Utilization of Electronic Health Records for the Assessment of Adiponectin Receptor Autoantibodies during the Progression of Cardio-metabolic Comorbidities
Pugia, M. J.; Pradhan, M.; Qi, R.; Eastes, D. L.; Geisinger, A.; Mills, B. J.; Baird, Z.; Wijeratne, A.; McAhren, S. M.; Mosley, A. L.; Shekhar, A.; Robertson, D. H.
Show abstract
BACKGROUNDDiabetes is a complex, multi-symptomatic disease that drives healthcare costs through its complications as the prevalence of this disease grows rapidly world-wide. Real-world electronic health records (EHRs) coupled with patient biospecimens, biological understanding, and technologies can lead to identification of new diagnostic markers. METHODSWe analyzed the 20-year EHRs of 1862 participants with midpoint samples (10-year) in an observational study of type 2 diabetes and cardiovascular arterial disease (CVAD) conducted by the Fairbanks Institute to test the diagnostic biomarkers. Participants were assigned to four cohorts (healthy, diabetes, CVAD, CVAD+diabetes) based on EHR data analysis. The immunoassay reference range for circulating autoantibodies against the C-terminal fragment of adiponectin receptor 1 (IgG-CTF) was determined and used to predict outcomes post-sample. RESULTSThe IgG-CTF reference range was determined [75-821 ng/mL] and out-of-range values of IgG-CTF values predicted increased likelihood of additional comorbidities and mortality determined from the EHRs 10 years after sample collection. The probability of mortality was lower in patients with elevated IgG-CTF >821 ng/mL [OR 0.49-0.0] and higher in patients with lowered IgG-CTF <75 ng/mL [OR 3.74-9.64]. Although many patients at the time of sample collection had other conditions (hypertension, hyperlipidemia, or elevated uristatin values), only hypertension correlated with increased likelihood of mortality (OR 4.36-5.34). CONCLUSIONSThis study confirms that retrospective analysis of biorepositories coupled with EHRs can provide insight into novel diagnostic markers and the IgG-CTF marker can predict the likelihood of progressing to additional comorbidities or mortality.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Covid-19 fatality prediction in people with diabetes and prediabetes using a simple score at hospital admission 94%
- Heterogeneity of Treatment Effects Across Nine Glucose-Lowering Drug Classes in Type 2 Diabetes: Extension of the LEGEND-T2DM Network Study 93%
- Metabolomic Risk Predictors of Diabetic Foot Complications: a longitudinal observational study 92%
Similar papers in this journal
- Estimating Heritability of Glycaemic Response to Metformin using Nationwide Electronic Health Records and Population-Sized Pedigree 92%
- Precision Prognostics for Cardiovascular Disease in Type 2 Diabetes: A Systematic Review and Meta-analysis 92%
- Systematic review of precision subclassification of type 2 diabetes 91%
Similar papers in this journal
- Vitamin B12 and Risk of Diabetes: New insight from Cross-Sectional and Longitudinal Analyses of the China Stroke Primary Prevention Trial (CSPPT) 93%
- A plasma metabolite score of three eicosanoids predicts incident type 2 diabetes – a prospective study in three independent cohorts 93%
- Predictors of Arterial Stiffness in Adolescents and Adults with Type 1 Diabetes: A Cross-Sectional Study 93%
Similar papers in this journal
- Factors associated with therapeutic inertia in individuals with type 2 diabetes mellitus started on basal insulin 93%
- Comorbid Diabetes Mellitus was Associated with Poorer Prognosis in Patients with COVID-19: A Retrospective Cohort Study 92%
- Data-driven subtypes of type 2 diabetes mellitus and risk of dementia, stroke, and brain structural changes in the UK Biobank 90%
Similar papers in this journal
- Replication and cross-validation of T2D subtypes based on clinical variables: an IMI-RHAPSODY study 93%
- Development and validation of a Trans-Ancestry polygenic risk score for Type 1 Diabetes 93%
- Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care setting 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.