| Objective To investigate the association between the red blood cell distribution width to albumin ratio (RAR) and osteoporosis (OP) in different age groups at the Xinjiang Uygur Autonomous Region People's Hospital, and to evaluate the predictive value of RAR for OP. Methods Patients who underwent osteoporosis screening were selected from October 26, 2020 to January 10, 2025. Covariates included common demographic characteristics, routine hematological parameters, and standard biochemical indices, with variables included in the analysis if their data completion rate exceeded 30%. Univariable and multivariable logistic regression analyses were performed to explore the relationships between various factors and osteoporosis outcomes, identifying potential influencing variables for further investigation. Multivariable logistic regression models and machine learning algorithms were employed to confirm the statistical significance of RAR in predicting OP after adjusting for other covariates. Finally, restricted cubic spline (RCS) curves were used to determine the cutoff point of RAR and assess its clinical significance. Results In this study, age, sex, alcohol consumption, BMI, white blood cell count, eosinophil percentage, neutrophil percentage, AST/ALT ratio, ALP, HDL-C, and RAR were significantly associated with osteoporosis. Similar findings were observed in the machine learning models. Conclusion RAR was positively correlated with the risk of osteoporosis, with an increased risk of osteoporosis when the RAR valve over 3.18. It is suggested that RAR can be used as a simple and economical and assistant indicator for early risk assessment of osteoporosis. |