A project to identify relevant clinical factors associated with diabetic peripheral neuropathy using the All of Us Research data.
Title: Risk Factors for Diabetic Peripheral Neuropathy in a Large, Diverse U.S. Cohort from the All of Us Research Program
Christianah Jemiyo, Brett A. McGregor, and Junguk Hur
Department of Biomedical Sciences, University of North Dakota School of Medicine and Health Sciences, Grand Forks, North Dakota
Diabetic peripheral neuropathy (DPN) is a common and debilitating complication of diabetes, yet early identification remains challenging. Prior studies have primarily relied on homogeneous populations and a limited range of clinical variables, limiting their generalizability. This study uses the NIH All of Us dataset to evaluate cross-sectional associations between DPN and clinical and laboratory features in a large, diverse cohort. We analyzed 58,776 adults with diabetes, excluding those with prediabetes, and harmonized measurements with each participant’s diabetes diagnosis date. Elastic Net logistic regression was used to identify key variables associated with DPN, followed by multivariable logistic regression to estimate adjusted associations. Seven variables were retained in the penalized model: hematocrit, albumin, alkaline phosphatase, respiratory rate, diabetes duration, Middle Eastern/North African (ME/NA) race, and vitamin D deficiency. In the multivariable analysis, lower serum albumin and vitamin D deficiency demonstrated the strongest associations with DPN. Common diabetes markers like HbA1c and glucose were excluded in the penalized model but remained significant in full models. These findings reinforce associations between known and emerging risk factors and demonstrate the value of diverse, population-scale data for exploring patterns related to diabetic complications.
Created by Junguk Hur