Mathematics education for students with and at-risk for a disability is important and high-quality, research-supported practices should be used for online teaching--whether intentional or as a result of a global pandemic. This single-case design multiple probe replicated across participants study explored the online delivery of an intervention package consisting of the virtual-abstract instructional sequence--taught via modified explicit instruction--and the system of least prompts to three upper elementary students with a disability or at-risk in solving equivalent fractions. Researchers determined a functional relation existed between the intervention package and student accuracy. Researchers also found students were independent and able to maintain accuracy when instruction did not proceed either following the intervention or with the support of boost sessions. Implications for providing mathematical interventions to students with disabilities or at-risk online exist. Further, the study lends support to virtual manipulative-based instructional sequences, as the setting, population, and implementation of explicit instruction differed in this study as compared to previous research.