3.3.3. Sensitivity analysis and publication bias
Sensitivity analysis showed that the omission of any of the studies from
the meta-analysis, create no change in the outcomes of the meta-analysis
on testosterone concentration whereas, the results on cortisol
concentration were sensitive to omitting 3 studies (7, 29, 30). Funnel
plots for cortisol and testosterone were visually symmetrical, and the
results of Egger’s test did not determine any evidence of publication
bias in studies that examine the effect of HMB consumption on cortisol
(Egger’s test, P = 0.337) and on testosterone (Egger’s test, P = 0.140).