This systematic review and meta-analysis appraised and synthesized mathematics interventions research addressing rational number concepts for students with mathematics difficulties (MD) in Grades 3-9. Included studies (n = 28) meet the What Works Clearinghouse Group Design standards (Version 4.0) for study quality. Across all studies, we found significant mean effects (g = 0.68) favoring intervention. Understanding which instructional components within interventions are associated with effects was a fundamental purpose of the study. Random-effects meta-regression models with robust variance estimation (RVE) were used to examine which instructional components moderated effects. The teaching and use of accurate mathematical language emerged as significant (b = 0.50, p = 0.042) when controlling for other instructional components (visual representations, review, student explanations, and fluency building activities) and three control variables (group size, duration, nature of the control). Intervention characteristics and study design features were also explored through moderator analysis. Group size, duration, nature of the control, type of measure, interventionist, provision of ongoing training, and grade level emerged as significant moderators (p < 0.05). [For a previous version of this report, see ED612822.]