The guide

Getting tested, and telling a real change from noise

By GyroscopeUpdated November 15, 20267 sources

Ask for ApoB to be added to your next cholesterol test, or use a home kit. You don’t need to fast. Retest at the same lab 6 to 12 weeks after a change, and only trust a change of more than about 15% as probably real, or about 23% as almost certainly real.

Why most people have never had one

A standard cholesterol panel reports total cholesterol, LDL, HDL and triglycerides. ApoB isn’t on it, so it only shows up if you or your doctor ask. The National Lipid Association’s 2024 consensus says the test is “readily available” at the major US labs, cheap to perform and accurate, and that it’s “underutilized in clinical practice”. A Northwestern University release in 2026 gave one reason, which is that it usually means an extra test on top of the standard panel.

What to ask for

Ask for “apolipoprotein B”, or ApoB, added to your cholesterol panel. It comes from the same blood draw. The 2026 US guideline says the test helps most in people on treatment once their LDL goal is met, especially with triglycerides over 200, diabetes or an LDL under 70. It also says it may be reasonable for anyone not on treatment who wants a sharper picture of their risk. Where can I get an ApoB test? covers doctors, labs and the home kit.

Fasting doesn’t change it

In a 2008 study of 33,391 adults in Copenhagen, researchers compared people’s lipids by how long since they’d last eaten. After a normal meal, triglycerides rose up to about 26 mg/dL and LDL fell up to about 8. ApoB didn’t change at all. The 2026 guideline says ApoB measurement is “standardized and unaffected by fasting status”.

The reason is where ApoB comes from. Your gut does send out its own particles after a meal, but the NLA says that even then, more than 95% of the ApoB in your blood is the liver’s kind. Eat breakfast. If your panel is done at the same visit, cholesterol.guide covers whether the rest of it needs fasting.

How much a result wobbles

Two ApoB tests a few weeks apart will almost never match, even if nothing about you changed. In 1992 researchers at Johns Hopkins measured lipid clinic patients on three separate visits before any treatment. Their ApoB varied by about 6.4% from visit to visit on its own. The lab machine adds its own wobble, which the NLA says is usually 5 to 6%. Put together, here’s how big a change needs to be:

A drop from 104 to 96 is 8 points, or 7.7%. It sits inside the band, so you can’t tell yet. From 120 to 95 is 21%, probably real. From 130 to 98 is 25%, a real change.

ApoB’s band is narrower than LDL’s. A calculated LDL needs about 29% to be almost certainly real, because it’s worked out from three other numbers that each wobble, while ApoB is measured directly.

Like stepping on a scale

Nobody judges a diet from one morning on the bathroom scale. Water and salt move it a pound or two, so you watch the trend over weeks. A blood test works the same way. You get one number every couple of months instead of one every morning, which makes it more important to know how far a single result can wander.

When to retest

• After a food change: 6 to 12 weeks after it’s become your normal. In the feeding trials ApoB had moved by four weeks.
• After starting or changing a medicine: the guideline checks a lipid profile 4 to 12 weeks later, then every 6 to 12 months.
• Too soon: under 4 weeks apart won’t tell you much.
• During fast weight loss: the guideline lists rapid weight loss among the things that can raise LDL for a while, so retest once your weight has held.
• Around menopause: ApoB jumps in the year either side of the last period, so two tests that straddle it can show a rise that isn’t your diet (menopause).
• In pregnancy: cholesterol rises as a normal part of pregnancy, and the usual goals don’t apply.

Same lab, same way

The test is standardized, and in the College of American Pathologists’ accuracy survey most ApoB tests were within 4 mg/dL of the true value. Labs can still differ by a few points, which eats into a small change. Use the same lab when you can. Units differ too. US labs print mg/dL and Canadian labs g/L, and you multiply g/L by 100 to compare (how to convert).

The catch with small changes

Every food change measured for ApoB is smaller than one retest’s noise. Psyllium takes off about 5, oats about 3 and halving saturated fat about 6 to 7. On a single retest you often can’t see them. Do them anyway, and test more than once, or average two draws, before you decide what worked. Did my ApoB really change? does the arithmetic on your own two numbers.