Sensitivity analysis and publication bias
Sensitivity analysis indicated that the removal of any of the studies
from the meta-analysis, create no change in the results of the
meta-analysis on leptin concentration whereas, the results on
adiponectin concentration were sensitive to omitting 1 study18. Funnel plots for
adiponectin and leptin were visually symmetrical (Figure 4 and
5) , and the results of Begg’s test did not determine any evidence of
publication bias in studies that examine the effect of soy protein
consumption on adiponectin (Begg’s test, P = 0.059) and on leptin
(Begg’s test, P = 0.861).