Considering Race and Ethnicity in the Management of Bone Health C ommunities of color have borne a disproportionate burden of infections, hospitalizations, and deaths throughout the global coronavirus disease 2019 (COVID-19) pandemic. (1) Along with a rising tide of anti-Asian xenophobia and social injustice protests, events of the past year and a half have triggered a reckoning on racism in the United States. The field of medicine is facing up to its own unfortunate history of racism, prompting hospitals and organizations to examine their own policies and practices for evidence of bias. (2) In this context the use of race and ethnicity in clinical decisionmaking is under scrutiny. (3) Health outcomes are influenced by myriad factors including genetics, health behaviors, socioeconomic status, and access to health care. Although the interactions of these factors are complex, it is clear that social determinants of health, the conditions in the environments in which individuals live, learn, and work, are important drivers of health outcomes. (4) The inclusion of race and ethnicity in clinical algorithms is beset by several challenges. Although a common implicit assumption is that race is a proxy for genetic or biological differences, race is a social construct and racial differences in health outcomes may instead reflect the effects of systemic racism. In addition, the use of broad racial and ethnic categories as standardized by the Office of Management and Budget aggregates heterogeneous subgroups and can mask substantial disparities within these communities.What is the role of race and ethnicity in the management of bone health? A commonly used clinical calculator, the Fracture Risk Assessment Tool (FRAX; https://www.sheffield.ac.uk/FRAX/) provides the 10-year probabilities of hip fracture or major osteoporotic fracture based on clinical risk factors and optional femoral neck bone mineral density (BMD). (5) Using algorithms developed from population-based cohorts in Europe, North America, Asia, and Australia, FRAX provides 73 country-specific models plus ethnicity-specific models in the United States, South Africa, and Singapore. (6) In the United States, race/ethnicity-specific probabilities are offered for white, Hispanic, black, and Asian adults, based on data demonstrating differing fracture rates between these groups despite similar BMD. (7) The FRAX algorithm returns lower fracture risk estimates for black, Hispanic, and Asian women, leading to concerns for the potential to delay osteoporosis therapy. (3) But given differences in the incidence atypical femur fractures, a rare complication of long-term bisphosphonate use, between racial and ethnic groups, lower risk estimates might appropriately prevent overtreatment. (8)