Is it efficacy or just improvement? A re-evaluation of the meta-analysis presented by Zhang et al