Depression, Insomnia, Suicide
Conditions
Brief summary
The ultimate goal is to facilitate the delivery of empirically supported treatments (ESTs) for mental disorders and to improve client mental health outcomes. Toward this end, this Phase II SBIR proposal seeks to complete the development and testing of PracticeGround, a comprehensive software system designed to integrate with electronic health records, and that contains multiple methods of training clinicians in ESTs and delivering ESTs to clients, continuous progress monitoring of client outcomes, and clinical support tools to guide clinicians and clients through delivery of the necessary EST. The investigators will conduct an 18-week randomized controlled trial (N=80) comparing PracticeGround (n=40) to care-as-usual (n=40) in depressed outpatient clients. PracticeGround clinicians will have full access to the software. Study clinicians and clients will be assessed once every six weeks (baseline, 6, 12, and 18 weeks). To ensure generalizability of findings, the investigators will not control for natural therapy variations (e.g.,session frequency, medication use, etc.). Primary outcomes include: depression, psychological distress, treatment satisfaction (clients and clinicians), and treatment drop out. Secondary clinician outcomes include: extent of PracticeGround use with clients across clinicians' caseload.
Interventions
BEHAVIORALBehavioral Activation for Depression, Cognitive Behavior Therapy for Insomnia, Collaborative Assessment and Management of Suicidality
Evidence-based interventions for treating depression and concurrent problems
BEHAVIORALCare-as-Usual Psychotherapy
Usual care psychotherapy for depression and concurrent problems
Sponsors
Evidence-Based Practice Institute, Seattle, WA
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
PATIENTS Inclusion Criteria: * Meets DSM-IV criteria for MDD and/or DD * Obtains a score of 11 or greater on the PHQ-9 * Currently receiving psychosocial treatment for depression by a study-participant clinician * Payment-eligible to receive at least 12 weeks of additional psychosocial treatment
Exclusion criteria
* Current diagnosis of bipolar disorder or another psychotic disorder * Active mania * Current diagnosis of alcohol and/or drug dependence * At imminent risk of suicide (as deemed by their clinician). PROVIDERS Inclusion Criteria: * English speaker * 18 years or older * Currently licensed mental health treatment provider (or a therapist working towards licensure under direct supervision of licensed mental health treatment provider) * Have an office with access to internet * Have access to computer with sound card and/or tablet and printer * Capacity to accept new patients seeking treatment for depressive symptoms
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Depression, Anxiety, and Stress Scale (DASS) | baseline, 4 weeks, 8 weeks, 12 weeks | Higher scores on subscales from this 23-item measure reflected higher depression, anxiety, stress, and suicidality. Other researchers added two suicidality items to the original 21-item measure with permission from the original authors. Each subscale score is computed by summing the subscale items, then multiplying by 2. Total possible scores for the depression, anxiety, and stress subscales range from 0-42. We categorized DASS depression, anxiety, and stress scores into severity levels coded as 0, 1, and 2: Normal (0 through 9), Mild/moderate (10 through 20), and Severe/extremely severe (21 through 42) (Psychology Foundation of Australia, 2018). The DASS suicide subscale ranged from 0-12 and not recategorized. |
| Beck Depression Inventory (BDI-II) | baseline, 4 weeks, 8 weeks, 12 weeks | Higher scores on this 21-item reflect greater depressive symptoms in the past two weeks. All items are summed to a total score (possible range 0-63). We categorized BDI scores into severity levels coded as 0, 1, and 2: Minimal (raw scores 0 through 13), Mild/moderate (14 through 28), and Severe (29 through 63). |
| Patient Health Questionnaire-9 (PHQ-9) | baseline, 4 weeks, 8 weeks, 12 weeks | Higher scores on this ten-item measure reflects more severe depression. The total score (possible range 0-27) is created by summing the first nine items. We categorized PHQ severity levels coded as 0, 1, and 2 based on authors' research as minimal (0 through 4), mild/moderate (5 through 14), and moderately severe/severe (15 through 27). The recategorized scores are reported here. |
