | 43 Participants | 20 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants | 2 Participants | 2 Participants |
| Race (NIH/OMB) Black or African American | 2 Participants | 3 Participants | 1 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 3 Participants | 3 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 23 Participants | 42 Participants | 19 Participants |
| Region of Enrollment United States | 25 participants | 50 participants | 25 participants |
| Sex/Gender, Customized Youth self-identified gender Female | 15 Participants | 30 Participants | 15 Participants |
| Sex/Gender, Customized Youth self-identified gender Male | 9 Participants | 17 Participants | 8 Participants |
| Sex/Gender, Customized Youth self-identified gender Non-binary | 1 Participants | 3 Participants | 2 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 25 | 0 / 25 |
| other Total, other adverse events | 5 / 25 | 4 / 25 |
| serious Total, serious adverse events | 0 / 25 | 0 / 25 |
Outcome results
Primary
Change in Linguistic Interpretation Bias as Assessed by the Word-sentence Association Paradigm for Youth (WSAP-Y)
The WSAP-Y is a computerized assessment of interpretation bias in which youth indicate whether word-sentence pairings are related. This measure provides information about the degree to which youth evidence interpretation bias, as well as a behavioral (reaction time) index for bias endorsement. For the primary outcome of percent threat interpretations endorsed, the possible range is 0 (zero) to 100%. Higher percentage indicates more threat interpretation bias (i.e., the youth endorsed a higher proportion of threat words as related to amiguous sentences); lower scores indicate lower threat interpretation bias. Greater decrease in the percent of threat interpretations endorsed and lower absolute scores indicate better outcome (i.e., more reduction in threat interpretation bias; less interpretation bias overall).
Time frame: 6 weeks; post-intervention time point
Population: The WSAP-Y analysis was a between-groups analysis of interpretation bias as measured by percent threat interpretations endorsed during the task. The outcome measures data below provides the percent threat interpretations endorsed at the post-treatment (6 weeks) endpoint, and does not reflect the complex linear, quadratic, and piecewise models of change for analyses in Aims 1 and 2.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) | Change in Linguistic Interpretation Bias as Assessed by the Word-sentence Association Paradigm for Youth (WSAP-Y) | 19.77 score on a scale | Standard Deviation 19.48 |
| Interpretation Control Condition (ICC) | Change in Linguistic Interpretation Bias as Assessed by the Word-sentence Association Paradigm for Youth (WSAP-Y) | 47.40 score on a scale | Standard Deviation 13.52 |
Primary
Change in Self-reported Interpretation Bias as Measured by the Children's Automatic Thoughts Scale (CATS)
The Children's Automatic Thoughts Scale is a youth self-report questionnaire which assesses presence and frequency of a variety of anxious thoughts from domains of: physical threat, social threat, personal failure, and anxious hostility. Total scores range from 0 to 160, with higher scores indicating more threat interpretation bias. Higher scores indicate worse outcome, or that youth self-report that they have more threatening thoughts and/or threatening thoughts at higher frequency; lower scores indicate better outcome or that youth self-report fewer threatening thoughts and/or threatening thoughts at lower frequency.
Time frame: 6 weeks; post-intervention time point
Population: The outcome measures data below provides the measure endorsed at the post-treatment (6 weeks) endpoint, and does not reflect the complex linear, quadratic, and piecewise models of change for analyses in Aims 1 and 2.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) | Change in Self-reported Interpretation Bias as Measured by the Children's Automatic Thoughts Scale (CATS) | 6.52 score on a scale | Standard Deviation 10.22 |
| Interpretation Control Condition (ICC) | Change in Self-reported Interpretation Bias as Measured by the Children's Automatic Thoughts Scale (CATS) | 19.00 score on a scale | Standard Deviation 16.47 |
Primary
Change in Visual Interpretation Bias as Assessed by the Ambiguous Faces Task
Youth view faces portraying neutral expressions or subtle emotional expressions (i.e., morphed faces ranging in intensity of emotional valence). Youth categorization of faces as neutral or threatening provides their sensitivity and bias for reporting presence of threat. The criterion mean value in outcomes below reflects a score from -1 to 1, with higher negative numbers reflecting greater bias toward angry faces, and higher positive values approaching one reflecting a bias toward happy faces. Theoretically, higher positive values indicate better outcome such that youth have more positive bias toward facial stimuli. H
Time frame: 6 weeks; post-intervention time point
Population: The faces analysis was a between-groups analysis of interpretation bias as measured using signal detection theory. The analysis was a between-groups analysis of interpretation bias for detecting anger versus happy faces. The outcome measures data below provides the measure endorsed at the post-treatment (6 weeks) endpoint, and does not reflect the complex linear, quadratic, and piecewise models of change for analyses in Aims 1 and 2.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) | Change in Visual Interpretation Bias as Assessed by the Ambiguous Faces Task | 0.86 score on a scale | Standard Deviation 0.87 |
| Interpretation Control Condition (ICC) | Change in Visual Interpretation Bias as Assessed by the Ambiguous Faces Task | 1.32 score on a scale | Standard Deviation 0.54 |
Secondary
Number of Trainings Completed of 16 Intended Sessions
This outcome is related to the feasibility of CBM-I aim. The number of CBM-I/ICC trainings completed of 16 intended sessions. Theoretically, completion of more training sessions is better and indicates that youth received a higher dose of training.
Time frame: 6 weeks; post-intervention time point
Population: The average number of 16 possible CBM-I/ICC trainings completed per group. Note that this number averages (i.e., Mean score) across all participants in the randomized group. Two participants (siblings) dropped from CBM-I following the second training session due to an unrelated medical emergency and hospitalization, and one participant was removed from the ICC group following randomization and before completion of any trainings due to participant not being eligible for study participation.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) | Number of Trainings Completed of 16 Intended Sessions | 16 number of 16 trainings completed | Standard Deviation 0 |
| Interpretation Control Condition (ICC) | Number of Trainings Completed of 16 Intended Sessions | 15.83 number of 16 trainings completed | Standard Deviation 0.82 |
Secondary
Participant/Parent Acceptability Questionnaire (PAQ)
The Participant Acceptability Questionnaire is a 10-item youth and parent report questionnaire, accompanied by an exit interview, that assesses burden (travel, boredom), credibility of computerized intervention techniques, and youth comprehension of the intervention. Parents and youth completed the PAQ separately.
Time frame: 6 weeks
Population: The PAQ is scored on a scale from 0 to 50, with higher scores indicating higher acceptability. Scores of 30-39 indicate moderate acceptability and scores of 40-50 indicate high acceptability.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Cognitive Bias Modification for Interpretations (CBM-I) | Participant/Parent Acceptability Questionnaire (PAQ) | Parent PAQ | 47.25 score on a scale | Standard Deviation 5.17 |
| Cognitive Bias Modification for Interpretations (CBM-I) | Participant/Parent Acceptability Questionnaire (PAQ) | Youth PAQ | 38.61 score on a scale | Standard Deviation 8.03 |
| Interpretation Control Condition (ICC) | Participant/Parent Acceptability Questionnaire (PAQ) | Parent PAQ | 42.91 score on a scale | Standard Deviation 4.03 |
| Interpretation Control Condition (ICC) | Participant/Parent Acceptability Questionnaire (PAQ) | Youth PAQ | 33.87 score on a scale | Standard Deviation 8.41 |