| Generalized Anxiety Disorder |
|---|
| Dahlin (2016)11 |
|---|
Randomization was performed with an online random-number service by an employee of the university with no connection to the study. During the nine weeks of treatment administration, the control group had no contact with study administrators. Intention-to-treat analysis was conducted. Drop-outs were relatively balanced between treatment and control groups.
|
The study was conducted in Sweden and the generalizability to patients in Canada is unclear. The majority of participants (76.7%) had university education. Treatment support was provided by trained graduate students rather than licensed therapists. In addition, the same students who provided guidance conducted the initial diagnostic telephone interview, which may have affected the type of support provided (based on knowledge of scores on the diagnosis questionnaires). The primary author is employed by the company that developed the treatment program, which is a potential conflict of interest.
|
| Panic Disorder |
|---|
| Ciuca (2018)12 |
|---|
Randomization was performed with software that implemented a minimization algorithm, which balanced groups with respect to disease severity and chronicity. Allocation was conducted by an independent researcher. Also, researchers involved in recruitment and screening had no knowledge or control over allocation. Treatment groups were overall balanced in baseline characteristics. Missing data were imputed as treatment failures, which resulted in conservative estimates. Intention-to-treat analysis was conducted.
|
The study was conducted in Romania and the generalizability to patients in Canada is unclear. Large number of drop-outs (27%). Blinding of assessors was compromised because some participants disclosed information about treatment during interviews. Authors of study had affiliations with the web-based software used to deliver the iCBT program.
|
| Ivanova (2016)13 |
|---|
Randomization was performed with a random number service and stratified by primary diagnosis (i.e. panic disorder and social anxiety disorder). The randomization was conducted by a researcher with no relation to the study. Intention-to-treat analysis was conducted.
|
The study was conducted in Sweden and the generalizability to patients in Canada is unclear. Therapists in training conducted the intervention, although they were supervised by a licensed clinical psychologist. Small sample size (N=39). Baseline characteristics combined for panic disorder and social anxiety disorders (unclear if characteristics were balanced for participants with panic disorder). Two of the authors were employed by the company that develops and distributes the research products used in the study.
|
| Social Anxiety Disorder |
|---|
| Johansson (2017)14 |
|---|
Randomization was performed with a random number service. An independent researcher conducted the randomization and allocation of participants. Intention-to-treat analysis was conducted. Small number of drop-outs at post-treatment (1/36 in treatment group and 1/36 in control). Long-term follow-up of treatment group (24 months). The CGI-I telephone interviewers were conducted by final-year clinical psychology students who were blind to treatment allocation.
|
The study was conducted in Sweden and the generalizability to patients in Canada is unclear. Therapists in training conducted the intervention, although they were supervised by more experienced therapists. The control group had more females than males (72.2% vs. 27.8%).
|
| Ivanova (2016)13 |
|---|
Randomization was performed with a random number service and stratified by primary diagnosis (i.e. panic disorder and social anxiety disorder). The randomization was conducted by a researcher with no relation to the study. Intention-to-treat analysis was conducted.
|
The study was conducted in Sweden and the generalizability to patients in Canada is unclear. Therapists in training conducted the intervention, although they were supervised by a licensed clinical psychologist. Baseline characteristics combined for panic disorder and social anxiety disorders (unclear if characteristics were balanced for participants with social anxiety disorder). Two of the authors were employed by the company that develops and distributes the research products used in the study.
|
| Schulz (2016)15 |
|---|
Randomization was performed with a computerized random number generator and concealed from investigators. Participants were informed about their group allocation via email. Intention-to-treat analysis was conducted. Drop-outs were regarded as treatment failures for post-treatment diagnostic status (for other outcomes, missing data were imputed with mixed-effect models, which use all available data on a subject). Treatment groups were balanced overall.
|
The study was conducted in Switzerland, Austria, and Germany; generalizability to patients in Canada is unclear. The assessors of post-diagnostic status were not blinded to treatment allocation. Large percentage of drop-outs (25%).
|
| Mixed |
|---|
| Silfvernagel (2017)10 |
|---|
Randomization was performed with an online random number-generation service independent of investigators and therapists. At post-treatment, semi-structured telephone interviews were conducted by blinded assessors who had no earlier contact with participants. Intention-to-treat analysis was conducted.
|
The study was conducted in Sweden and the generalizability to older adults in Canada is unclear. The majority of participants (53%) had college or university education. Some imbalances were present in the treatment and control groups: more participants were employed in the control group (30.3% vs. 9.1%) and more participants in the control group had no experience with psychotherapy (46% vs. 24.2%). Although unclear, this suggests that potentially allocation to randomized groups may have been affected. A large percentage of participants were lost to follow-up in the treatment group (33.3%). Data were assumed to be missing at random, however, given the larger number of drop-outs in treatment group vs. control, this assumption may not be accurate. The control group was administered general weekly e-mail support, however details of this intervention were not provided (e.g. did all participants in the control group receive the same support or were there any overlaps with the treatment group). Details about the nature of therapist guidance were not provided.
|
| Dear (2015)16 |
|---|
Permuted block randomization sequence was generated with a random number generator by an independent researcher at another institution. Allocation assignments were kept in sealed envelopes. Overall, groups were balanced in baseline characteristics. Intention-to-treat analysis was conducted (assumed that data missing at random) Small number of withdrawals or post-treatment non-responders (10% in treatment group and 13.5% in control).
|
The study was conducted in Australia; generalizability to patients in Canada is unclear. An initial inclusion criterion of ≥8 on the Generalized Anxiety Disorder-7 Item Scale was removed during early stages of recruitment because many applicants did not meet this cut-off value. However, analyses were conducted separately for the subgroup of participants that did meet the criterion. Two participants randomized to treatment group were subsequently excluded from analyses (reason unclear). Two of the authors are developers of the internet-delivered cognitive behavior therapy course used in the study.
|
