Deep exploration and precise identification of key risk factors for diabetic peripheral neuropathy using the random...

Area of Science:

• Endocrinology and Metabolism
• Neurology
• Biostatistics

Background:

• Diabetic peripheral neuropathy (DPN) is a common and debilitating complication of diabetes mellitus, increasing disability and mortality rates.
• Early intervention and effective management are crucial for reducing the risk and progression of DPN.

Purpose of the Study:

• To develop and validate a risk prediction model for DPN in hospitalized patients with type 2 diabetes mellitus (T2DM).
• To identify key risk factors associated with DPN for improved clinical risk assessment and personalized management strategies.

Main Methods:

• Retrospective cohort study of 1,004 hospitalized T2DM patients.
• Development of a DPN risk prediction model using the Random Forest (RF) algorithm.
• Logistic regression analysis to identify significant risk factors for DPN.

Main Results:

• Five significant risk factors for DPN were identified: increasing age, longer diabetes duration, elevated HbA1c, and higher GSP levels.
• Lower serum 25(OH)D3 levels were found to be protective against DPN.
• The RF-based DPN risk prediction model demonstrated excellent discriminatory performance with an AUC of 0.829.

Conclusions:

• The developed Random Forest model effectively predicts DPN risk by incorporating key clinical and biochemical factors.
• This model provides a valuable tool for personalized DPN prevention and management strategies in T2DM patients.
• The findings support the integration of this risk assessment tool into clinical practice for proactive diabetes care.