The uncertainty of public health emergencies leads to the diffusion of rumors on social media, and the extensive forwarding of rumor-refuting information has become a way of rumor management. Determining the characteristics of rumor-refuting information publishers that are conducive to rumor information being forwarded has become a concern. This article crawls 3,035 rumor-refuting information data about the COVID-19 event. Based on social capital theory, from the perspective of rumor-refuting information publishers, the characteristic variables of social capital of publishers are selected from the three dimensions of structure, relationship, and cognition to construct the model of influencing factors of information-forwarding behavior of rumor-refuting information, and the negative binomial regression model is used for data analysis and robustness test. Results show that the number of followers, disclosure of geographical location information, gaining membership, and obtaining Weibo professional certification will increase the number of forwarding of the rumor-refuting information, and whether or not having a personal introduction will not significantly affect the forwarding of rumor-refuting information. At the same time, the number of Weibo and fans will reduce the retransmission of the rumor-refuting information. Research results of this paper can help the platform make targeted choices and encourage users to release rumor-refuting information, expand the scope of rumor-refuting information dissemination, and effectively achieve rumor governance.