Table 2.

Meta-analysis and pooled results of the main outcome in included studies.

Study ID	Adverse reaction	Risk of bias	Follow-up	Random method	Overall effect OR (95% CI)	TCM syndrome effect	Hp OR (95% CI)	Gastroscope OR (95% CI)	Pathology OR (95% CI)	Overall effect P value
Chen and Lai 2013 [7]	Not mentioned	Low	Not mentioned	Not mentioned	3.08 (1.10, 8.62)	OR 6.67 (1.41, 31.59)				0.03
Fu et al. 2013 [8]	Not found	Low	Not mentioned	Random number table	5.21 (1.28, 21.24)	MD −5.85 (−7.71, −3.99)				0.02
Zhang 2013 [9]	Not found	Low	Not mentioned	Random number table	3.58 (0.89, 14.39)	OR 9.75 (1.16, 82.11)	6.02 (1.43, 25.40)	3.78 (1.29, 11.06)	2.83 (1.00, 7.98)	0.07
Shi 2010 [10]	Not mentioned	Low	Not mentioned	Not mentioned	3.86 (1.41, 10.57)					0.009
Li and Xu 2009 [11]	Not mentioned	Low	Not mentioned	Not mentioned	1.64 (0.51, 5.33)					0.41
L. Liu and Y. Liu 2014 [12]	Not mentioned	Low	Not mentioned	Random number table	2.62 (0.99, 6.94)					0.05
Ni et al. 2013 [13]	Not found for intervention; rash and anaphylactoid purpura of legs for 1 case in control group; increase of eosinophil for 1 case	Low	Not mentioned	Not mentioned	2.80 (1.35, 5.82)					0.006
Li 2013 [14]	In intervention group, epigastric pain, fullness, belching, and poor appetite were all less than control group	Low	Not mentioned	Not mentioned	3.62 (0.90, 14.63)					0.07
Tang and Hong 2003 [15]	Most in control group, poor appetite, upper abdominal discomfort, nausea and vomiting, and so forth; especially used metronidazole, must add metoclopramide and anisodamine, and so forth, impacted quality of lives	Low	Not mentioned	Not mentioned	1.58 (0.25, 9.95)					0.63