| Behavioral Activation for Depression Scale Short Form (BADS-SF) | baseline, 4 weeks, 8 weeks, 12 weeks | The Behavioral Activation for depression Scale Short Form has 9 questions each having a scale from 0 (Not at All) to 6 (Completely) with a total range of 0-54. Higher scores on this nine-item scale reflect lower engagement in avoidance and higher engagement in activation behaviors over the course of BA for depression. |
| Mental Health Services Satisfaction Survey (MHSSS) | After first study therapy session, 4 weeks, 8 weeks, 12 weeks | We created a ten-item measure in which higher scores reflected greater satisfaction with services they received. The measure had 10 questions each having a scale of 1(Strongly Disagree) to 5(Strongly Agree) with a total range of 10-50. Higher scores indicated a better outcome. |
| Working Alliance Inventory (WAI) | 4 weeks, 8 weeks, 12 weeks | We modified this measure to include items from the BHS from Group Health. The resulting 12-item scale measures the therapeutic relationship with higher scores indicating stronger alliance. With 12 questions and each having a scale of 1(Never) to 7(Always) the total range was 12-84. |
| Therapy Task Checklist (TTC) - Patient Version | 4 weeks, 8 weeks, 12 weeks | We developed the TTC. The Therapy Task Checklist-Patient Version with 25 questions and each having a scale of 0 (Never/Very Rarely) to 7 (All of the Time) has a total range of 0-100. Higher scores reflected more frequent use of evidence-based therapy tasks, positive outcomes. |
| Modified Practice Attitudes Scale (MPAS) | Baseline, 12 weeks | The Modified Practice Attitudes Scale with eight questions, each having a scale of 0 (Not at all) to 4(To a very great extent), has a total range of 0-32. Higher scores on this eight-item scale reflect more positive attitudes towards evidence-based practices. |
| Perceptions of Computer-Assisted Therapy Questionnaire (PCAT) | Baseline, 12 weeks | The Perceptions of Computer-Assisted Therapy Questionnaire has 34 questions each having a scale of 1(Strongly Disagree) to 7 Strongly Agree). Thus, the total range was 34-238. Higher scores on this 34-item scale indicate stronger agreement with the benefits of computer-assisted therapy, or a better outcome. |
| ASA-Monitoring and Feedback Version (ASA-MF) | Baseline, 4 weeks, 8 weeks, 12 weeks | The ASA-Monitoring and Feedback Version (ASA-MF) measure has 18 questions each having a scale of 1(Strongly Disagree) to 5(Strongly Agree) with a total range of 18-90. Higher scores on this 18-item scale reflect positive attitudes about standardized and routine progress monitoring, or positive outcomes. |
| Monitoring and Feedback Attitudes Scale (MFA) | Baseline, 4 weeks, 8 weeks, 12 weeks | The Monitoring and Feedback Attitudes Scale has 14 questions and each has a scale of 1(Strongly Disagree) to 5(Strongly Agree), thus the total range was 14-70. Higher scores on this 14-item measure indicate more positive attitudes about routine progress monitoring and providing feedback to patients about treatment progress, or positive outcomes. |
| Routine Monitoring (RM) | Baseline, 4 weeks, 8 weeks, 12 weeks | We created a three-item measure where the total score is found by summing the items. The first two items are yes(scored 1)/no(scored 0) questions. The last item is a count measure that can be any number greater than or equal to 0. Thus, the total score can have any integer value greater than or equal to 0, with higher scores reflecting greater use of routine progress monitoring. |
| Current Assessment Practice Evaluation-Revised (CAPER) | Baseline, 4 weeks, 8 weeks, 12 weeks | We modified the original measure into nine items. The Current Assessment Practice Evaluation-Revised has 9 questions with each having a scale of 1(None, 0%) to 5 (Nearly All, 81-100%). Thus, the total range was 9-45 and higher scores reflect greater provider use of measurement-based care or positive outcomes. |
| Therapy Task Checklist (TTC) - Provider Version | 4 weeks, 8 weeks, 12 weeks | The Therapy Task Checklist -Provider Version has 22 questions and each have a scale of 0(Never/Very Rarely) to 4 (All of the time) with a total range of 0-88. Higher scores reflected more frequent use of evidence-based therapy tasks, or positive outcomes. |
Countries
United States
Participant flow
Recruitment details
We recruited providers and their patients from 17 behavioral health clinics, outpatient clinics, and private practices. After consent, providers were randomized to WILLOW or TAU condition. Providers referred their existing new patients to the study; those deemed eligible and who consented were assigned to the same study condition as their provider.
