Purpose: The purpose of this study is to evaluate the evidence for direct and indirect interventions for communication in people with moderate–severe dementia. Method: A systematic search of the literature was conducted, as per the Preferred Reporting Items for Systematic Reviews and Meta-Analysed guidelines, across 8 electronic databases. Studies were included if they included direct or indirect interventions, which could be administered by a speech-language pathologist to people with moderate–severe dementia (defined as having Mini-Mental State Examination of ≤ 15; Folstein, Folstein, & McHugh, 1975). Studies also were required to include outcome measures, which reported on communication function or participation and/or well-being related to communication. Included studies were evaluated for methodological quality using the McMaster critical appraisal tool (Law et al., 1998). Results: Eleven studies met the inclusion criteria. Ten of these studies related to direct interventions and included cognitive stimulation approaches using group (n = 5) or individual therapy (n = 1); cognitive training, including naming therapy (n = 1) and spaced retrieval training (n = 1); and cognitive rehabilitation approaches using augmentative and alternative communication (n = 2). One study reported an indirect intervention: conversation partner training. Due to the heterogeneity of studies, a meta-analysis was unable to be conducted. A descriptive synthesis of results indicated that interventions generally resulted in positive changes to communication and related quality-of-life outcomes compared with baseline or control groups. Conclusions: Preliminary evidence was found to support communication interventions for people with moderate–severe dementia. The use of cognitive stimulation approaches, which use a group treatment model and conversation, as a therapy medium show promise as direct intervention options. Implications for clinical practice for speech-language pathologists and future research are discussed.