U-shaped association between BMI and cognitive impairment in middle-aged and older adults with type 2 diabetes: effect modification by lifestyle and exercise.

Objective: To explore the correlation between body mass index (BMI) and cognitive impairment in type 2 diabetes patients through a cross-sectional observational study. Methods: Data on basic information and cognitive impairment of type 2 diabetes were collected through questionnaires, and the correlation between BMI and cognitive impairment of type 2 diabetes was analyzed using logistic regression model, restricted cubic spline (RCS) model and subgroup analysis. At the same time, the interaction between BMI and exercise, living status and other factors was tested. Results: A total of 565 valid samples were included in this study and 51.15% had cognitive impairment. The mean BMI score was 24.51 ± 2.16 kg/m2. An inverse association between BMI score and cognitive impairment in patients with was observed in all three models. Subsequent regression analysis using RCS confirmed this nonlinear association and found two inflection points at 23.72 kg/m2 and 27.77 kg/m2. Specifically, cognitive impairment increased with decreasing BMI at BMI scores <23.72 kg/m2, was least expressed in the interval 23.72-27.77 kg/m2, and increased with increasing BMI scores >27.77 kg/m2. In addition, the interaction between BMI and factors such as exercise and lifestyle was examined, and the results showed that the interaction did not reach the statistical significance level. Conclusion: Observations indicate that the U-shaped relationship between cognitive impairment and BMI observed in middle-aged and older adults with type 2 diabetes was more pronounced in those who live alone and are physically inactive. Although the interaction test was not significant, the subgroup analysis suggested that middle-aged and older adults with type 2 diabetes who live alone and are physically inactive may need to manage their BMI more rigorously.