Profiling Inflammatory Biomarkers following Curcumin Supplementation: An Umbrella Meta-Analysis of Randomized Clinical Trials

Table 4

Assessing the quality of evidence using the GRADE approach.

Outcome measures	Summary of findings		Quality of evidence assessment (GRADE)
	No. of patients (meta-analysis)	Effect (95% CI)	Risk of bias^a	Inconsistency^b	Indirectness^c	Imprecision^d	Publication bias^e	Quality of evidence^f
Inflammatory biomarkers
CRP	3,271 (7)	−0.74 (−1.11, −0.37)	Not serious	Not serious	Serious	Not serious	Not serious	Moderate
TNF-α	3,224 (6)	−1.92 (−2.64, −1.19)	Not serious	Not serious	Serious	Not serious	Not serious	Moderate
IL-6	2,972 (6)	−1.07 (−1.71, −0.44)	Not serious	Not serious	Serious	Not serious	Not serious	Moderate

CRP: C-reactive protein; TNF: tumor necrosis factor; IL-6: interleukin-6. (a) Risk of bias based on the according to AMSTAR results. (b) Downgraded if there was a substantial unexplained heterogeneity (I² > 50%, ) that was unexplained by meta-regression or subgroup analyses. (c) Downgraded if there were factors present relating to the participants, interventions, or outcomes that limited the generalizability of the results. Participants of the included studies were from different health conditions. (d) Downgraded if the 95% confidence interval (95% CI) crossed the minimally important difference (MID) for benefit or harm. MIDs used for each outcome were 3.16 mg/l for CRP, 7.9 pg/ml for TNF-α, and 2 pg/ml for IL-6. (e) Downgraded if there was evidence of publication bias using funnel plot. (f) Since all included studies were meta-analysis of randomized controlled trials, the certainty of the evidence was graded as high for all outcomes by default, and then downgraded based on prespecified criteria. Quality was graded as high, moderate, low, or very low.