Pre-assignment details
98 providers completed eligibility screening; 91 were eligible and 76 consented to participate. 29 participating providers enrolled 1+ patients. Providers referred 119 patients to the study; 100 completed screening, 73 met eligibility criteria, 66 provided consent, and 56 attended 4+ therapy sessions and completed all data collection points.
Participants by arm
| Arm | Count |
|---|---|
| WILLOW Clinicians have access to WILLOW's dynamic progress monitoring, clinical decision support, rich visual displays of client outcomes, online training modules in ESTs, just-in-time training for guided real-time assistance in delivering ESTs, educational videos, and a client portal
Behavioral Activation for Depression, Cognitive Behavior Therapy for Insomnia, Collaborative Assessment and Management of Suicidality: Evidence-based interventions for treating depression and concurrent problems | 49 |
| Treatment-as-Usual Usual care without access to WILLOW.
Care-as-Usual Psychotherapy: Usual care psychotherapy for depression and concurrent problems | 36 |
| Total | 85 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Dropped/Excluded by PI | 6 | 6 |
| Overall Study | Lost to Follow-up | 2 | 5 |
| Overall Study | No participating patient | 16 | 13 |
| Overall Study | Participant withdrew | 7 | 2 |
Baseline characteristics
| Characteristic | Treatment-as-Usual | Total | WILLOW |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 1 Participants | 2 Participants | 1 Participants |
| Age, Categorical Between 18 and 65 years | 35 Participants | 83 Participants | 48 Participants |
| Age, Continuous | 41.2 years | 42.5 years | 43.4 years |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) Asian | 2 Participants | 4 Participants | 2 Participants |
| Race (NIH/OMB) Black or African American | 4 Participants | 9 Participants | 5 Participants |
| Race (NIH/OMB) More than one race | 1 Participants | 3 Participants | 2 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 1 Participants | 2 Participants | 1 Participants |
| Race (NIH/OMB) White | 28 Participants | 66 Participants | 38 Participants |
| Region of Enrollment United States | 36 participants | 85 participants | 49 participants |
| Sex: Female, Male Female | 20 Participants | 44 Participants | 24 Participants |
| Sex: Female, Male Male | 16 Participants | 41 Participants | 25 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 80 | 0 / 62 |
| other Total, other adverse events | 0 / 80 | 0 / 62 |
| serious Total, serious adverse events | 0 / 80 | 0 / 62 |
Outcome results
Primary
ASA-Monitoring and Feedback Version (ASA-MF)
The ASA-Monitoring and Feedback Version (ASA-MF) measure has 18 questions each having a scale of 1(Strongly Disagree) to 5(Strongly Agree) with a total range of 18-90. Higher scores on this 18-item scale reflect positive attitudes about standardized and routine progress monitoring, or positive outcomes.
Time frame: Baseline, 4 weeks, 8 weeks, 12 weeks
Population: Provider participants. Sample sizes are smaller because this measure was e not used in the pilot study.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | ASA-Monitoring and Feedback Version (ASA-MF) | Baseline | 68.0 score on a scale | Standard Deviation 6.24 |
| WILLOW | ASA-Monitoring and Feedback Version (ASA-MF) | Week 4 | 72.1 score on a scale | Standard Deviation 7.95 |
| WILLOW | ASA-Monitoring and Feedback Version (ASA-MF) | Week 8 | 72.2 score on a scale | Standard Deviation 7.56 |
| WILLOW | ASA-Monitoring and Feedback Version (ASA-MF) | Week 12 | 71.1 score on a scale | Standard Deviation 7.6 |
| Treatment-as-Usual | ASA-Monitoring and Feedback Version (ASA-MF) | Week 12 | 66.9 score on a scale | Standard Deviation 3.9 |
| Treatment-as-Usual | ASA-Monitoring and Feedback Version (ASA-MF) | Baseline | 64.8 score on a scale | Standard Deviation 6.83 |
| Treatment-as-Usual | ASA-Monitoring and Feedback Version (ASA-MF) | Week 8 | 68.3 score on a scale | Standard Deviation 5.88 |
| Treatment-as-Usual | ASA-Monitoring and Feedback Version (ASA-MF) | Week 4 | 64.8 score on a scale | Standard Deviation 6.47 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.14Chi-squared
Primary
Beck Depression Inventory (BDI-II)
Higher scores on this 21-item reflect greater depressive symptoms in the past two weeks. All items are summed to a total score (possible range 0-63). We categorized BDI scores into severity levels coded as 0, 1, and 2: Minimal (raw scores 0 through 13), Mild/moderate (14 through 28), and Severe (29 through 63).
