QUICKI Calculator

What QUICKI estimates

The Quantitative Insulin Sensitivity Check Index, abbreviated QUICKI, combines fasting glucose and fasting insulin in a logarithmic equation. Higher values generally indicate greater insulin sensitivity, while lower values indicate lower sensitivity. The index describes the fasting relationship between glucose and insulin rather than directly measuring glucose uptake by muscle.

Katz and colleagues introduced QUICKI after comparing it with the euglycemic hyperinsulinemic clamp, a controlled research procedure that infuses insulin and glucose. The original study included people across a range of body weight and glucose tolerance. Later studies found useful group-level correlations, although performance differs by population and reference method.

QUICKI is mainly a research and adjunctive metabolic index. It is not an ADA diagnostic test for prediabetes or diabetes.

NIDDK notes that insulin resistance often has no symptoms and that clinicians diagnose prediabetes with glucose-based tests rather than a direct routine test for insulin resistance.

[1][2][4][6]

QUICKI formula, logarithm, and units

The original equation is QUICKI = 1 / [log10(fasting insulin in μU/mL) + log10(fasting glucose in mg/dL)]. It uses common, base-10 logarithms. Natural logarithms produce a different number and cannot be substituted. Glucose must be in mg/dL for published QUICKI values to be comparable.

If glucose is reported in mmol/L, convert it first: glucose in mg/dL = glucose in mmol/L × 18.0182. Insulin in mIU/L is conventionally the same numerical value as μU/mL on many reports. An insulin result in pmol/L requires assay-specific conversion information and should not be entered directly.

Both inputs should come from the same fasting draw. Combining insulin from one date with glucose from another breaks the paired steady-state assumption.

QUICKI has no physical unit. Extra decimal places do not correct uncertainty in the assays.

[1][3][5]

Worked QUICKI examples

For fasting glucose 95 mg/dL and insulin 10 μU/mL, log10(95) is 1.978 and log10(10) is 1.000. QUICKI = 1 / (1.978 + 1.000) = 0.336. The calculator may display 0.336 or 0.337 depending on intermediate rounding.

For glucose 5.2 mmol/L, convert first: 5.2 × 18.0182 = 93.69 mg/dL. With insulin 6 μU/mL, QUICKI = 1 / [log10(93.69) + log10(6)] = 1 / (1.972 + 0.778) = 0.364.

Keeping the glucose as 5.2 would create an invalid result.

[1][5]

Reference values are population-specific

There is no ADA-endorsed QUICKI cutoff.

The original Katz paper reported mean QUICKI values near 0.382 in nonobese participants, 0.331 in obese participants, and 0.304 in participants with diabetes. These are study-group summaries, not diagnostic boundaries. Their distributions overlap, so a single value cannot assign a disease state.

Some publications use thresholds near 0.33 or 0.34, while others derive local percentiles or outcome-based cutoffs. The chosen reference method, age, pubertal status, pregnancy, body composition, ancestry, glucose tolerance, and insulin assay all shift the distribution. A cutoff from one cohort should not be presented as universally normal or abnormal.

A useful report names the equation, assay, fasting protocol, population reference, and clinical question. If a laboratory provides a local interval, that interval is more relevant than an unsourced online category. It still does not turn QUICKI into a diabetes diagnostic test.

[1][2][3]

QUICKI compared with HOMA-IR

HOMA1-IR multiplies fasting insulin and glucose, so its value rises as the product rises. QUICKI takes the reciprocal of the sum of their logarithms, so its value falls. The two indices therefore usually rank people in opposite numerical directions while using the same inputs. QUICKI's logarithm compresses high values and can reduce skew in statistical analyses.

Neither index directly measures peripheral insulin action, and both depend heavily on fasting insulin. Their apparent agreement does not provide independent confirmation because the equations reuse the same measurements.

A clamp or frequently sampled intravenous glucose tolerance method addresses different physiology but is too demanding for routine population screening. HOMA2 is a nonlinear computer model and should not be mixed with HOMA1. QUICKI, HOMA1-IR, and HOMA2-IR each have separate scales. Serial comparisons should use the same index, laboratory, assay, preparation, and clinical setting.

[1][2][3][5]

Diagnostic and monitoring context

Prediabetes and diabetes are diagnosed with validated glucose criteria. For nonpregnant people, prediabetes includes A1C 5.7% to 6.4%, fasting plasma glucose 100 to 125 mg/dL, or two-hour glucose 140 to 199 mg/dL after a 75 g oral glucose tolerance test. Diabetes thresholds are A1C at least 6.5%, fasting glucose at least 126 mg/dL, or two-hour glucose at least 200 mg/dL, usually confirmed.

