Table 4.
Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) in included randomized controlled trials
| Outcomes | Participants (studies) | Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | Overall quality of evidence |
|---|---|---|---|---|---|---|---|
| Weight | 1518 (27 studies) | no serious risk of bias | serious1 | serious2 | no serious imprecision | reporting bias strongly suspected3 |
⊕⊕⊝⊝ LOW 1,2,3 due to inconsistency, indirectness, publication bias, dose-response gradient |
| HDL | 1352 (23 studies) | no serious risk of bias | no serious inconsistency1 | no serious indirectness | no serious imprecision | undetected |
⊕⊕⊕⊕ HIGH |
| LDL | 1283 (22 studies) | no serious risk of bias | serious1 | no serious indirectness | no serious imprecision | reporting bias strongly suspected3 |
⊕⊕⊕⊝ LOW 1.3 due to inconsistency |
| TG | 549 (23 studies) | no serious risk of bias | serious1 | no serious indirectness | no serious imprecision | undetected |
⊕⊕⊕⊝ MODERATE 1 due to inconsistency |
| Fasting insulin | 935 (20 studies) | no serious risk of bias | very serious1 | no serious indirectness | no serious imprecision | undetected |
⊕⊕⊝⊝ LOW 1 due to inconsistency |
| Fasting glucose | 1086 (19 studies) | no serious risk of bias | serious1 | no serious indirectness | no serious imprecision | undetected |
⊕⊕⊕⊝ MODERATE 1 due to inconsistency |
1 The effect sizes varied between studies, rated down by one level for high I2
2 There is a statistical difference between the direct comparison group and the final result (P<0.05)
3 Less eating time helps with weight loss
4 Inspection of funnel plot suggest publication bias