OBJECTIVES: (1) To appraise and synthesize the literature on dance interventions for individuals with Parkinson disease (PD); (2) to provide information regarding the frequency, intensity, duration, and type of dance used in these programs; and (3) to inform the development of future studies evaluating dance interventions in this population. DATA SOURCES: Eight databases (MEDLINE, Cumulative Index to Nursing and Allied Health Literature [CINAHL], the Allied and Complementary Medicine Database [AMED], SPORTDiscus, PubMed, PubMed Central, Sage, and ScienceDirect) were electronically searched in April 2014. The references lists from the included articles were also searched. STUDY SELECTION: Studies retrieved during the literature search were reviewed by 2 reviewers independently. Suitable articles were identified by applying inclusion criteria. DATA EXTRACTION: Data regarding participants and the frequency, intensity, duration, and type of dance form used were extracted. The effect that each dance program had on defined outcomes and the feasibility of each program were also reviewed. DATA SYNTHESIS: Thirteen articles were identified. The quality of studies varied, and methodological limitations were evident in some. The evidence evaluated suggests that two 1-hour dance classes per week over 10 to 13 weeks may have beneficial effects on endurance, motor impairment, and balance. CONCLUSIONS: Dance may be helpful for some people with PD. This article provides preliminary information to aid clinicians when implementing dance programs for people with PD. Higher-quality multicenter studies are needed to determine the effect of other dance genres and the optimal therapy volume and intensity.