Minimal model analysis of intravenous glucose tolerance test (IVGTT) glucose and insulin concentrations offers a validated approach to measuring insulin sensitivity, but model identification is not always successful. Improvements may be achieved by using alternative settings in the modelling process, although results may differ according to setting and care must be exercised in combining results. IVGTT data (12 samples, regular test) from 533 men without diabetes was modelled by the traditional non-linear regression (NLR) approach, using 5 different permutations of settings. Results were evaluated with reference to the more robust Bayesian Hierarchical (BH) approach to model identification and to the proportion of variance they explained in known correlates of insulin sensitivity (age, BMI, blood pressure, fasting glucose and insulin, serum triglyceride, HDL cholesterol and uric acid concentration). BH analysis was successful in all cases. With NLR analysis, between 17 and 35 IVGTTs were associated with parameter coefficients of variation (PCVs) for minimal model parameters S_I (insulin sensitivity) and S_G (glucose effectiveness) of >100%. Systematic use of each different approach in combination reduced this number to 5. Mean (inter-quartile range) S_I^NLR was then 3.14 (2.29-4.63) min^-1.mU^-1.l x10^-4 and 2.56 (1.74-3.83) min^-1.mU^-1.l x10^-4 for S_I^BH (correlation 0.86, p<0.0001). S_I^NLR explained, on average, 10.6% of the variance in known correlates of insulin sensitivity, whereas S_I^BH explained 8.5%. In a large body of data, which BH analysis demonstrated could be fully identified, use of alternative modelling settings in NLR analysis could substantially reduce the number of analyses with PCVs>100%. S_I^NLR compared favourably with S_I^BH in the proportion of variance explained in known correlates of insulin sensitivity.