Electroacupuncture Is Effective for Peripheral Facial Paralysis: A Meta-Analysis

Table 3

Evidence quality of the results of responding rate.

Results	Test type	Downgrade factor					Upgrade factor			Quality
		DD	In	HE	RI	PB	SE	BSE	DE
EA versus “controls”
Total	RCT	−1^A	0	−1^B	0	0	0	0	0	Low
Embedding	RCT	−1^A	0	0	−2^C	0	0	0	0	Very low
Massage	RCT	−1^A	0	0	−2^C	0	0	0	0	Very low
Needle acupuncture	RCT	−1^A	0	0	0	0	0	0	0	Moderate
Rood technique	RCT	−1^A	0	0	−2^C	0	0	0	0	Very low
Stuck needling	RCT	−1^A	0	0	−2^C	0	0	0	0	Very low
Warm needling	RCT	−1^A	0	−1^B	−1^D	0	0	0	0	Very low
Current	RCT	−1^A	0	0	0	−1^F	0	0	0	Low
Current characteristics
Low frequency	RCT	−1^A	0	0	−1^D	0	0	0	0	Low
Low-high frequency	RCT	−1^A	0	0	0	0	0	0	0	Moderate
High frequency	RCT	−1^A	0	0	−1^D	0	0	0	0	Low
DBC	RCT	−1^A	−1^E	0	0	0	0	0	0	Low
EA as an adjunct	RCT	−1^A	0	0	0	−1^F	0	0	0	Low

Note. DD: design defects. IN: indirectness. HE: heterogeneity. RI: data sparse or incomplete. PB: publication bias. SE: significant effect. BSE: bias subtractive effect. DE: dose effect. DBC: differences between current characteristics. A: study design defects may affect the results; B: cannot ignore heterogeneity; C: very few participants seriously affect the results; D: very few participants may affect the results; E: indirect comparison results; F: publication bias may affect the results.