Table.

Multivariable meta-regression analysis of bias in 89 placebo controlled trials of homoeopathy

Study characteristic	Ratio of odds ratios* (95% CI)	P value
Asymmetry coefficient†	0.20 (0.11 to 0.37)	<0.001
Language (non-English versus English)	0.73 (0.55 to 0.98)	0.038
Study quality:
Allocation concealment (not adequate versus adequate)	0.98 (0.73 to 1.30)	0.87
Blinding (not double blind versus double blind)	0.35 (0.20 to 0.60)	<0.001
Handling of withdrawals (not adequate versus adequate)	1.10 (0.80 to 1.51)	0.56
Publication type (not indexed by Medline versus indexed by Medline)	0.91 (0.67 to 1.25)	0.57

^*Odds ratio with characteristic divided by odds ratio without characteristic. Ratios below 1 correspond to smaller treatment odds ratio for trials with characteristic and hence larger apparent benefit of homoeopathic treatment. 

^†Ratio of odds ratios per unit increase in standard error of log odds ratio, which measures funnel plot asymmetry.