Increasing Access to Mental Health Care With Breathe, an Internet-Based Program for Anxious Adolescents: Study Protocol for a Pilot Randomized Controlled Trial

Background There is a demand to make first-line treatments, including cognitive behavioural therapy (CBT) for adolescent anxiety disorders, more widely available. Internet-based CBT is proposed to circumvent access and availability barriers and reduce health care system costs. Recent reviews suggest more evidence is needed to establish the treatment effects of Internet-based CBT in children and adolescents and to determine related economic impacts. Objective This pilot trial aims to collect the necessary data to inform the planning of a full-scale RCT to test the effectiveness of the Internet-based CBT program Breathe (Being Real, Easing Anxiety: Tools Helping Electronically). Methods We are conducting a 27-month, 2-arm parallel-group, pilot randomized controlled trial (RCT). Outcomes will inform the planning of a full-scale RCT aimed to test the effectiveness of Internet-based CBT with a population of adolescents with moderate to mild anxiety problems. In the pilot RCT we will: (1) define a minimal clinically important difference (MCID) for the primary outcome measure (total anxiety score using the Multidimensional Anxiety Scale for Children); (2) determine a sample size for the full-scale RCT; (3) estimate recruitment and retention rates; (4) measure intervention acceptability to inform critical intervention changes; (5) determine the use of co-interventions; and (6) conduct a cost-consequence analysis to inform a cost-effectiveness analysis in the full-scale RCT. Adolescents aged 13-17 years seeking care for an anxiety complaint from a participating emergency department, mobile or school-based crisis team, or primary care clinic are being screened for interest and eligibility. Enrolled adolescents are being randomly allocated to either 8 weeks of Internet-based CBT with limited telephone and e-mail support, or a control group with access to a static webpage listing anxiety resources. Adolescents are randomly assigned using a computer generated allocation sequence. Data are being collected at baseline, treatment completion, and at a 3-month follow-up. Results Currently, adolescents are being enrolled in the study. Enrolment is taking place between March 2014 and February 2016; data collection will conclude May 2016. We expect that analysis and results will be available by August 2016. Conclusions In many communities, the resources available for front-line anxiety treatment are outweighed by the need for care. This pilot RCT is an essential step to designing a robust RCT to evaluate the effectiveness of an Internet-based CBT program for adolescents with moderate to mild anxiety problems. Trial Registration Clinicaltrials.gov NCT02059226; http://clinicaltrials.gov/ct2/show/NCT02059226 (Archived by WebCite at http://www.webcitation.org/6epF8v7k4)


SYNOPSIS OF PROPOSAL
The overall purpose of the proposed project is to develop and carry out preliminary testing of an Internet-based cognitive behavioural therapy (CBT) for anxious youth (13-17 years) having visited an emergency department (ED). It is a 3 year project entailing 3 studies: Study 1: Internet-based intervention development (goal 1) Study 2: Real-time intervention evaluation of usability (goal 2) Study 3: Preliminary data for a future multi-site RCT (goal 3) Applicants have provided their response to reviewers regarding their earlier submission.

