Table 3 Risk of Bias summary according to ROBINS-I in the non-randomized studies
1stauthor, year, study design | Bias due to confounding | Bias in selection of participants into the study | Bias in classification of interventions | Bias due to deviations from intended interventions | Bias due to missing data | Bias in measurement of outcomes | Bias in selection of the reported results | Overall bias |
|---|---|---|---|---|---|---|---|---|
Haggård, 2015, Sweden, Controlled cohort retrospective study [31] | Low risk Controlled statistically for baseline recidivism risk that might confound the association between treatment status and recidivism | Low risk All participants who would have been eligible for the target trial were included in the study | Low risk The intervention group (IG) and control group (CG) are clearly defined. Information about intervention status was obtained retrospectively | Low risk Followed an Intention-To-Treat (ITT)-approach. Other co-interventions that might have affected the outcome were balanced across IG and CG. The assessor extracting data was blinded to recidivism data | Low risk Attrition from treatment was described (IG 27%). The study had complete outcome measurement based on registry information | Low risk Retrospective study with already reported outcomes | Moderate Performed the study after the intervention was finished. There is no published protocol, making it difficult to know if the outcomes were pre-defined | Moderate |
Boira,2013, Spain, Quasi-experimental study [2] | Low risk The participants were selected from the target population. The study had controlled for possible baseline confounding, and the three groups are comparable for sociodemographic characteristics | Low risk All participants who would have been eligible for the target trial were included in the study | Low risk The four groups are clearly defined | No information Insufficient information with respect to ITT-analysis and adherence to interventions | Moderate There is unclear information about recidivism data on the intervention groups at 18-months follow up (the outcome is presented as total participants, N = 44, making it impossible to separate the effects between the four groups status). Low attrition. It is unclear how missing data was analysed | Moderate Lack of blind outcome assessments. The methods of outcome assessment were comparable across three of the four groups for pre- and post-assessments. Errors of measurement occurs at 18-months follow-up (non-differential measurements are presented with respect to conviction) | Serious There is no published protocol, making it difficult to know if the outcomes were pre-defined The lack of differentiation between treatment modalities in presenting the results at 18-month follow-up makes it difficult to judge whether the observed effect is associated to group treatment | Serious |