- 01A real change vs scanner noise
Two DEXA scans of the same body a week apart do not give identical numbers. A change counts as real only when it is bigger than the clinic's least significant change, which is its precision error times 2.77. At the minimum precision the International Society for Clinical Densitometry (ISCD), the body that sets DEXA standards, accepts, that is about 5% for bone density at the spine and hip, about 8% for fat mass and about 5.5% for lean mass. A 3% drop in hip bone density or a 5% drop in fat mass between two scans can be noise.
- 02T-score vs Z-score
A T-score compares your bone density with a healthy young adult at peak bone mass, and it is the score that diagnoses osteoporosis in postmenopausal women and men aged 50 and over. A Z-score compares your bone density with people of your own age and sex, and it is the score to read before menopause, in men under 50 and in children. A Z-score of -2.0 or lower is a cue to look for a medical cause of bone loss, not a diagnosis of osteoporosis.
A DEXA (dual-energy X-ray absorptiometry) report compares each of your numbers with a reference, so reading it comes down to knowing what each number is compared against. The machine sends two low-dose X-ray beams through the body; bone, fat and lean tissue absorb them differently, which is how one scan measures all three. Start at the top of the page. If the images show your lower spine and one hip, and the table has T-score and Z-score columns, you had a bone density scan and the first half of this guide reads it. If the image is a whole-body outline and the table lists fat, lean and bone mass by arm, leg and trunk, you had a body composition scan and the second half reads it. Some clinics print both. Either way, the same rule decides whether a change between two scans is real, and it is the rule most reports leave out. As our guide to bone tests puts it, "a DEXA scan measures how much bone, blood tests explain why" (bone markers explained).
Which DEXA did you have?
| Bone density scan | Body composition scan | |
|---|---|---|
| What is scanned | Lower spine (L1 to L4) and one or both hips; sometimes the forearm | The whole body, head to feet |
| What the table prints | BMD in g/cm2, T-score, Z-score, often FRAX | Fat mass, lean mass, bone mineral content, body fat %, regional table, visceral fat estimate, indexes |
| Can it diagnose osteoporosis? | Yes, from the spine and hip sites 1 | No; a whole-body bone reading is not a diagnostic site |
| Usual reason to have one | Screening after menopause or at 65+, a fracture, steroid use | Tracking fat and muscle through training, weight loss or ageing |
Reading a bone density report: BMD, T-score and Z-score
The results table on a bone density report has one row per site (each lumbar vertebra, the total of L1 to L4, the femoral neck and the total hip) and four columns that matter. The femoral neck is the narrow top of the thigh bone just below the ball of the hip joint, where most hip fractures happen. BMD (bone mineral density) is the raw measurement in grams of mineral per square centimetre of bone seen by the X-ray. A BMD value depends on the machine brand, so a 0.850 on a Hologic machine and a 0.850 on a GE Lunar machine (the two main brands) are not the same bone.
| Column on the report | Example | What it means |
|---|---|---|
| BMD (g/cm2) | 0.812 | Bone mineral per area at that site; compare across years only on the same machine |
| T-score | -1.8 | Standard deviations (SD) above or below a healthy young adult's average; 0 means average peak bone |
| Z-score | -0.6 | SD above or below people your age and sex |
| Young-adult % or age-matched % | 78% | The same comparisons as T and Z, written as a percentage |
A standard deviation is the typical spread around an average, so a T-score of -1.0 means bone density one typical step below a young adult's. For hip T-scores the ISCD recommends one reference group: white women aged 20 to 29 from NHANES III, a large US national health survey 1. The T-score applies to postmenopausal women (and women in the menopausal transition) and men aged 50 and over; the Z-score applies to premenopausal women, men under 50 and children 1.
What does my T-score mean?
