In this paper, we investigate the global threshold dynamics of a stochastic SIS epidemic model incorporating media coverage. We give the basic reproduction number R s 0 and establish a global threshold theorem by Feller's test: if R s 0 ≤ 1, the disease will die out a.s.; if R s 0 > 1, the disease will persist a.s. In the case of R s 0 > 1, we prove the existence, uniqueness, and global asymptotic stability of the invariant density of the Fokker-Planck equations associated with the stochastic model. Via numerical simulations, we find that the average extinction time decreases with the increase of noise intensity σ , and also find that the increasing σ will be beneficial to control the disease spread. Thus, in order to control the spread of the disease, we must increase the intensity of noise σ .