QUICKI may appear in studies of obesity, polycystic ovary syndrome, fatty liver disease, or intervention response because it can rank a large cohort at low cost. It does not tell whether a person meets criteria for those conditions, and it has no treatment target comparable to an individualized A1C goal.

A repeated QUICKI value can move when fasting insulin or glucose changes. That movement is not automatically a validated measure of clinical improvement. Review established outcomes such as A1C, fasting glucose, blood pressure, lipids, symptoms, and medication effects with the original reason for testing.

[2][4][6]

Fasting, assay, and medication limitations

Use measurements from one morning draw after the fasting period specified by the ordering service, commonly at least eight hours without calories. Acute illness, alcohol, poor sleep, strenuous exercise, smoking, and a recent meal can disturb the basal state. Insulin is secreted in pulses, so one sample also contains biological timing variation.

Glucose can decline in an unprocessed tube as blood cells consume it. Insulin stability depends on collection, processing, and storage. Insulin immunoassays lack complete harmonization. Antibody specificity, calibrators, and proinsulin cross-reactivity can change the measured concentration and therefore QUICKI.

Injected insulin complicates the result because assays differ in detection of insulin analogs and cannot always separate endogenous from exogenous insulin. Glucocorticoids and several other medicines alter glucose or insulin physiology.

Do not use this index to adjust prescribed drugs.

[2][3][5]

Population and physiologic factors

Children and adolescents undergo changes in insulin sensitivity during puberty, so adult values do not apply directly. Pregnancy also changes insulin physiology and has separate glucose targets. Older age, menopause, liver disease, kidney dysfunction, and reduced pancreatic beta-cell reserve can change one or both fasting inputs.

Advanced type 2 diabetes may produce lower insulin despite substantial insulin resistance because beta-cell output has fallen. Type 1 diabetes and pancreatic disease also violate the assumptions behind a fasting insulin surrogate. A numerically higher QUICKI in these settings does not necessarily indicate healthy insulin sensitivity.

QUICKI should not be race-corrected.

Published differences among ancestry groups can reflect body composition, social and environmental exposures, assay methods, and cohort selection. Use a relevant validated reference and the individual's clinical findings. Body weight alone does not determine insulin sensitivity. People at the same BMI can differ in visceral fat, liver fat, muscle mass, fitness, sleep, and medication exposure. QUICKI cannot identify which factor explains a low value, so interpretation should stay tied to the original clinical question.

[2][4][5]

How to use this QUICKI calculator

Enter fasting glucose and fasting insulin from the same sample. Confirm glucose is in mg/dL or let the calculator convert from mmol/L. Confirm insulin is in μU/mL or numerically equivalent mIU/L. Do not enter C-peptide, post-meal insulin, natural logarithms, or an insulin value in pmol/L.

Record the collection date, fasting duration, laboratory, and equation. Read the output as a continuous research-style index without assigning a universal normal or resistant label.

Compare serial values only when the laboratory method and collection conditions are similar. Discuss the result with the clinician who ordered fasting insulin. Ask whether validated screening with A1C, fasting plasma glucose, or an oral glucose tolerance test is indicated. Seek timely care for excessive thirst, frequent urination, unexplained weight loss, vomiting, confusion, or high glucose with illness regardless of QUICKI.

[2][4][6]

How it works

QUICKI = 1 / (log₁₀(fasting glucose) + log₁₀(fasting insulin)). Normal >0.357.

Frequently asked questions

• QUICKI vs HOMA-IR?Both assess insulin sensitivity from fasting labs. QUICKI uses logarithmic transformation; they correlate strongly.

Related calculators

References

1. Katz et al.. Quantitative insulin sensitivity check index: a simple, accurate method
2. Gutch et al.. Assessment of insulin sensitivity and resistance
3. Wallace et al.. Use and abuse of HOMA modeling
4. National Institute of Diabetes and Digestive and Kidney Diseases. Insulin Resistance and Prediabetes
5. Antuna-Puente et al.. Validity and reproducibility of HOMA-IR, QUICKI and McAuley indexes
6. American Diabetes Association Professional Practice Committee. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes 2026
7. National Institute of Diabetes and Digestive and Kidney Diseases. Diabetes Tests and Diagnosis