Where an example hip T-score of -1.8 fallsT-score
| T-score | Category | What usually happens next |
|---|---|---|
| -1.0 or higher | Normal | No treatment for bone density; the clinician decides when, or whether, to rescan |
| Between -1.0 and -2.5 | Low bone mass (osteopenia) | FRAX (10-year fracture risk) guides the choice: treatment is usually considered at a 10-year hip fracture risk of 3% or more, or a major fracture risk of 20% or more 2 |
| -2.5 or lower | Osteoporosis | Treatment is usually recommended, plus a check for medical causes 2 |
| -2.5 or lower with a fragility fracture (a break from a fall at standing height or less) | Severe (established) osteoporosis | Treatment is recommended in guidelines; at very high risk the clinician may choose a bone-building medicine such as teriparatide, abaloparatide or romosozumab, plus fall prevention 2 |
Four worked examples show how the bands play out. A T-score of -1.8 at the total hip sits in low bone mass; on its own it is not usually treated, and guidelines use the FRAX result and any past fracture to decide whether to consider treatment, a decision made with the clinician who ordered the scan. A T-score of -2, a common result, is also low bone mass: one step short of the osteoporosis line. A T-score of -2.6 at the femoral neck crosses into osteoporosis by 0.1 SD, and in the US the Bone Health and Osteoporosis Foundation guide recommends treatment from -2.5 down 2. A T-score of -3.5 is a full SD into the osteoporosis range: bone density is 3.5 typical steps below a young adult's, and treatment is usually recommended.
By that meta-analysis, a T-score of -3.5 at the hip carries roughly six to seven times the hip fracture risk of a T-score of -1.5, since two steps of 2.6 multiply. The number is still one input. A fragility fracture of the hip or spine means osteoporosis whatever the T-score reads 2.
Why are my spine and hip scores different?
Bone density varies from site to site, and the diagnosis uses the lowest T-score among the lumbar spine, total hip and femoral neck 1. Do not average the sites. The spine often reads better than the hip in people over 60 for a mechanical reason: spinal arthritis, worn discs, calcium in the aorta, an old compression fracture and metal from surgery all add density the X-ray counts as bone.
The radiologist drops any vertebra affected this way; at least two must remain, and a jump of more than 1.0 T-score between neighbouring vertebrae suggests an artifact 1. When the hip or spine cannot be read, or in hyperparathyroidism, a spot on the forearm bone a third of the way up from the wrist (the 33% radius site) is used instead 1.
Before menopause or a man under 50: reading the Z-score
A Z-score of -2.0 or lower is defined as 'below the expected range for age' and a Z-score above -2.0 is 'within the expected range for age.'International Society for Clinical Densitometry, 2023 Adult Official Positions 1
Osteoporosis is not diagnosed from bone density alone in this group; the ISCD states that it cannot be diagnosed in men under 50 on BMD alone 1. A Z-score of -2.0 or lower is the cue for your clinician to look for a secondary cause, meaning an illness or medicine that is removing bone. The usual list: vitamin D deficiency, hyperparathyroidism, hyperthyroidism or too high a dose of thyroid hormone for hypothyroidism, celiac disease, glucocorticoid (steroid) use, Cushing's syndrome, low testosterone or low estrogen, eating disorders, some antiseizure medicines (phenytoin, carbamazepine), aromatase inhibitors for breast cancer and hormone therapy for prostate cancer.
The first blood tests behind a low Z-score
FRAX, TBS and VFA: the extra lines on some reports
| Line on the report | What it is | When it matters |
|---|---|---|
| FRAX major osteoporotic | 10-year chance of a spine, hip, forearm or shoulder fracture, from age, sex, BMI, risk factors and femoral neck BMD | In the US, 20% or more with low bone mass usually means treatment is considered 2 |
| FRAX hip | 10-year chance of a hip fracture | 3% or more with low bone mass usually means treatment is considered 2 |
| TBS (trabecular bone score) | A texture score from the spine image that reflects the inner structure of bone | Can adjust the FRAX result 1; most often used where BMD understates fracture risk, as in type 2 diabetes |
| VFA (vertebral fracture assessment) | A low-dose side view of the spine that finds silent vertebral fractures | Offered when the T-score is below -1.0 with age 70+ (women) or 80+ (men), height loss over 4 cm, or steroid use 1 |
A vertebral fracture found on a VFA changes the picture more than any T-score: it counts as a fragility fracture, and in the BHOF guide a vertebral fracture is a reason for treatment on its own 2.
Has my bone density really changed?
