Bone Mineral Density (BMD): What the Test Actually Measures

What does a bone mineral density test actually measure?

It measures the bone mineral inside a defined patch of your skeleton, divided by the flat area of that patch — and nothing more. The Royal Osteoporosis Society puts the plain version first: "A bone density scan measures your bone density," taking images of "your spine and hips - two of the most common places where bones break" Royal Osteoporosis Society.

Underneath that sentence sits a specific piece of arithmetic. The scanner measures bone mineral content — the grams of mineral it detects — and area, the size of the region it outlined. Areal bone mineral density is then defined as bone mineral content divided by that projected area Skeletal Radiology, 2013. That single division is the whole quantity; everything else on a report is interpretation layered on top of it.

What does g/cm² actually mean on a bone density report?

It means grams of bone mineral spread across each square centimetre of the region the scanner saw. Nothing in the unit refers to how thick the bone is, how well its internal struts are arranged, or how much force it would take to break. Those things matter enormously to real bone strength — they are simply not what this number counts.

Notice too what the unit is not. Density, in the everyday physical sense, is mass per unit volume — grams per cubic centimetre. A bone density report gives grams per square centimetre. The field keeps these strictly apart: areal BMD is recorded in g/cm², volumetric BMD in g/cm³ Skeletal Radiology, 2013.

Why is bone density called "areal" rather than "volumetric"?

Because a standard bone density scan produces a flat image, so the depth of the bone is never measured. A 2021 review in the Journal of Nuclear Medicine states it directly: "DXA is a 2-dimensional measurement of bone mass, or a measurement of area. Since bone depth is not a factor, bone size can affect apparent BMD" Journal of Nuclear Medicine, 2021.

Picture pressing a bone flat onto a page and asking how much mineral sits in each square centimetre of the outline. The answer is perfectly reproducible — and still tells you nothing about whether the bone was thin or substantial. That missing third dimension is the origin of almost every quirk in bone density interpretation. For the engineering of the scanner itself, see how a DEXA scanner works.

Does being tall or short change your bone density result?

Bone size does affect an areal measurement, and larger bones tend to read higher. The same 2021 review spells out the mechanism: "Essentially, 2 vertebrae with identical volumetric densities can have different areal densities based on size. As a result, the larger the bone the higher the apparent BMD" Journal of Nuclear Medicine, 2021. Or, more bluntly: "Body or skeletal size affects the results of DXA because of its two-dimensional geometry" Skeletal Radiology, 2013.

This is a known property of the measurement, not a fault in any particular machine — and it is why size-corrected indices exist. The best known is bone mineral apparent density (BMAD), an estimate of volumetric density derived from a standard scan by dividing areal density by the square root of the projected area. It was developed to help interpret results in growing skeletons, but has also been applied to adjust adult areal measurements Skeletal Radiology, 2013.

What this does not license is a rule of thumb about your own height. What it justifies is scepticism about treating any single absolute figure as a verdict.

Areal BMD, volumetric BMD and BMAD: what is the difference?

	Areal BMD (aBMD)	Volumetric BMD (vBMD)	BMAD
What it is	Bone mineral content ÷ projected area	Bone mineral content ÷ bone volume	Volumetric density estimated from an areal scan
Units	g/cm²	g/cm³	g/cm³
How it is obtained	Standard two-dimensional bone density scan	Volumetric imaging such as quantitative CT	Calculated from areal density and the projected area
Main limitation	No depth information, so bone size influences the result	Higher cost, lower accessibility, larger X-ray exposure	An estimate derived from a flat scan, not a direct volumetric measurement

Definitions, units and limitations above: Skeletal Radiology, 2013.

