Risk of bias for analysis 1.1 All‐cause mortality at longest follow‐up.
| 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 | |
| Lin 2004 | Some concerns | The review authors noted that the only information reported about randomisation is a statement that the patients were randomly divided into experimental or control groups. There is no reference to concealment of allocation sequence. The review authors judged there to be no imbalance between the two study groups according to baseline characteristics. | Some concerns | Given the nature of an exercise‐based CR intervention, the review authors judged participants were aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. The triallists did not provide any information to help us ascertain whether people may or may not have deviated from the intended intervention. The triallists did not state if any analysis was carried out to estimate the effect of assignment to the intervention. The review authors judged that there was no reason to believe that participants were excluded from the analysis or assessed in the wrong intervention group. | High risk of bias | The triallists reported that only participants who did not drop out and were not lost to follow‐up were included in analysis. The triallists state that they used available case analysis. Even though the outcome was mortality, which is registered centrally, because the trial only included participants who were not lost to follow‐up, this might have affected the number of deaths recorded in this trial. The triallists did not provide enough information to allow us to assess further whether or not the missingness depended on its true value. | Low risk of bias | Mortality was reported from a central database; because of the nature of the outcome and due to it being reported centrally, measurement or ascertainment could not have differed between the two groups. The triallists did not provide any information with regard to blinding of the outcome assessors/assessment. The review authors judged that knowledge of the assigned intervention was unlikely to influence all‐cause mortality as this was an observer‐reported outcome that did not involve judgement. | Low risk of bias | Although there was no protocol published for this study and the triallists made no mention of whether there was a prespecified analysis plan. Furthermore, all‐cause mortality was centrally recorded and the review authors judged that there can not have been bias in selection of this outcome. | High risk of bias | At least one risk of bias domain judged as 'High'. |
| Pressler 2016 | Low risk of bias | The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). | Low risk of bias | Given the nature of an exercise‐based cardiac rehabilitation intervention, participants were aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not specifically state if there were any deviations from the intended intervention, the flow of participants through the study (Figure 1 of 2016 paper and Figure 1 of 2018 paper) showed that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. dropout due to unrelated causes) or consistent with what could occur outside the trial context (i.e. dropout due to unwillingness to continue participation). Triallists tested their data for their distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (standard deviation). Baseline differences between groups were analysed using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed by the triallists. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as a dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fitted to the data by the triallists. |
Low risk of bias | Triallists presented outcome data for all participants. | Low risk of bias | Mortality was reported from a central database; because of the nature of the outcome and due to it being reported centrally, the review authors judged that measurement or ascertainment could not have differed between the two groups. The triallists stated that "All endpoints were performed and supervised by experienced staff blinded to original group assignments". | Low risk of bias | The trial was registered on clinicaltrials.gov before start of the study (NCT01935297) with researchers’ prespecified intentions available in sufficient detail. As all‐cause mortality was centrally recorded, the the review authors that there cannot have been selection bias of this outcome. The triallists provided results for all time points at which this outcome was assessed, with all extra analysis intended as exploratory analysis to provide additional information on the group‐specific impact of the intervention. The triallists reported all results. | Low risk of bias | All risk of bias domains have judgement of low risk of bias. |