Time frame: baseline, 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Beck Depression Inventory (BDI-II) | Baseline | 1.25 score on a scale | Standard Deviation 0.63 |
| WILLOW | Beck Depression Inventory (BDI-II) | Week 4 | 1.16 score on a scale | Standard Deviation 0.82 |
| WILLOW | Beck Depression Inventory (BDI-II) | Week 8 | 1.00 score on a scale | Standard Deviation 0.78 |
| WILLOW | Beck Depression Inventory (BDI-II) | Week 12 | .71 score on a scale | Standard Deviation 0.86 |
| Treatment-as-Usual | Beck Depression Inventory (BDI-II) | Week 12 | 1.04 score on a scale | Standard Deviation 0.74 |
| Treatment-as-Usual | Beck Depression Inventory (BDI-II) | Baseline | 1.52 score on a scale | Standard Deviation 0.65 |
| Treatment-as-Usual | Beck Depression Inventory (BDI-II) | Week 8 | 1.08 score on a scale | Standard Deviation 0.76 |
| Treatment-as-Usual | Beck Depression Inventory (BDI-II) | Week 4 | 1.12 score on a scale | Standard Deviation 0.78 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.036Chi-squared
Primary
Behavioral Activation for Depression Scale Short Form (BADS-SF)
The Behavioral Activation for depression Scale Short Form has 9 questions each having a scale from 0 (Not at All) to 6 (Completely) with a total range of 0-54. Higher scores on this nine-item scale reflect lower engagement in avoidance and higher engagement in activation behaviors over the course of BA for depression.
Time frame: baseline, 4 weeks, 8 weeks, 12 weeks
Population: Patient Participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Baseline | 18.8 score on a scale | Standard Deviation 7.37 |
| WILLOW | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 4 | 23.4 score on a scale | Standard Deviation 9.42 |
| WILLOW | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 8 | 24.6 score on a scale | Standard Deviation 9.05 |
| WILLOW | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 12 | 27.7 score on a scale | Standard Deviation 8.42 |
| Treatment-as-Usual | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 12 | 23.6 score on a scale | Standard Deviation 7.88 |
| Treatment-as-Usual | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Baseline | 17.3 score on a scale | Standard Deviation 6.47 |
| Treatment-as-Usual | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 8 | 21.3 score on a scale | Standard Deviation 7.52 |
| Treatment-as-Usual | Behavioral Activation for Depression Scale Short Form (BADS-SF) | Week 4 | 20.4 score on a scale | Standard Deviation 6.68 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.764Chi-squared
Primary
Current Assessment Practice Evaluation-Revised (CAPER)
We modified the original measure into nine items. The Current Assessment Practice Evaluation-Revised has 9 questions with each having a scale of 1(None, 0%) to 5 (Nearly All, 81-100%). Thus, the total range was 9-45 and higher scores reflect greater provider use of measurement-based care or positive outcomes.