Every clinic measures its own precision error, how much the same person's BMD varies when scanned twice. The least significant change (LSC), the smallest change that is 95% likely to be real, is that precision error times 2.77 1. Ask the clinic for its LSC; at the minimum precision the ISCD accepts, it works out as follows.
| Site | ISCD minimum precision | Least significant change |
|---|---|---|
| Lumbar spine | 1.9% | 5.3% |
| Total hip | 1.8% | 5.0% |
| Femoral neck | 2.5% | 6.9% |
Worked example: total hip BMD went from 0.812 to 0.790 g/cm2 between two scans. That is a 2.7% fall, inside a 5.0% LSC, so the report should call it stable rather than a confirmed loss. Compare BMD in g/cm2 on the same machine and software, never T-scores across machines, and never a Hologic result against a GE Lunar one. A repeat scan is usually about 2 years apart, sooner during treatment or long-term steroid use, for the same reason that a trend beats a single reading. Bone loss speeds up in the years around menopause, so a woman's first scan after menopause sets the baseline later scans are read against.
Reading a body composition report
A body composition report starts with a whole-body image, then a table that splits the body into arms, legs, trunk, android (belly) and gynoid (hips and thighs) regions, then a summary of indexes. Most reports place each value on a percentile or colour bar against a reference population, often NHANES; a percentile tells you where you sit among people of your age and sex, which is a different question from what is healthy.
| Line on the report | Unit | What it is | What to watch |
|---|---|---|---|
| Total body fat % | % | Fat mass as a share of total weight | Your trend; a change smaller than about 5.5% of the value itself (about 1.5 points at 28% body fat) can be noise 1 |
| Fat mass | kg or lb | All fat tissue | A change of more than about 8% between scans (2.77 x the 3% precision the ISCD accepts) 1 |
| Lean mass | kg or lb | Muscle plus organs, skin and water | A change bigger than about 5.5% 1; it moves with hydration and glycogen, so it is not pure muscle |
| Bone mineral content (BMC) | g | Total mineral in the skeleton | Not a diagnosis; use a hip and spine scan for bone |
| Regional table | kg per region | Fat and lean in each arm, leg and the trunk | Left-right gaps of a few percent are normal |
| Android/gynoid (A/G) ratio | ratio | Belly fat divided by hip fat | Read it beside the visceral fat line |
| Visceral fat (VAT) | g, lb or cm2 | Estimated fat around the organs | Your own trend on one machine |
| FMI | kg/m2 | Fat mass divided by height squared | Published classes by sex 6 |
| FFMI or LMI | kg/m2 | Lean mass divided by height squared | Height-adjusted muscle plus organs |
| ALMI or RSMI | kg/m2 | Arm plus leg lean mass divided by height squared | The muscle number with sarcopenia cutoffs 5 |
What is a good body fat percentage on DEXA?
No guideline sets a single healthy body fat percentage, and a DEXA result often differs by several points from a tape measure, smart scale or calipers, so a DEXA 28% and a scale 24% can describe the same body. Fat mass index is the more useful number because it adjusts for height the way BMI does, but counts only fat. Kelly and colleagues proposed FMI classes from US NHANES DEXA scans 6.
| FMI class | Women (kg/m2) | Men (kg/m2) |
|---|---|---|
| Normal | about 5 to 9 | about 3 to 6 |
| Excess fat | over 9 to 13 | over 6 to 9 |
| Obese | over 13 | over 9 |
Where an example fat mass index of 10.4 falls (women)kg/m2
CalculatorBody fat calculatorA cheap way to follow your trend between scans with a tape measure. Compare tape result with tape result; the number will not match your DEXA percentage.Work it out Is my muscle mass low? ALMI and sarcopenia
Total lean mass mixes muscle with organs and water, so the muscle proxy is appendicular lean mass (ALM), the lean mass of the arms and legs. The European Working Group on Sarcopenia in Older People (EWGSOP2) sets low muscle quantity at an ALM under 20 kg in men or 15 kg in women, or an ALMI (ALM divided by height squared) under 7.0 kg/m2 in men or 5.5 kg/m2 in women 5. The US Foundation for the National Institutes of Health (FNIH) Sarcopenia Project used ALM divided by BMI instead, with cutoffs of 0.789 in men and 0.512 in women 7. Low muscle on a DEXA becomes sarcopenia only when strength is also low, measured by grip strength or a timed chair rise 5.
These cutoffs were set for older adults, so they fit a 70-year-old better than a 30-year-old. If your ALMI is under the cutoff, a clinician can check grip strength to see whether it matters; resistance training two to three times a week with enough protein is the usual first step, and a rescan after about 6 months shows whether it is working. A low ALMI, or lean mass or weight falling over months without a change in diet or training, is a result to take to a clinician rather than to explain away as water.
