Published online Nov 15, 2022. doi: 10.4239/wjd.v13.i11.986
Peer-review started: June 13, 2022
First decision: August 1, 2022
Revised: August 20, 2022
Accepted: October 27, 2022
Article in press: October 27, 2022
Published online: November 15, 2022
Processing time: 151 Days and 1.9 Hours
Yunnan province has a high prevalence of diabetic retinopathy (DR). Accordingly, it is of great significance to explore the DR-related factors and to construct an economic and intuitive clinical prediction model.
The research motivation is early intervention using the DR-related risk factors from the perspective of a predictive model to reduce the prevalence of DR in patients with type 2 diabetes mellitus (T2DM).
The research intends to establish a prediction model that allows clinically early prevention and treatment of DR.
A total of 1654 Han population with T2DM were recruited in this study and were grouped in the without DR and DR groups. The DR group was further subgrouped according to the severity of DR. Then, univariate analysis, logistic regression analysis, and clinical decision tree models of clinical data were performed.
Based on the decision tree model constructed in this study, DR classification outcomes were obtained by evaluating diabetes duration followed by stages of chronic kidney disease, supine systolic blood pressure (SBP), standing SBP, and body mass index.
Personalized interventions for DR-related risk factors based on a decision tree model may potentially reduce the prevalence of DR.
In this study, patients with T2DM in Western China were taken as samples to analyze the influencing factors of DR and build a clinical prediction model. In the future, it is hoped that the prediction model can produce certain social and economic benefits in clinical practice. In addition, when comparing with other clinical studies on DR, we found some controversies, such as the impact of sex and body mass index on DR, which opened up a new direction for future research.