| Study |
Bias |
| Randomisation process |
Deviations from intended interventions |
Missing outcome data |
Measurement of the outcome |
Selection of the reported results |
Overall |
| Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
| Bell 2019 |
High risk of bias |
It is indicated that "Twenty‐four eligible adults were enrolled on a first‐come, first‐served basis and alternately assigned between the LZNF group and HRVB group according to the order in which they returned their prescreening materials." Thus, this was quasi‐randomisation and there was a lack of allocation concealment. "The largest baseline differences were in the HRV measures, for which the LZNF group had higher initial levels." |
Low risk of bias |
Participants and intervention deliverers probably knew assignment but analysis likely appropriate. |
Low risk of bias |
Data available for all or nearly all. |
Low risk of bias |
Method of measuring outcome not inappropriate. |
Low risk of bias |
Probably analysed in accordance with pre‐specified plan. |
High risk of bias |
Although this outcome is about dropout from the trial and many features lead to confidence about that result, there remains the issue that the randomisation process was susceptible to bias. |
| Bomyea 2015 |
Low risk of bias |
Individuals who completed all baseline assessments were randomly assigned to the HIC condition or the LIC condition based on a computer‐generated random number system prior to attending the first training session. Conditions were assigned by an independent third party using computer software, so that participants and research personnel remained blind to subjects’ conditions. Participants in the two groups did not differ on age or measures of clinical characteristics and no differences were identified in participants' ethnic background, education, income, or marital status. |
Low risk of bias |
Unaware of assignment. |
Low risk of bias |
7/22 (intervention) v 10/20 (control) ‐ based on flowchart we selected the Ns that did not receive the intervention and started but discontinued and use the randomised ns as the denominators. |
Some concerns |
However it is not clear if treatment completion meant all 8 sessions were attended and this was important to state. |
Some concerns |
For dropout, it is fairly straightforward. |
Some concerns |
This study adequately measures and reports treatment dropout though more clarity on required number of sessions (definition of completion) would have been useful. |
| Brady 2021 |
Some concerns |
“After diagnostic assessment and completion of baseline outcome measures, a research assistant randomised participants to either trial condition using a virtual coin toss programme.” “Randomisation did not yield equal allocation across the two groups. There were also some notable group differences in the distribution of other participant characteristics.” |
Low risk of bias |
The interventions were delivered effectively to the best of our knowledge. |
Low risk of bias |
Dropout from the study interventions was low (as distinct from missing data on outcomes). |
Low risk of bias |
We had no concerns that the researchers were able to accurately report on the numbers who exited the intervention which was 2/15 (intervention) and 3/10 (minimally active control) |
Low risk of bias |
No concerns that there was biased reporting on dropout. |
Some concerns |
It is conceivable that the imbalance at baseline on characteristics of participants could have interacted with outcomes including susceptibility to leaving the intervention. |
| Foa 2006 |
Some concerns |
There was a lack of information about randomisation. "At the start of the study, participants were randomly assigned to either four sessions weekly of B‐CBT or AC." On the other hand, participants in the assessment condition and in the CBT (Brief CBT) did not seem to differ in terms of initial psychopathology or demographic variables. |
Low risk of bias |
Completer analysis was undertaken however the current outcome is not so much affected by this |
Low risk of bias |
No obvious risk to ascertaining the dropout rates. |
Some concerns |
9/31 v 10/30, however did not detail how many sessions required to consider completion. It was unclear what thresholds were used for brief CBT v assessment control condition. |
Some concerns |
Numerical result here not assessed based on a selection made by the researchers. |
Some concerns |
Information on number of sessions lacking by group and what precisely constituted dropout from treatment. |
| Littleton 2016 |
Low risk of bias |
"Participants were randomized to the interactive program or psychoeducational website based on a computerized coin flip." |
Low risk of bias |
Participants and intervention deliverers were likely aware of assignment but appropriate analysis appears to have been used. There is evidence that "External ratings of the competence of the feedback provided by therapists were uniformly high." |
Low risk of bias |
Data available for all or nearly all participants. |
High risk of bias |
Some concern that there was a low threshold for treatment completion, suggesting the 'dose' ended up as low for most participants. e.g. just "Six participants (15.8%) completed the entire program." A higher expectation about 'completion level' would have meant greater dropout from treatment. It is difficult to align what is in text with the numbers in the flow diagram for the study and so we selected the Ns reported the study's Figure 1 for non initiation and discontinued. It is not entirely clear how dropout was ascertained in the control (minimal intervention). |
Some concerns |
There are several ways of expressing treatment dropout and the authors may have selected a perspective that made the result look more favourable. |
High risk of bias |
The study reports a favourable rate of treatment adherence when in fact only 15.8% of people in the treatment group went through the full program. |