Time frame: Baseline, 4 weeks, 8 weeks, 12 weeks
Population: Provider participants. Sample sizes are smaller because the measure was not used in the pilot study.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Current Assessment Practice Evaluation-Revised (CAPER) | Baseline | 17.6 score on a scale | Standard Deviation 8.07 |
| WILLOW | Current Assessment Practice Evaluation-Revised (CAPER) | Week 4 | 21.7 score on a scale | Standard Deviation 9.66 |
| WILLOW | Current Assessment Practice Evaluation-Revised (CAPER) | Week 8 | 21.9 score on a scale | Standard Deviation 9.8 |
| WILLOW | Current Assessment Practice Evaluation-Revised (CAPER) | Week 12 | 22.3 score on a scale | Standard Deviation 11.26 |
| Treatment-as-Usual | Current Assessment Practice Evaluation-Revised (CAPER) | Week 12 | 13.4 score on a scale | Standard Deviation 5.38 |
| Treatment-as-Usual | Current Assessment Practice Evaluation-Revised (CAPER) | Baseline | 12.4 score on a scale | Standard Deviation 5.04 |
| Treatment-as-Usual | Current Assessment Practice Evaluation-Revised (CAPER) | Week 8 | 13.4 score on a scale | Standard Deviation 5.66 |
| Treatment-as-Usual | Current Assessment Practice Evaluation-Revised (CAPER) | Week 4 | 13.2 score on a scale | Standard Deviation 5.13 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.001Chi-squared
Primary
Depression, Anxiety, and Stress Scale (DASS)
Higher scores on subscales from this 23-item measure reflected higher depression, anxiety, stress, and suicidality. Other researchers added two suicidality items to the original 21-item measure with permission from the original authors. Each subscale score is computed by summing the subscale items, then multiplying by 2. Total possible scores for the depression, anxiety, and stress subscales range from 0-42. We categorized DASS depression, anxiety, and stress scores into severity levels coded as 0, 1, and 2: Normal (0 through 9), Mild/moderate (10 through 20), and Severe/extremely severe (21 through 42) (Psychology Foundation of Australia, 2018). The DASS suicide subscale ranged from 0-12 and not recategorized.
Time frame: baseline, 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (anxiety subscale) | .74 score on a scale | Standard Deviation 0.77 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (anxiety subscale) | .58 score on a scale | Standard Deviation 0.81 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Baseline (stress subscale) | 1.55 score on a scale | Standard Deviation 0.68 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (stress subscale) | 1.32 score on a scale | Standard Deviation 0.7 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (stress subscale) | 1.29 score on a scale | Standard Deviation 0.64 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (stress subscale) | 1.10 score on a scale | Standard Deviation 0.75 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Baseline (suicide subscale) | 3.42 score on a scale | Standard Deviation 3.43 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (suicide subscale) | 2.32 score on a scale | Standard Deviation 3.31 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (suicide subscale) | 1.68 score on a scale | Standard Deviation 2.32 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (suicide subscale) | 1.74 score on a scale | Standard Deviation 3.26 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Baseline (depression subscale) | 1.55 score on a scale | Standard Deviation 0.68 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (depression subscale) | 1.32 score on a scale | Standard Deviation 0.79 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (depression subscale) | 1.1 score on a scale | Standard Deviation 0.75 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (depression subscale) | .94 score on a scale | Standard Deviation 0.94 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Baseline (anxiety subscale) | .97 score on a scale | Standard Deviation 0.8 |
| WILLOW | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (anxiety subscale) | .81 score on a scale | Standard Deviation 0.75 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (anxiety subscale) | .76 score on a scale | Standard Deviation 0.78 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (anxiety subscale) | .76 score on a scale | Standard Deviation 0.83 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (suicide subscale) | 1.04 score on a scale | Standard Deviation 1.84 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (anxiety subscale) | .72 score on a scale | Standard Deviation 0.74 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (depression subscale) | 1.08 score on a scale | Standard Deviation 0.76 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Baseline (stress subscale) | 1.60 score on a scale | Standard Deviation 0.58 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (suicide subscale) | 1.52 score on a scale | Standard Deviation 3.18 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (stress subscale) | 1.16 score on a scale | Standard Deviation 0.69 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Baseline (anxiety subscale) | .84 score on a scale | Standard Deviation 0.75 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 8 (stress subscale) | 1.24 score on a scale | Standard Deviation 0.72 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Baseline (depression subscale) | 1.68 score on a scale | Standard Deviation 0.56 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (stress subscale) | 1.00 score on a scale | Standard Deviation 0.76 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 12 (depression subscale) | 1.08 score on a scale | Standard Deviation 0.76 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Baseline (suicide subscale) | 1.92 score on a scale | Standard Deviation 2.68 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (depression subscale) | 1.4 score on a scale | Standard Deviation 0.65 |
| Treatment-as-Usual | Depression, Anxiety, and Stress Scale (DASS) | Week 4 (suicide subscale) | 1.28 score on a scale | Standard Deviation 1.99 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.734Chi-squared
Comparison: We used a Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.27Chi-squared
Comparison: We used a Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.75Chi-squared
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.812Chi-squared
Primary
Mental Health Services Satisfaction Survey (MHSSS)
We created a ten-item measure in which higher scores reflected greater satisfaction with services they received. The measure had 10 questions each having a scale of 1(Strongly Disagree) to 5(Strongly Agree) with a total range of 10-50. Higher scores indicated a better outcome.
