Table 3.
Isometric exercise training and resting blood pressure: author panel consensus on the certainty rating of evidence
| GRADE domain | Judgement | Concerns about certain domains | |||
|---|---|---|---|---|---|
| Methodological limitations of the studies | Many of the studies, despite randomisation, provided unclear information regarding allocation concealment. Blinding of outcome assessors was not reported in most of the studies. However, this limitation was mitigated using automatic devices for blood pressure assessments. Some studies failed to adequately account for all patients in the analysis. Selective outcome reporting and other limitations were not disclosed. Approximately 30% of the studies exhibited a high risk of bias | Serious | |||
| Indirectness | Assessment of blood pressure was performed directly using validated methods. We judged there is no evidence of indirectness | Not suspected | |||
| Imprecision | The sample size of all studies was ~ 1400 subjects. Significant reductions in systolic and diastolic blood pressure were reported in most trials with different sample sizes | Not suspected | |||
| Inconsistency | There were no variations in the direction of the estimated effect. The mean magnitude of effect for systolic blood pressure was − 8.1 mmHg (95% CI − 6.5 to − 9.7 mmHg) and for diastolic blood pressure was − 3.7 mmHg (95% CI − 2.9 to − 4.5 mmHg). The inter-study variability in the magnitude of effect is considered primarily attributable to protocol and population differences | Not suspected | |||
| Publication bias | Although unclear, we did not strongly suspect publication bias because most registered trials were published | Not suspected | |||
|
Outcome Resting systolic and diastolic blood pressure |
Effect Most studies show reductions in both systolic and diastolic blood pressure |
Number of participants/studies 1424 subjects (28 randomised controlled trials) |
Certainty in the evidence MODERATEa ⊕ ⊕ ⊕ O (due to serious risk of bias) |
||
aSerious risk of bias across studies because of unclear or inadequate allocation concealment, blinding and adequately accounting for all patients in the analysis