About
Tools and calculators
A calculator is only as good as the line it compares you against, and several of the lines in common use came from a different question entirely. This page is about which line each of the tools uses. The tools themselves say where their own numbers came from and what those numbers can't settle.
What a calculator can and can't do for you
The sums in these tools are the easy part, and that isn't where the difficulty lies at all. HOMA-IR multiplies two blood results and divides by a fixed number. The hard question is what the answer means, and that depends entirely on the threshold you compare it against.
Those thresholds are choices rather than facts of biology. One Spanish study of general adults produced HOMA-IR cut-offs running from 1.58 to 3.46, all of them out of one dataset, and what changed between them was the rule applied. One rule took a percentile, another used a metabolic syndrome definition, and the age and sex of the person moved it again. A calculator that returns a single verdict hides all of that from you.
So the pages here spend more time on where a number came from than on the arithmetic. Each one says where its numbers came from, which group of people they came out of, and what question they were built to answer. Several of those questions turn out not to be the one a rheumatology patient is asking.
Why the thresholds count for more than the formulas
The clearest example is the omega-3 index, where the 8 percent target gets quoted a great deal and sold alongside testing. It came from one thing only, which was death from coronary heart disease as measured in heart studies. The authors who proposed it called it an informed estimate and said so openly, not one rheumatic outcome fed into it, and no trial in any rheumatic disease has aimed anyone at reaching it.
Insulin resistance has a similar problem from a different direction. No HOMA-IR threshold has been worked out in any rheumatic disease, so even the best study here borrowed its cut-offs. That study compared insulin resistance in rheumatoid arthritis and lupus, its cut-offs came from that Spanish survey of general adults, and the study says so itself. It's worth knowing that the number in your own report rests on that borrowing.
None of this makes the measurements useless, and nothing here is an argument against having them. It makes them worth reading as one input among several rather than as a verdict on their own. It also makes one question more useful than the value itself, which is the question of which threshold your clinician is using and where it came from.
Your data stays in your browser
Any calculator on this site runs entirely in the page, so nothing is sent anywhere and nothing is stored. That's why the privacy policy names no analytics provider and no outside origin, and it's a claim you can check by reading it. A tool that sent your fasting insulin to a server would have to be disclosed there, and none of these do anything of the kind.
The fonts are held on this site for the same reason, and the citation links route through doi.org or PubMed only rather than a dozen publisher domains. That isn't really a security feature so much as a decision about what a medical site should ask of a reader. Somebody looking up an insulin result shouldn't be tracked for it.
If a future tool ever needs to send anything anywhere, the policy will say so before the tool ships. That sequence is written into how this site gets maintained. The build refuses to complete if a processor appears in the code without appearing in the policy alongside it.
Questions patients ask.
Why do the tools here spend so long on where the numbers came from?
Because that's where nearly all of the trouble is. The sums in these tools are simple and they always give the same answer, while the line between normal and abnormal is a choice somebody made, using one statistical rule on one group of people. Two fair rules on one dataset gave HOMA-IR thresholds running from 1.58 to 3.46, which is wide enough to put one person on both sides of it.
Why don't some of these pages have a working calculator yet?
Because a calculator has to return a verdict, and a verdict needs a threshold this site can trace to somebody. Where no threshold exists that we can stand behind, the page explains the measurement and what moves it instead of producing a number with a color next to it. Which cut-offs get published is a decision for the physician whose name is on this site, and the calculators follow that.
Will these tools send my numbers anywhere?
No, because anything worked out here runs entirely in your browser. Nothing you type is sent anywhere, stored, or seen by us, and that's a deliberate choice rather than an accident of how it was built. It's also why the privacy policy names no analytics provider and no outside origin at all, which is a claim you can check by reading the policy.
Can a calculator tell me whether I have a problem?
It can put a number in context and it can't diagnose anything. Every result here depends on three things, which are the group it was compared against, the lab test itself, and the threshold applied, and one tube of blood can give different answers in two laboratories. What the number is good for is a better conversation with whoever ordered the test, and the question worth asking there is which threshold they're using.
Which of these is most worth reading?
The omega-6 to omega-3 page, and especially so if anybody has told you to get supplement testing. It sets out where the 8 percent omega-3 index target came from, which was death from coronary heart disease in heart studies. It also shows what a supplement does, which is to reliably move the fat readings in your blood while leaving the disease activity scores unchanged.
Are more tools coming?
Yes, and each one waits on the two things named above. It needs a threshold this site can trace to a source, and it needs a physician's sign-off on the wording around it. A calculator without both of those is a number with false confidence attached, which is worse than no calculator at all.