How long-term changes in general and abdominal obesity indices in prediabetic subjects predict the incidence of diabetes in future: results from a 16-year prospective cohort study among first-degree relatives of type 2 diabetic patients

Background: Obesity indices abnormality pervasively is associated with the risk of type 2 diabetes mellitus (T2DM). This study aimed to identify the patterns of changes in obesity indices over time in prediabetic subjects and to classify these subjects as either having a low, moderate, and high risk for developing diabetes in future. Methods: This prospective 16-year cohort study was conducted among 1228 prediabetic subjects. The patterns of changes in general and abdominal obesity indices based on three measurements including first, mean values during the follow-up period and last visit from these indices were evaluated by using the latent Markov model (LMM). Results: The mean (standard deviation) age of subjects was 44.0 (6.8) years and 73.6% of them were female. The LMM identified three latent states of subjects in terms of change in all anthropometric indices: a low, moderate, and high tendency to progress diabetes with the state sizes (29%, 45%, and 26%), respectively. LMM showed that the probability of transitioning from a low to a moderate tendency to progress diabetes was higher than the other transition probabilities. Conclusions: Based on a long-term evaluation of patterns of changes in general and abdominal obesity indices, we classified prediabetic subjects into three groups (low, moderate, and high risk to progress diabetes in future). Also, the method used enabled us to estimate the transition probabilities from low- to moderate and to high-risk states and vice versa. Our results reemphasized the values of all five obesity indices in clinical settings for identifying prediabetic subjects as a high-risk people for progressing diabetes and the need for more effective obesity prevention strategies.