Work

Using metabolic syndrome severity in the “real world”: Associations with diabetes and cardiovascular disease in all of us vs. NHANES

Year: 2025

Type: article

Abstract: While the severity of the metabolic syndrome (MetS) is associated with type 2 diabetes and cardiovascular disease (CVD) in research cohorts, it is unclear if this association remains when using clinically-collected data, which is more subject to error and bias. We used a previously-validated MetS-severity z-score (MetS-Z) to compare the degree of cardiometabolic derangement between clinically-collected data from the electronic health-record (EHR) in the All-of-Us cohort (N=101,676) with research-collected data from NHANES 2017-2020 (n=3470). We assessed (separately) the odds of current diabetes and CVD in each cohort based on MetS z-score, adjusted for sex, race/ethnicity, education and income. Mean MetS-Z-scores in All of Us (vs. NHANES) were higher overall (0.41 vs. 0.15), including being slightly higher among those without diabetes (0.03 vs. -0.09) and similar among those with diagnosed diabetes (1.42 vs. 1.50). In All of Us (vs. NHANES) MetS-Z was higher among those without CVD (0.36 vs. 0.10) but similar among those with CVD (0.72 vs. 0.66). Adjusted odds of diagnosed diabetes and CVD based on MetS-Z remained significant when using clinically-collected data in All of Us (diabetes: 3.41 [95% confidence interval 3.34, 3.49]; CVD: 1.20 [1.18, 1.21]), though not as high as in NHANES (diabetes: 5.83 [4.52, 7.51]; CVD: 1.43 [1.27, 1.61]). While EHR data is limited by selection bias and data accuracy concerns (e.g. lack of fasting laboratory testing), we found overall similar relationships between MetS-severity, diabetes and CVD in EHR-based and research-based cohorts, supporting utility of these data in risk algorithms. • Metabolic syndrome (MetS) severity is linked to diabetes & CVD in research cohorts. • This has not been studied using clinically-collected electronic health record data. • We evaluated MetS severity risks from EHR data in the All of Us cohort vs. NHANES. • Despite limitations of EHR data, MetS was linked to diabetes & CVD in All of Us. • This supports clinically-collected data as valid for EHR-based risk calculators.

Source: Metabolism Open

DOI: 10.1016/j.metop.2025.100415

Authors: Mark D. DeBoer, Matthew J. Gurka, Marieke K. Jones, Mark E. Smolkin

Institution: University of Virginia

Language: English

Topic: Diabetes, Cardiovascular Risks, and Lipoproteins

Subfield: Endocrinology, Diabetes and Metabolism

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,120