Background: Evidence on the prevalence, determinants, and health outcomes of physical activity in disabled people is limited. It is possible that the limited availability of high-quality scientific evidence is due to the extent and nature of disability assessment in physical activity research. This scoping review explores how disability has been measured in epidemiological studies that included accelerometer-based measurement of physical activity. Methods: Data sources: MEDLINE (Ovid), Embase, PsychINFO, Health Management Information Consortium, Web of Science, SPORTDiscus, and CINAHL. Eligibility criteria: Prospective and cross-sectional studies that included an accelerometer measurement of physical activity. Survey instruments used in these studies were obtained, and questions relating to the International Classification of Functioning, Disability and Health domains of (1) health conditions, (2) body functions and structures, and (3) activities and participation, were extracted for analysis. Results: Eighty-four studies met the inclusion criteria, from which complete information on the 3 domains was obtained for 68. Seventy-five percent of studies (n = 51) captured whether a person had at least one health condition, 63% (n = 43) had questions related to body functions and structures, and 75% (n = 51) included questions related to activities and participation. Conclusion: While most studies asked something about one of the 3 domains, there was substantial diversity in the focus and style of questions. This diversity indicates a lack of consensus on how these concepts should be assessed, with implications for the comparability of evidence across studies and subsequent understanding of the relationships between disability, physical activity, and health.