In one sentence
Your fasting glucose multiplied by your fasting insulin and scaled so that a young, healthy reference population reads 1; the higher it is, the more insulin it is taking to hold your glucose, which is what insulin resistance means.
Why it matters
The model behind it treats fasting glucose and insulin as the two halves of a feedback loop, and when it was validated against the euglycemic clamp, the reference method, the estimate correlated with the clamp at 0.88, better than fasting insulin alone, with a precision the authors put at a coefficient of variation of 31 percent (Matthews 1985). Twenty years on, its originators reviewed its use in more than 500 publications and set out where it belongs (cohort and epidemiological studies, and following one person over time) and where it does not (reading beta-cell function in isolation), with the reminder that the index is only as good as the insulin assay beneath it (Wallace 2004). For a member it is the one number that turns the fasting insulin into something that can be followed, and it moves years before HbA1c does.
What moves it
Whatever moves fasting insulin, mostly: body fat and inactivity up, weight loss and training down, a short night and a stressful week up, the second half of the menstrual cycle up; and the length of the fast, which is why it has to be the same every time.
How to measure it well
Both inputs from one morning draw after a 10- to 12-hour fast, no hard training or alcohol for 48 hours, the same laboratory every time, because insulin assays differ by more than the changes the index is meant to show.
How Vitality Tracker uses it
Computed only when fasting glucose and fasting insulin come from one collection and the insulin can be expressed in mU/L (VT applies the laboratory's stated factor to a pmol/L result, and does not compute the index when the factor is unknown). Shown in the glycemic panel beside the two inputs. VT's working reference points (v0.1), marked working: below 1.5 as optimal and above 2.5 as the insulin-resistance range, both population-derived and shown as such, never as a personal verdict; a laboratory that prints its own HOMA range has its range shown first. The index is not in the Vitality Score. Re-test every three months with HbA1c while you are changing something, otherwise yearly.
How far to trust the number
Calculated: it carries the insulin assay's between-laboratory differences and the 31 percent imprecision of the model, so a change of less than about a third between two draws is inside the noise. The direction over several draws in one laboratory is the signal.
Sources
- Matthews DR, et al. Diabetologia 1985;28:412–9. doi:10.1007/BF00280883
- Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. doi:10.2337/diacare.27.6.1487
- Vitality Tracker, Clinician report content specification (derived indices); optimal set v0.1 (working).