How to Calculate QUICKI From Fasting Glucose and Insulin
QUICKI is 1 divided by log10 fasting insulin plus log10 fasting glucose. Here is the formula, the units, the bands, and the free calculator.
QUICKI is 1 divided by (log10 fasting insulin in µIU/mL plus log10 fasting glucose in mg/dL). The logs are base 10, not natural logs. A higher number means your fasting insulin and glucose combination points toward better insulin sensitivity.
How do I calculate QUICKI from fasting glucose and insulin?
Take fasting insulin and fasting glucose from the same blood draw, after an 8 to 12 hour overnight fast. Convert insulin to µIU/mL and glucose to mg/dL. Take the base-10 log of each number. Add the two logs. Divide 1 by that sum.
QUICKI = 1 / (log10 insulin + log10 glucose)
For example, fasting insulin 6 µIU/mL and fasting glucose 90 mg/dL:
log10 6 = 0.778 log10 90 = 1.954 0.778 + 1.954 = 2.732 1 / 2.732 = 0.366
That calculation is the entire index. You do not need body weight, age, or sex. You do need true fasting numbers.
The units are not optional
The standard formula was built on µIU/mL for insulin and mg/dL for glucose. Lab reports often differ.
- If your insulin is in pmol/L, divide by 6.945 to get µIU/mL.
- If your glucose is in mmol/L, multiply by 18.018 to get mg/dL.
- Use the base-10 logarithm. On a calculator keypad that is usually
LOG, notLN.
Using natural log, or leaving insulin in pmol/L, produces a number that can be off by enough to place you in the wrong band.
What the number says
The insulin number drives QUICKI more than glucose does in most fasting states. A person with normal glucose but fasting insulin of 15 µIU/mL has a much larger log insulin term than a person with insulin of 5 µIU/mL, so the denominator rises and QUICKI falls. That is the mechanism: insulin resistance forces the pancreas to secrete more insulin to hold glucose in range, and the log sum captures that.
In the original QUICKI paper from Katz et al. (2000), group means ran about 0.38 in normal-weight insulin-sensitive people, about 0.33 in obese insulin-resistant people, and about 0.30 in type 2 diabetes. These are group means, not diagnostic cutoffs. A value of 0.35 does not diagnose anything. But the direction is reliable: higher means more insulin sensitive, lower means more insulin resistant.
QUICKI = 1 ÷ (log10 fasting insulin in µIU/mL + log10 fasting glucose in mg/dL). Higher QUICKI means better insulin sensitivity; lower means more insulin resistance.
The common mistake is treating QUICKI like a pass/fail lab with a fixed normal range. It is a continuous index. Use it to follow change over time, not to classify a single draw. On a graph, QUICKI usually moves opposite to fasting insulin. If fasting insulin falls over months while glucose stays roughly the same, QUICKI climbs. That is the trend worth watching, and it is easier to see with two or three fasting draws than with one.
QUICKI versus HOMA-IR
Both use the same two fasting numbers. HOMA-IR, from Matthews et al. (1985), is:
HOMA-IR = (fasting glucose in mg/dL × fasting insulin in µIU/mL) / 405
HOMA-IR goes up as insulin resistance worsens. QUICKI goes down. The QUICKI log transform compresses the effect of very high insulin and glucose values, so it can be less dominated by an extreme insulin reading than HOMA-IR. In practice they are strongly related, and both are fasting surrogates for the glucose clamp. The QUICKI original paper reported a better linear correlation with clamp-derived insulin sensitivity in its sample than HOMA-IR did, but both remain approximations.
If you already have HOMA-IR on a report, QUICKI is not telling you a wholly different story. It is a different curve fit from the same two numbers.
When the simple answer does not hold
QUICKI assumes a fasting steady state. A coffee with milk, a hard workout the evening before, or an acute infection can all move fasting insulin. Alcohol the night before can also shift fasting glucose. If the draw was not truly fasting, the number is not interpretable.
The index is also less reliable when beta cells are failing. In advanced type 2 diabetes or type 1 diabetes, insulin can be low despite high glucose because the pancreas cannot secrete enough. That combination can produce a misleading QUICKI that looks mid-range while the underlying physiology is very different. QUICKI was developed and validated mainly in adults, not in pregnancy or acute illness.
If your fasting insulin is very high or your QUICKI is well outside the band you expected, that is worth discussing with a doctor. The index gives a number; it does not tell you what to do.
What if my lab did not measure insulin?
You cannot calculate QUICKI from fasting glucose alone. The formula needs insulin. If your report has fasting triglycerides and glucose, the TyG index is an alternative that does not require insulin. But for QUICKI specifically, no insulin means no result.
Run it on your own numbers
To skip the arithmetic, open the QUICKI calculator, enter your fasting insulin and glucose, and it returns the result and its band in your browser. Nothing you type is sent or stored. If you have a full lab PDF or photo, uploading it at healthosx.com/blood works out every index your report allows at once and, on a second upload, shows how each marker moved.
Written by
HealthOSX Research