Short-term swings are a different matter: lean mass on a DEXA moves with water. A low-carbohydrate diet empties glycogen (the carbohydrate your muscles and liver store), which is held with water, and can shift lean mass noticeably within a week; starting creatine pulls water into muscle and can raise it. Neither change is muscle gained or lost, so read lean mass trends across months. For goals around strength and ageing, see body performance and longevity; the lean body mass calculator estimates lean mass between scans, and the protein calculator sets an intake to support it.
The supplement that shows up on a lean mass reading
Visceral fat on a DEXA
Visceral fat (VAT) is the fat packed around the liver, gut and other organs. DEXA estimates it from a band across the abdomen rather than imaging it directly, and brands calculate it differently, so there is no single normal visceral fat number on DEXA that holds across machines; it works best as your own trend on one machine. What counts as high, and what lowers it, is on our visceral fat page, and the waist-to-height ratio calculator is a free check between scans. A high visceral fat estimate or a high A/G ratio is the reason to look at metabolic blood work, because visceral fat travels with insulin resistance, fatty liver and metabolic syndrome; for more, see menopause weight gain and weight loss.
Blood work to read beside a high visceral fat number
Why your next scan might disagree
The ISCD minimum precision for body composition is 3% for fat mass, 2% for lean mass and 2% for body fat percentage 1. Multiplied by 2.77, as for bone, that gives a least significant change of about 8% for fat mass and about 5.5% for lean mass and body fat percentage, so a smaller change is not yet a confirmed change; ask the clinic for its own figure. Worked example: fat mass went from 24.0 to 22.8 kg, a 5% fall, inside an 8% LSC, so it is not yet a confirmed loss. Hydration, a large meal, a hard workout the day before, glycogen, creatine and time of day all shift lean and fat mass. Rescan when the change you expect is larger than that noise, usually after a 3 to 6 month phase of training or weight loss rather than monthly.
What to check in your blood next
On the bone side, a low T-score or Z-score usually leads to 25-OH vitamin D, calcium and PTH, then TSH (thyroid-stimulating hormone), and testosterone or estradiol when relevant. Bone turnover markers (CTX and P1NP) matter most before and during treatment; the bone markers guide linked above reads them. Vitamin D and calcium supplements can support bone in people who are deficient or eat too little calcium, though trials have not shown that they prevent fractures in everyone; they do not reverse a low T-score on their own, and too much calcium from supplements can raise the risk of kidney stones. Whether you need either is a question for your clinician, and vitamin D and calcium explain who benefits.
On the body composition side, creatinine reads higher in people with more muscle, so a muscular person's kidney result can look worse than it is; a cystatin C vs creatinine check gives a kidney reading that muscle does not skew. If lean mass falls for no clear reason, testosterone is one of the things to ask your clinician about (see testosterone).
Free, no accountRead your blood testGot blood work to go with your DEXA, such as vitamin D, PTH or a metabolic panel? Read every line of it, free.Read my results
Citations
- International Society for Clinical Densitometry. 2023 ISCD Official Positions: Adult
- LeBoff MS et al. The clinician's guide to prevention and treatment of osteoporosis. Osteoporos Int. 2022;33(10):2049-2102
- US Preventive Services Task Force. Screening for osteoporosis to prevent fractures: US Preventive Services Task Force recommendation statement. JAMA. 2025;333(6):498-508
- Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 1996;312(7041):1254-1259
- Cruz-Jentoft AJ et al. Sarcopenia: revised European consensus on definition and diagnosis (EWGSOP2). Age Ageing. 2019;48(1):16-31
- Kelly TL, Wilson KE, Heymsfield SB. Dual energy X-ray absorptiometry body composition reference values from NHANES. PLoS One. 2009;4(9):e7038
- Studenski SA et al. The FNIH Sarcopenia Project: rationale, study description, conference recommendations, and final estimates. J Gerontol A Biol Sci Med Sci. 2014;69(5):547-558
- RadiologyInfo.org (RSNA and ACR). Bone Densitometry (DEXA, DXA)
Educational context only, not medical advice or a diagnosis. Always discuss your results with a clinician.







