Inclusion criteria for the review were clearly defined. Several relevant databases were searched. Publication bias was not assessed and could not be ruled out. Attempts were made to reduce reviewer error and bias throughout study selection and data extraction; it was unclear whether this was also the case for quality assessment. Quality assessment was undertaken using a standard tool which indicated that most trials were of moderate quality.
There was some discrepancy between text and tables for reporting of patient numbers. The authors noted that the comparator was not defined clearly. Data were synthesised narratively and this seemed appropriate for the type of data. The synthesis was very limited and no quantitative data or significance levels were given, so it was impossible to determine the significance of findings. The goal setting component seemed limited. The authors noted that there were differences in the study populations and interventions. The authors also noted that the limitations of methodological quality and lack of homogeneity meant it was not possible to identify the most successful elements of the intervention.
The limited narrative synthesis and differences across trials and potential quality issues mean that the authors’ conclusions should be viewed with caution as they may not be reliable.