| 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 |
| Grote 2012 |
Some concerns |
Method used to generate random sequence is adequate: "Using a permuted block design stratified by race, we randomly assigned participants to receive either enhanced IPT‐B or UC"
Methods used to conceal allocation not described
No significant differences reported between groups: "Analyses show that participants in the IPT‐B and UC groups did not differ significantly on any of the baseline demographic or clinical characteristics" Note group size in total sample across IPT‐B and UC is equal, however group size is unequal when looking only at 'more exposure' (i.e., n = 6 in IPT‐B, n = 18 in UC). However, it would be more problematic if larger sample size was in IPT‐B group, as could indicate that preference was given to putting pts in the intervention group |
Low risk of bias |
Participants and people delivering the interventions would have likely been aware of their group allocation due to differences in the interventions
No deviation from intended intervention specified, unlikely that the outcome would be impacted by lack of blinding
Modified ITT excluding single participant with missing outcome data |
Low risk of bias |
Data available for almost all participants (data seems to be missing from one participant only) |
Some concerns |
Validated outcome measure used
Same measurement and assessment timepoints used for all participants
Measurement tool is self‐report so participants are assessors and cannot be blinded. Also states in limitation section: "raters may have been able to determine intervention due to differences in ease on contact to conduct follow up assessment"
The outcome assessment could be potentially influenced by the knowledge of the intervention received. It is unlikely that the reporting was influenced by the treatment received |
Low risk of bias |
No published protocol, however outcomes in methods section align with those in the results
All reported results correspond to intended outcome measurements
All reported results correspond to intended analyses |
Some concerns |
Overall some concerns due to no information about allocation concealment and potential bias in the measurement of outcome (self‐report) |
| Grote 2015 |
Low risk of bias |
Method used to generate random sequence and conceal allocation is adequate: "Randomization to the MOMCare collaborative care intervention or to MSS public health services proceeded by means of an adaptive block randomization scheme to ensure that intervention and MSS groups maintained equal sizes. Randomization was stratified on initial depression severity (SCL‐20 ≤ 40 or > 40) and gestational age (< 22 weeks or ≥ 22 weeks). Within each of the 4 strata we created random orders of block sizes of either 2 or 4 study arm assignments in order to insure balance of intervention participants within each strata. Stratified block randomization was developed by our statistician and carried out via a computerised program by the depression care specialist (DCS) responsible for screening in 3‐4 public health centres. For each eligible patient, the DCS first entered the stratification variables into her laptop computer program and then received immediate notification of the patient's group assignment"
No significant differences reported between groups: "The study groups were well‐balanced on socio‐demographic and clinical variables at baseline, except that MOMCare participants were more likely to be unemployed, a difference for which we controlled in the analyses" |
Low risk of bias |
Participants and people delivering the interventions would have likely been aware of their group allocation due to differences in the interventions – add any additional sentences
No deviation from intended intervention specified, unlikely that the outcome would be impacted by lack of blinding
Modified ITT excluding participants with missing outcome data |
Some concerns |
Dropout range from 0‐19% (based on unpublished subgroup data provided by authors), with more dropout in TAU (MTT+) group (19%) vs MOMCare (0%)
No evidence that the results was not biased by the missing data; copy and paste quote if included
Reasons for dropout not reported |
Low risk of bias |
Validated outcome measure used
Same measurement and assessment timepoints used for all participants
Outcome assessors were blinded to treatment condition, even though conventionally the outcome is self‐report: "All assessments were conducted via phone by an interviewer blinded to intervention status" |
High risk of bias |
Published protocol, trial registered (NCT01045655)
Reported results do not correspond to intended outcome measurements, suggesting missing outcome data for non‐significant effect: "Treatment group differences refer to a comparison of the mean of the 3‐, 6‐, 12‐, and 18‐month follow‐up assessments, controlling for baseline values. In the event of a significant main effect for group, we report the pattern of means at each time point" Results section states: "In Tables 2 and 3, we show the significant main effects for group for each variable, with the relevant P‐values at each time point".
All reported results correspond to intended analyses |
High risk of bias |
Overall some concerns due to missing outcome data (unbalanced dropout) and risk that outcomes were selected after the analysis were completed |
| Madigan 2015 |
Some concerns |
Method used to generate random sequence and to conceal allocation not described
No significant differences reported between groups: "There were no significant differences between the intervention and control group, or the participating and dropped out subjects in terms of socio‐demographic variables, the main study criteria or other mental health variables." |
Low risk of bias |
Unclear if participants were aware of intervention assignment as the intervention consisted of trauma focused CBT and parenting program while the control group received the parenting program. All participants informed the study was piloting two interventions. No details provided on participant blinding.
Carers and people delivering the interventions would have likely been aware of their group allocation due to differences in the interventions
No deviation from intended intervention specified, unlikely that the outcome would be impacted by lack of blinding
Modified ITT excluding participants with missing outcome data |
High risk of bias |
40% dropout (from total sample)
No evidence that the results was not biased by the missing data
Although reasons for dropout not reported, likely that missing mechanism is related to the true value of the outcome: "Difficulty with treatment adherence is frequently reported in adolescents, and pregnant adolescents in particular, as their desire to engage in research is generally poor" |
Some concerns |
Validated outcome measure used
Same measurement and assessment timepoints used for all participants
Measurement tool is self‐report so participants are assessors and cannot be blinded
The outcome assessment could be potentially influenced by the knowledge of the intervention received. It is unlikely that the reporting was influenced by the treatment received |
Low risk of bias |
No published protocol, however outcomes in methods section align with those in the results
All reported results correspond to intended outcome measurements
All reported results correspond to intended analyses |
High risk of bias |
Overall some concerns due to no information on the randomization process, missing outcome data (large dropout) and potential bias in the measurement of outcome (self‐report) |
