Background This study aimed to analyse which influencing factors may be more effective to achieve diabetes management targets in the community by the adaptive-lasso logistic regression model.
Methods A cross-sectional study (N=1,127) was adopted to establish the adaptive-lasso logistic regression model of influencing factors for community management based on multi-stage cluster sampling data among patients with diabetes in China. Patient’s fasting blood glucose level, blood pressure, and triglycerides was collected.
Results Overall, 90.6% of included people had a fasting glucose level higher than 6.1mol/L, and 9.4% of them were below 6.1mol/L. By cross-validation, after folding eight times, the variables involved in the adaptive lasso-logistic regression model include age, education level, main source of income, marital status, average monthly income, free medical service, basic medical insurance for residents, hospital history, number of follow-up evaluations by family doctor team, voluntary participation in community blood glucose measurement. The Akaike Information Criterion and Bayesian Information Criterion of adaptive lasso-logistic regression model were 1980 and 2021, which were lower than the full-variable logistic model (2041, 2245) and the ridge logistic model (2043, 2348). The adaptive-lasso logistic regression model was better than the other two models regarding time cost.
Conclusions The adaptive-lasso logistic regression model can analyse the influencing factors of community management in patients with diabetes. Community intervention and intensive management measures can significantly improve the blood glucose status of patients with diabetes.