Latent class analysis suggests four classes of persons with type 2 diabetes mellitus based on complications and comorbidities in Tianjin, China: a cross-sectional analysis

Abstract

The aim of this study was to explore a new classification way in persons with type 2 diabetes mellitus based on complications and comorbidities using Latent Class Analysis, moreover, finding out the factors associated with different latent classes and making specific suggestions. In this study, 5,500 patients with type 2 diabetes mellitus from ten hospitals in Tianjin, China were selected, and the response rate was 96.2%. Latent Class Analysis was used to cluster patients. After compared the baseline characteristics, multinomial logistic regression was applied. Patients with type 2 diabetes mellitus were classified into four classes. In the univariate analysis, all variables were significant (p<0.05). According to multinomial logistic regression, we found longer duration of type 2 diabetes mellitus, family history of diabetes, older age, obesity and central obesity, female menopause, living in a suburb, having a higher 2hPG at diagnosis, smoking and drinking were associated with the prevalence of complications and comorbidities. In conclusion, LCA was shown to be an effective method for grouping patients with T2DM, which presented a nuanced approach to data reduction. Further research using LCA may be especially useful to investigate causal relationships between complications and the significant factors identified in our study.