Comparative evaluation of simple indices of insulin resistance

Metabolism. 2004 Dec;53(12):1522-6. doi: 10.1016/j.metabol.2004.05.017.


Various surrogate methods for the quantification of insulin sensitivity have been proposed. A comparative evaluation is lacking and is relevant for the standardization of investigative methods and comparability of results. The aims of the study were to perform a comparative validation of fasting insulin, homeostasis model assessment (HOMA), Quantitative Insulin Sensitivity Check Index (QUICKI), and revised-QUICKI (R-QUICKI) against minimal model derived estimates of insulin sensitivity (SI(MM)) in nondiabetic people and to carry out a comparative evaluation of the ability of these indices as means for the identification of individuals with the metabolic syndrome (MS) on a population basis. We used 2 data sets defined as "validation sample" and "prevalence sample". Validation sample: a total of 162 healthy men and women aged 30 to 65 years were studied by frequently sampled intravenous glucose tolerance test (FSIVGTT). SI(MM) was calculated with the Minmod program. Prevalence sample: a total of 2,731 nondiabetic men and women aged 35 to 65 years were studied. In both samples, anthropometry, blood pressure, fasting glucose, insulin, triglycerides, high-density lipoprotein (HDL) cholesterol, and free fatty acid (FFA) were measured. HOMA, QUICKI, and R-QUICKI were calculated. The MS was defined according to the Adult Treatment Panel III. Validation sample: insulin, HOMA, QUICKI, and R-QUICKI significantly correlated with SI(MM) (r = -0,53, -0.52, 0.41, 0.33; all P < .001). The finding was confirmed in obese (body mass index [BMI] > or =25 kg/m(2)), but in the normal weight, the correlation coefficient for QUICKI was significantly smaller than for the other indices. Receiver operator characteristic (ROC) curve analysis performed with SI(MM) below or above the lowest 25th percentile (ie, insulin resistance yes, no) as the outcome variable and each of the 4 indices as the test variable showed no significant differences in the areas under the curve. Prevalence sample: prevalence of the MS progressively increased across quartiles of insulin resistance as evaluated by any of the 4 indices, with no significant differences between them. The novel indices QUICKI and R-QUICKI do not perform better than HOMA and fasting insulin as surrogate measures of insulin resistance or means for the identification of people with MS in the general population.

Publication types

  • Comparative Study
  • Validation Study

MeSH terms

  • Adult
  • Aged
  • Blood Pressure / physiology
  • Body Weights and Measures
  • Cholesterol, HDL / blood
  • Female
  • Glucose / analysis
  • Glucose / metabolism
  • Glucose Tolerance Test / methods
  • Homeostasis
  • Humans
  • Insulin / blood*
  • Insulin Resistance / physiology*
  • Male
  • Metabolic Syndrome / blood
  • Metabolic Syndrome / epidemiology
  • Metabolic Syndrome / physiopathology
  • Middle Aged
  • Models, Statistical
  • Prevalence
  • ROC Curve
  • Severity of Illness Index
  • Statistics as Topic
  • Triglycerides / blood


  • Cholesterol, HDL
  • Insulin
  • Triglycerides
  • Glucose