Time frame: After first study therapy session, 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Mental Health Services Satisfaction Survey (MHSSS) | Baseline | 37.6 score on a scale | Standard Deviation 4.91 |
| WILLOW | Mental Health Services Satisfaction Survey (MHSSS) | Week 4 | 37.6 score on a scale | Standard Deviation 6.77 |
| WILLOW | Mental Health Services Satisfaction Survey (MHSSS) | Week 8 | 39.2 score on a scale | Standard Deviation 4.85 |
| WILLOW | Mental Health Services Satisfaction Survey (MHSSS) | Week 12 | 38.9 score on a scale | Standard Deviation 7.14 |
| Treatment-as-Usual | Mental Health Services Satisfaction Survey (MHSSS) | Week 12 | 38.3 score on a scale | Standard Deviation 7.65 |
| Treatment-as-Usual | Mental Health Services Satisfaction Survey (MHSSS) | Baseline | 38.8 score on a scale | Standard Deviation 7.24 |
| Treatment-as-Usual | Mental Health Services Satisfaction Survey (MHSSS) | Week 8 | 37.5 score on a scale | Standard Deviation 9 |
| Treatment-as-Usual | Mental Health Services Satisfaction Survey (MHSSS) | Week 4 | 37.8 score on a scale | Standard Deviation 7.14 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.44Chi-squared
Primary
Modified Practice Attitudes Scale (MPAS)
The Modified Practice Attitudes Scale with eight questions, each having a scale of 0 (Not at all) to 4(To a very great extent), has a total range of 0-32. Higher scores on this eight-item scale reflect more positive attitudes towards evidence-based practices.
Time frame: Baseline, 12 weeks
Population: Provider participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Modified Practice Attitudes Scale (MPAS) | Baseline | 24.8 score on a scale | Standard Deviation 3.06 |
| WILLOW | Modified Practice Attitudes Scale (MPAS) | Week 12 | 24.6 score on a scale | Standard Deviation 3.93 |
| Treatment-as-Usual | Modified Practice Attitudes Scale (MPAS) | Baseline | 24.9 score on a scale | Standard Deviation 4.84 |
| Treatment-as-Usual | Modified Practice Attitudes Scale (MPAS) | Week 12 | 23.6 score on a scale | Standard Deviation 5.77 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.15Chi-squared
Primary
Monitoring and Feedback Attitudes Scale (MFA)
The Monitoring and Feedback Attitudes Scale has 14 questions and each has a scale of 1(Strongly Disagree) to 5(Strongly Agree), thus the total range was 14-70. Higher scores on this 14-item measure indicate more positive attitudes about routine progress monitoring and providing feedback to patients about treatment progress, or positive outcomes.
Time frame: Baseline, 4 weeks, 8 weeks, 12 weeks
Population: Provider participants. Sample sizes are smaller because the measure was not used in the pilot study.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Monitoring and Feedback Attitudes Scale (MFA) | Baseline | 57.9 score on a scale | Standard Deviation 4.48 |
| WILLOW | Monitoring and Feedback Attitudes Scale (MFA) | Week 4 | 59.7 score on a scale | Standard Deviation 5.67 |
| WILLOW | Monitoring and Feedback Attitudes Scale (MFA) | Week 8 | 58.9 score on a scale | Standard Deviation 5.29 |
| WILLOW | Monitoring and Feedback Attitudes Scale (MFA) | Week 12 | 60.6 score on a scale | Standard Deviation 5.88 |
| Treatment-as-Usual | Monitoring and Feedback Attitudes Scale (MFA) | Week 12 | 60.1 score on a scale | Standard Deviation 5.41 |
| Treatment-as-Usual | Monitoring and Feedback Attitudes Scale (MFA) | Baseline | 55.1 score on a scale | Standard Deviation 5.68 |
| Treatment-as-Usual | Monitoring and Feedback Attitudes Scale (MFA) | Week 8 | 58.0 score on a scale | Standard Deviation 6.71 |
| Treatment-as-Usual | Monitoring and Feedback Attitudes Scale (MFA) | Week 4 | 57.2 score on a scale | Standard Deviation 5.83 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.22Chi-squared
Primary
Patient Health Questionnaire-9 (PHQ-9)
Higher scores on this ten-item measure reflects more severe depression. The total score (possible range 0-27) is created by summing the first nine items. We categorized PHQ severity levels coded as 0, 1, and 2 based on authors' research as minimal (0 through 4), mild/moderate (5 through 14), and moderately severe/severe (15 through 27). The recategorized scores are reported here.
