Abstract: Objective To explore the prevalence of osteoporosis and its correlative factors in adults over 65 in a community in Shanghai.Methods A total of 594 elderly people were investigated by filling out questionnaires including their basic information, daily behavior habits, etc. Meanwhile, their lower limb myodynamia, and calcaneus ultrasonic bone mineral density were measured. Then, the questionnaires of 297 elderly people from the total were drawn randomly for analysis of correlative factors for osteoporosis by using multiple logistic regression analysis, which was also used to inspect the accuracy of equation prediction.Results The prevalence of osteoporosis was shown to be 60.6% in residents aged 65 or older.The multiple logistic regression analysis showed that gender, age, body mass index (BMI), intake of dairy products, activities of daily life, physical exercise and lower limb myodynamia were the main correlative factors of senile osteoporosis, and the standardized regression coefficient of the gender was found to be greater than others (B=0.300). Besides, the area under ROC curve was 0.896, 95% CI was 0.86~0.93. The model has sensitivity and specificity of 88.3% and 77.1% respectively, with Younden index being 0.654 and prediction accuracy 82.8%.Conclusion Gender, age, body mass index, dairy consumption, activities of daily life and physical exercise, and lower limb myodynamia can be used as judge indexes of osteoporosis, and the equation based on the indexes has a fine prediction effect, which possesses a certain reference value for osteoporosis prevention.