Table 2.

Grade quality of evidence of fire-needle moxibustion treatment for KOA.

Outcome	Effect		Number of participants (studies)	Quality of the evidence (GRADE)
	Relative effect (95% CI)	Absolute effect (95% CI)
Recovery rate	RR 1.56 (1.34 to 1.81)	162 more per 1000 (from 98 more to 234 more)	1114 (12 studies)	⊕⊕⊖ ⊖  Low1,2
Recovery and marked-improvement rate	RR 1.50 (1.36 to 1.64)	250 more per 1000 (from 180 more to 319 more)	1051 (11 studies)	⊕⊕⊖  ⊖  Low1,2
Pain		SMD 0.72 SD lower (1.23 lower to 0.22 lower)	390 (6 studies)	⊕⊖⊖  ⊖  Very low1,2,3

¹None of the trials were blinded; most of them did not mention randomization process and allocation concealment.

²Published evidence is limited due to a small number of trials, all of which are showing benefits.

³Confidence intervals with minimal overlap; the heterogeneity is significant.