Time frame: baseline, 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Patient Health Questionnaire-9 (PHQ-9) | Baseline | 1.48 score on a scale | Standard Deviation 0.57 |
| WILLOW | Patient Health Questionnaire-9 (PHQ-9) | Week 4 | 1.23 score on a scale | Standard Deviation 0.62 |
| WILLOW | Patient Health Questionnaire-9 (PHQ-9) | Week 8 | 1.24 score on a scale | Standard Deviation 0.64 |
| WILLOW | Patient Health Questionnaire-9 (PHQ-9) | Week 12 | .97 score on a scale | Standard Deviation 0.71 |
| Treatment-as-Usual | Patient Health Questionnaire-9 (PHQ-9) | Week 12 | 1.08 score on a scale | Standard Deviation 0.76 |
| Treatment-as-Usual | Patient Health Questionnaire-9 (PHQ-9) | Baseline | 1.44 score on a scale | Standard Deviation 0.58 |
| Treatment-as-Usual | Patient Health Questionnaire-9 (PHQ-9) | Week 8 | 1.28 score on a scale | Standard Deviation 0.61 |
| Treatment-as-Usual | Patient Health Questionnaire-9 (PHQ-9) | Week 4 | 1.24 score on a scale | Standard Deviation 0.6 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.664Chi-squared
Primary
Perceptions of Computer-Assisted Therapy Questionnaire (PCAT)
The Perceptions of Computer-Assisted Therapy Questionnaire has 34 questions each having a scale of 1(Strongly Disagree) to 7 Strongly Agree). Thus, the total range was 34-238. Higher scores on this 34-item scale indicate stronger agreement with the benefits of computer-assisted therapy, or a better outcome.
Time frame: Baseline, 12 weeks
Population: Provider participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Perceptions of Computer-Assisted Therapy Questionnaire (PCAT) | Baseline | 160.1 score on a scale | Standard Deviation 19.39 |
| WILLOW | Perceptions of Computer-Assisted Therapy Questionnaire (PCAT) | Week 12 | 170.4 score on a scale | Standard Deviation 24.6 |
| Treatment-as-Usual | Perceptions of Computer-Assisted Therapy Questionnaire (PCAT) | Baseline | 157.4 score on a scale | Standard Deviation 18.7 |
| Treatment-as-Usual | Perceptions of Computer-Assisted Therapy Questionnaire (PCAT) | Week 12 | 146.4 score on a scale | Standard Deviation 21.7 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: <0.001Chi-squared
Primary
Routine Monitoring (RM)
We created a three-item measure where the total score is found by summing the items. The first two items are yes(scored 1)/no(scored 0) questions. The last item is a count measure that can be any number greater than or equal to 0. Thus, the total score can have any integer value greater than or equal to 0, with higher scores reflecting greater use of routine progress monitoring.
Time frame: Baseline, 4 weeks, 8 weeks, 12 weeks
Population: Provider participants. Sample sizes are smaller because the measure was not used in the pilot study.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Routine Monitoring (RM) | Baseline | 1.6 score on a scale | Standard Deviation 1.76 |
| WILLOW | Routine Monitoring (RM) | Week 4 | 4.6 score on a scale | Standard Deviation 2.7 |
| WILLOW | Routine Monitoring (RM) | Week 8 | 3.6 score on a scale | Standard Deviation 2.39 |
| WILLOW | Routine Monitoring (RM) | Week 12 | 3.3 score on a scale | Standard Deviation 2.25 |
| Treatment-as-Usual | Routine Monitoring (RM) | Week 12 | 1.9 score on a scale | Standard Deviation 1.36 |
| Treatment-as-Usual | Routine Monitoring (RM) | Baseline | .60 score on a scale | Standard Deviation 1.06 |
| Treatment-as-Usual | Routine Monitoring (RM) | Week 8 | 1.3 score on a scale | Standard Deviation 1.16 |
| Treatment-as-Usual | Routine Monitoring (RM) | Week 4 | 1.8 score on a scale | Standard Deviation 1.36 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: <0.001Chi-squared
Primary
Therapy Task Checklist (TTC) - Patient Version
We developed the TTC. The Therapy Task Checklist-Patient Version with 25 questions and each having a scale of 0 (Never/Very Rarely) to 7 (All of the Time) has a total range of 0-100. Higher scores reflected more frequent use of evidence-based therapy tasks, positive outcomes.