RESEARCH APPROACH Through the project,
, an Internet-based CBT program for youth will be developed. Each of the three Wired! studies has specific objectives, design, and methods. , 8 CBT modules will be developed for which will include process (e.g. assessment of In Study 1 Wired! module upon completion), core content (e.g. coping skill activities, relapse prevention), and tailored (e.g. mood/behaviour, assertiveness training) modules. The approach follows Dr. McGrath's Feeling Better program development. A Youth Panel (n=10) and a Clinical Expert Panel (n=10) will be formed through maximum variation sampling and convened through (online interaction). Through either Elluminate Live focus groups or individual semi-structured interviews, panel members will provide input on module content. For the youth panel, youth 17-24 years will be recruited. However, the focus of the intervention is on youth Data will be entered into N-Vivo. 13-17 years. What is the rationale for the age variation?
Is use of a computer assisted qualitative data analysis necessary given the sample sizes above (10+ 10) and specific objectives of this phase of the study? , the usability (efficiency, errors, and acceptability) of will be evaluated with participants In Study 2 Wired! from the Youth Panel (n=8 to 10) with assistance from a usability analyst. The evaluation process is innovative (building on previous work conducted by Dr. McGrath) and iterative (2-3 cycles or until no further problems are identified).
, entails a demonstration RCT with an Internet-delivered CBT through (experimental, n=40) Study 3 Wired! group and a resource webpage (control, n=40) group (with retention goal of 20 per group at 3 month follow-up). The goal is to evaluate methodological processes and outcomes (e.g. sample size estimation, changes in anxiety levels pre and post-intervention) in preparation for a full-scale RCT in the future by gathering relevant information at this phase. Participants will be recruited from an Edmonton and a Halifax health care site.
Among the strengths of the proposal are its originality and potential impact (see below sections). The project builds on previous work conducted in the field and will contribute to new knowledge development. Also two study potential challenges are considered (1. Trial may affect usual ED treatment, and 2. Risk management concerns in treatment trials for youth). Appendix 1 outlines information on the proposed research staff training workshop.
As per information presented in the proposal's background section, anxiety disorders (AD) are among the most prevalent mental health problems in children and youth. This section also notes that AD rates are comparable among boys and girls. It would be helpful to elaborate how the authors came to this conclusion given other bodies of literature pointing to gender differences, with females being at higher risk of AD. Otherwise, the proposal has a comprehensive and well-written review of literature of AD in youth and challenges to use of computer-assisted treatments Given the multi-study aspect of the proposed project and specific outcomes expected for each of the 3 studies, a potential threat to it may be its timely completion in three years. Does the research team have ? contingency plans in case of unexpected delays for each of the studies ORIGINALITY OF PROPOSAL Delivering CBT to youth with AD is a new area of research and practice. However, the proposal draws from work in related fields and individualizes it to youth with AD. To this end, the multi-stage innovative project ENVIRONMENT FOR RESEARCH Content development for will take place in Edmonton (led by Dr. Newton) and web development will Wired! take place in Halifax (led by Dr. McGrath). Recruitment of youth will take place in Edmonton and Halifax. Team members have prior and developed connections with agencies and professionals both in the clinical and technological aspects of the proposed project.
IMPACT OF RESEARCH Given the potential high appeal of the on-line medium for most youth, lack of comprehensive follow-up for youth with AD following their ED visit, and potential for improvement in outcome through CBT, the project can be highly relevant to youth (and their families) with AD.  This is a resubmission of a previous application. The rationale for this proposal is that anxiety disorders are quite prevalent in youth and are a common cause for emergency room visits. However, due to lack of resources in the community, there is delayed time for post crisis treatment with a median length of 35 days. Internet based delivery of cognitive behavior therapy for anxiety may help to reduce anxiety related crisis and reduce repeat emergency room visits. This is a 3-phase application, with year one focusing on development, refinement and pre-testing of a web-based intervention. The second part of the research plan will involve a qualitative study with stakeholders helping to refine the content and delivery features. This will then be pre-tested by youth. Lastly, a pilot randomized control trial will form the basis for the remaining aspect of the study. Youth ages 13 to 17 will be recruited following discharge from the emergency department after presenting for anxiety. They will be randomized to either the internet CBT group or a control group, which is a website providing general information on resources for anxiety. Pre and post measures will include change in youth anxiety from a validated measure (MASC) from baseline to 3 months. The length of the intervention is weeks. Data from this will also be used to estimate a sample recruitment and retention rate for future large scale RCT to determine important differences in anxiety change measure intervention acceptability and determine the use of co-interventions.
This application has numerous strengths. The applicants have revised their application to address the previous reviewer's comments in a very detailed and thoughtful manner. This has significantly strengthened their current application. The significance of the issue is significant for Canadian youth and this study will fill a gap in this area for this particular age group who have not yet been studied using similar methodologies. The applicants have thought through many of the details extensively as evidenced by the appendices with the grant. Other positive inclusions are the use of CBT as a treatment method and the ability of the program to be tailored to the individual youth, along with safety measures that have been included for the subjects.
The investigators seem well established and have the necessary expertise to carry out this study. The investigators modestly assume that 50% of potential subjects will decline to enter, however given the acute situation and the interest in receiving care, one may actually assume that a higher participation rate will ultimately occur. Statistical methods appear appropriate.
It was actually very difficult to find the duration of the intervention in the grant application as it is not stated in the methods (8 weeks was seen in one of the appendices). Given the median time is 35 days to receive treatment, one would assume that the majority of the participants will be receiving other additional therapy outside of the study design. The authors do not address this nor how this co-intervention will be accounted for. Secondly, the background does highlight that many of these youth may have repeat visits to the emergency room prior to receiving care in the community. It would be helpful to know the actual statistics in regards to this and whether this outcome will also be captured in this study.
Budget seems adequate for the scope of the study.
The principal applicant is a relatively new investigator with an interest in emergency utilization and mental health. She has published 38 manuscripts, almost half within the last 2 years. She holds a CIHR New Investigator Salary Award Research award and a CIHR Canadian Child Health Clinician Scientist Program Career Development Award. She is well supported by the co-PI who is a senior investigator with experience in clinical trials and recent research activity in the mental health field.