TABLE 4.

GRADE evidence profile.

Outcome (no. of studies)	Quality assessment					No. Of patients		RR/SMD (95% CI)	Certainty	Importance
	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	XYS combination therapy	WM
TCE (6 RCTs)	Serious a	Not serious	Not serious	Not serious	None	302/346 (87.3%)	235/318 (73.9%)	RR = 1.17 (1.09, 1.27)	⊕⊕⊕○	CRITICAL
									Moderate
DSS (4 RCTs)	Serious a	Not serious	Not serious	Not serious	None	262	239	SMD = −0.72 (−0.90, −0.53)	⊕⊕⊕○	CRITICAL
									Moderate
Abdominal distension (2 RCTs)	Serious a	Serious b	Not serious	Not serious	None	87	68	SMD = −1.15 (−1.97, −0.33)	⊕⊕○○	CRITICAL
									Low
Upper abdominal pain (2 RCTs)	Serious a	Not serious	Not serious	Not serious	None	81	60	SMD = −0.98 (−1.62, −0.33)	⊕⊕⊕○	CRITICAL
									Moderate
Belching (2 RCTs)	Serious a	Serious b	Not serious	Serious c	None	96	76	SMD = −0.33 (−1.06, 0.39)	⊕○○○	CRITICAL
									Very low
Early satiety (2 RCTs)	Serious a	Serious b	Not serious	Serious c	None	96	76	SMD = −1.47 (−3.70, 0.75)	⊕○○○	CRITICAL
									Very low
Adverse events (2 RCTs)	Serious a	Not serious	Not serious	Not serious	Strong association d	4/96 (4.2%)	15/76 (19.7%)	RR = 0.20 (0.07, 0.63)	⊕⊕⊕⊕	CRITICAL
									High

CI, confidence interval; DSS, Dyspepsia-related symptom score; RCT, randomized controlled trial; RR, risk ratio; SMD, standardized mean difference; TCE, total clinical efficacy rate; WM, western medicine; XYS, Xiaoyao-san.

^a Most studies had an unclear risk of selection and detection biases. Performance bias was high in the trials. Therefore, the evidence was downgraded by one level

^b The results were inconsistent across studies (I² > 75%). Therefore, the evidence was downgraded by one level.

^c The 95% confidence interval overlapped with no effect. Therefore, the evidence was downgraded by one level.

^d The effect was large (RR < 0.5). Therefore, the evidence was upgraded by one level.