Does the geometry actually matter? In that same study, lumbar (L3) vertebrae from male cadaver donors aged 22 to 81 were scanned and then crushed in a mechanical testing rig, with 20 specimens analysed. Areal density correlated with the stress at which the bone failed at r = 0.779, volumetric density measured by CT did better at r = 0.883, and the size-corrected BMAD index sat between the two at r = 0.799. Adjusting for donor age weakened every coefficient but left the same ordering intact — areal density r = 0.511, CT volumetric density r = 0.648, BMAD r = 0.567 Skeletal Radiology, 2013. This was laboratory work on 20 isolated male vertebrae, not a clinical trial — but the direction is consistent: adding depth or size information improves how well the number predicts real mechanical strength.

Why is your result given as a T-score instead of the raw g/cm² figure?

Because the raw figure does not travel. Manufacturers use different reference population databases to calculate scores at several skeletal sites, so the same person can end up with a different score depending on which system measured them Journal of Nuclear Medicine, 2021. A T-score solves this by reporting how far you sit from a reference range — it "compares your bone density to the normal range found in young healthy adults" Royal Osteoporosis Society. UK guidance is specific about which reference is used where: femoral neck and total hip T-scores come from a young healthy female reference population, while lumbar spine and forearm references are usually the scanner manufacturer's own. Z-scores instead compare your result with people of the same age, sex and ethnicity NOGG, 2024.

The scores themselves are covered in depth elsewhere — see T-scores explained, Z-score versus T-score and what osteopenia means.

How much does bone mineral density tell you about your fracture risk?

A lot, and yet not the whole story — and UK guidance says both halves out loud. NOGG's 2024 guideline states, at its highest evidence grade, that "the risk of fracture increases approximately two-fold for each standard deviation (SD) decrease in BMD." In the same document: "BMD is now viewed as one, albeit very important, risk factor to be considered when assessing fracture risk" NOGG, 2024.

Then comes the sentence that reframes everything. In the UK, 6.8% of men and 21.8% of women aged 50 and over have a femoral neck T-score of −2.5 or below — yet "the majority of people who sustain a fragility fracture will have a femoral neck BMD T-Score above -2.5" NOGG, 2024. Most broken bones happen in people the density threshold would not have flagged. The Royal Osteoporosis Society reaches the same conclusion in patient language: "your bone density scan results do not give a complete picture of your bone strength" Royal Osteoporosis Society. Which is exactly why density is combined with a formal risk calculation — see fracture risk and FRAX explained.

What can push a bone density reading up or down for the wrong reasons?

Several things unrelated to bone strength — which is why an unexpected result deserves a conversation, not a reaction. NOGG notes that in older people, spinal degenerative changes are common and "artefactually increase the BMD value", so the spine is not always the reliable site — and a disproportionately low spine reading compared with the hip is itself informative NOGG, 2024. Separately, the Royal Osteoporosis Society notes that some people over 75 are offered treatment without a scan, "because the scans are not as accurate when you get older" Royal Osteoporosis Society. Genuine physiological change can mislead too: "substantial weight loss can lead to reduced BMD, despite effective therapy" Journal of Nuclear Medicine, 2021.

Is the absolute number or the change over time more useful?

For most people, the change. Densitometry centres calculate a least significant change to separate a genuine shift from measurement noise: it is 2.77 times the centre's precision error and "is best expressed as an absolute value (g/cm2)." For that reason, patients "are encouraged to return to the same DXA scanner as was used for their prior measurement" Journal of Nuclear Medicine, 2021. NOGG adds that for monitoring treatment the lumbar spine shows the largest treatment-related changes and is the preferred site, unless degenerative change makes the hip the better choice NOGG, 2024.

Which reframes the whole measurement. One figure in g/cm² tells you where you stand today, with every areal caveat above attached. Two figures, taken the same way, tell you which direction you are travelling — and direction is what you can act on. Our guide to how often to have a bone density scan covers sensible intervals. The NHS notes you may need a scan if you are "over 50 with a risk of developing osteoporosis" or "under 50 with other risk factors" NHS; if that does not describe you but you still want a baseline, that is what a private assessment is for.

---

This article is part of our screening technology series.