Objective@#To predict the risk of cognitive decline among elderly patients with type 2 diabetes mellitus (T2DM), so as to provide insights into the early identification of cognitive decline in elderly patients with T2DM.@*Methods@#The elderly patients with T2DM hospitalized in the Qingpu Branch of Zhongshan Hospital Affiliated to Fudan University were selected as subjects. General information was collected by a questionnaire survey, including demographics and nutritional status, and cognitive decline was assessed using the frailty Phenotype Scale, the Montreal Cognitive Assessment and the Clinical Dementia Rating. Multivariable logistic regression model was used to analyze the influencing factors of cognitive decline among elderly patients with T2DM. A nomogram was established, and was verified with Bootstrap resampling method and receiver operation characteristic curve.@*Results@#A total of 270 questionnaires were sent out and 262 valid questionnaires were collected, with an effective rate of 97.04%. There were 137 males (52.29%), 146 patients at ages of 60 to 69 years (55.73%), 179 married patients (68.32%), and 85 patients with cognitive decline (32.44%). Multivariable logistic regression analysis identified gender (female, OR=3.118), age (70 to 79 years, OR=3.218; 80 years and older, OR=3.058), educational level (high school or technical secondary school, OR=0.335; junior college, OR=0.130; college and above, OR=0.300), regular exercise (OR=0.083), memory (decreased, OR=29.723), nutritional status (risk of malnutrition, OR=16.307; malnutrition, OR=39.469), activity of daily living (decreased, OR=6.804) and depressive symptoms (OR=8.609) as factors affecting cognitive decline among elderly patients with T2DM (all P<0.05). The nomogram based on the above factors fit well and indicated strong predictive ability.@*Conclusion@#Gender, age, educational level, regular exercise, memory, nutritional status, activity of daily living and depressive symptoms can effectively predict the risk of cognitive decline among elderly patients with T2DM.