Time frame: 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Therapy Task Checklist (TTC) - Patient Version | Week 4 | 45.5 score on a scale | Standard Deviation 18.76 |
| WILLOW | Therapy Task Checklist (TTC) - Patient Version | Week 8 | 46.6 score on a scale | Standard Deviation 16.95 |
| WILLOW | Therapy Task Checklist (TTC) - Patient Version | Week 12 | 49.6 score on a scale | Standard Deviation 19.7 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Patient Version | Week 4 | 37.9 score on a scale | Standard Deviation 18.84 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Patient Version | Week 8 | 42.3 score on a scale | Standard Deviation 19.48 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Patient Version | Week 12 | 39.0 score on a scale | Standard Deviation 19.69 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.19Chi-squared
Primary
Therapy Task Checklist (TTC) - Provider Version
The Therapy Task Checklist -Provider Version has 22 questions and each have a scale of 0(Never/Very Rarely) to 4 (All of the time) with a total range of 0-88. Higher scores reflected more frequent use of evidence-based therapy tasks, or positive outcomes.
Time frame: 4 weeks, 8 weeks, 12 weeks
Population: Provider participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Therapy Task Checklist (TTC) - Provider Version | Week 4 | 48.7 score on a scale | Standard Deviation 13.19 |
| WILLOW | Therapy Task Checklist (TTC) - Provider Version | Week 8 | 51.3 score on a scale | Standard Deviation 13.33 |
| WILLOW | Therapy Task Checklist (TTC) - Provider Version | Week 12 | 50.8 score on a scale | Standard Deviation 13.37 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Provider Version | Week 4 | 40.7 score on a scale | Standard Deviation 12.51 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Provider Version | Week 8 | 43.7 score on a scale | Standard Deviation 10.06 |
| Treatment-as-Usual | Therapy Task Checklist (TTC) - Provider Version | Week 12 | 42.9 score on a scale | Standard Deviation 7.91 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.99Chi-squared
Primary
Working Alliance Inventory (WAI)
We modified this measure to include items from the BHS from Group Health. The resulting 12-item scale measures the therapeutic relationship with higher scores indicating stronger alliance. With 12 questions and each having a scale of 1(Never) to 7(Always) the total range was 12-84.
Time frame: 4 weeks, 8 weeks, 12 weeks
Population: Patient participants
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| WILLOW | Working Alliance Inventory (WAI) | Week 4 | 71.7 score on a scale | Standard Deviation 10.44 |
| WILLOW | Working Alliance Inventory (WAI) | Week 8 | 73.7 score on a scale | Standard Deviation 9.13 |
| WILLOW | Working Alliance Inventory (WAI) | Week 12 | 73.6 score on a scale | Standard Deviation 10.46 |
| Treatment-as-Usual | Working Alliance Inventory (WAI) | Week 4 | 71.9 score on a scale | Standard Deviation 11.54 |
| Treatment-as-Usual | Working Alliance Inventory (WAI) | Week 8 | 70.5 score on a scale | Standard Deviation 11.23 |
| Treatment-as-Usual | Working Alliance Inventory (WAI) | Week 12 | 72.7 score on a scale | Standard Deviation 10.85 |
Comparison: We used the Generalized Linear Model to examine the between-condition change over time on study outcomes. We computed the design effects due to this nested design to determine the extent to which the intraclass correlation due to clustering and size of the cluster affected the analyses. Design effects ranged up to 1.8. Because all were under 2.0, we did not adjust analyses to account for clustering effects.p-value: 0.09Chi-squared