In obese adults, METS-IR demonstrated a stronger association with comorbidity burden than HOMA-IR, but both indices showed limited discriminatory performance for T2DM and HT and should not be used as standalone tools for treatment guidance or diagnostic classification.
Key Findings
Results
HOMA-IR and METS-IR were only moderately correlated with each other in obese adults.
Spearman correlation coefficient ρ = 0.342, p < 0.001
The moderate correlation suggests the two indices capture partially different aspects of insulin resistance
Study included 481 adults with BMI ≥ 30 kg/m2 across four age strata (18-40, 40-65, 65-75, and ≥75 years)
This was a retrospective, single-center study
Results
METS-IR was significantly associated with comorbidity burden in obese adults, whereas HOMA-IR was not.
METS-IR adjusted odds ratio for comorbidity burden: OR 1.84; 95% CI 1.26-2.69; p < 0.01
HOMA-IR adjusted odds ratio for comorbidity burden: OR 1.12; 95% CI 0.92-1.36; p = 0.256
Comorbidity burden was defined as 0-3 chronic conditions and modeled using ordinal logistic regression
Analysis was per 1-SD increase in z-scores, adjusted for age, sex, and BMI
Results
Both METS-IR and HOMA-IR showed limited discriminatory performance for identifying type 2 diabetes mellitus in obese adults.
AUC for T2DM: HOMA-IR 0.500 vs. METS-IR 0.537
An AUC of 0.500 indicates no discriminatory ability beyond chance
Performance was assessed using receiver operating characteristic analysis
Authors concluded neither index should be used as a standalone diagnostic tool for T2DM
Results
Both METS-IR and HOMA-IR showed limited discriminatory performance for identifying hypertension in obese adults.
AUC for HT: HOMA-IR 0.471 vs. METS-IR 0.519
An AUC below 0.500 for HOMA-IR indicates performance worse than chance for hypertension discrimination
Performance was assessed using receiver operating characteristic analysis
Authors noted isolated use of these indices for hypertension classification 'may be misleading and fail to identify disease in specific patient subgroups'
Results
Despite limited discriminatory performance, METS-IR remained independently associated with both T2DM and hypertension in multivariable logistic regression.
METS-IR association with T2DM: OR 2.51; 95% CI 1.60-3.93; p < 0.001
METS-IR association with HT: OR 1.81; 95% CI 1.13-2.91; p < 0.05
Associations were assessed using multivariable logistic regression
The discrepancy between AUC and OR values highlights that statistical association does not equate to diagnostic utility
Methods
The study compared IR indices across four distinct age strata in obese adults, examining how clinical utility may vary by age.
Age strata were: 18-40, 40-65, 65-75, and ≥75 years
Total sample size was n = 481 obese adults (BMI ≥ 30 kg/m2)
The study was designed to examine whether IR index utility varies by age group
Authors concluded indices should be interpreted 'alongside age, clinical, and biochemical findings'
What This Means
This research studied two common measures of insulin resistance — HOMA-IR (which requires a blood test measuring insulin and glucose) and METS-IR (which uses easily available measurements like BMI, fasting glucose, and triglycerides) — in 481 obese adults across different age groups. The study found that these two measures are only moderately related to each other, suggesting they are not interchangeable. METS-IR was better at predicting overall disease burden (having multiple chronic conditions like diabetes and high blood pressure together) compared to HOMA-IR, making it potentially more useful as a general health risk indicator in obese patients.
However, a critical finding was that neither measure was reliably accurate at distinguishing which individual patients had type 2 diabetes or hypertension. The statistical measure of accuracy (AUC) was close to 0.5 for both conditions and both indices, meaning they performed barely better than random chance at identifying these specific diseases. This is an important limitation, since a useful diagnostic tool would need substantially higher accuracy. While METS-IR did show a statistical association with these conditions in more complex analyses, this did not translate into practical diagnostic usefulness.
This research suggests that while METS-IR may be a better overall marker of cardiometabolic risk than HOMA-IR in obese adults — particularly because it does not require an insulin blood test — neither index should be used alone to diagnose or guide treatment of diabetes or high blood pressure. Instead, these tools are best used as one piece of information among many, interpreted together with a patient's age, clinical history, and other laboratory results. Relying solely on these scores could miss disease in certain groups of patients, particularly as they age.
Timurkaan E, Altuntaş G, Uslu M, Ayyıldız H, Timurkaan M. (2026). Age-related changes in METS-IR and HOMA-IR in obese adults and their relationship with cardiometabolic comorbidities.. Archives of endocrinology and metabolism. https://doi.org/10.20945/2359-4292-2026-0059