T-scores and Z-scores are numbers you get from a bone density test (called a DXA scan) that tell you how your bone strength compares to other people. Both scores use the same unit of measurement, called a standard deviation, but they compare your bones to different groups. A T-score compares you to a healthy young adult at peak bone strength. A Z-score compares you to people your own age, sex, and ethnicity. Your doctor uses one or both to assess your fracture risk and decide whether you need treatment.
How Standard Deviations Work
Before the bone density numbers make sense, it helps to understand what a standard deviation actually means. It’s a way of measuring how far any single value sits from the average. If your score is 0, you’re exactly at the average. A score of +1 means you’re one standard deviation above it, and a score of -1 means you’re one standard deviation below.
In any normal distribution of data, about 68% of all values fall within one standard deviation of the average (between -1 and +1). About 95% fall within two standard deviations (between -2 and +2), and 99.7% fall within three. So a score of -2.5 puts you in the far tail of the distribution, well below where most people land. That’s exactly the threshold that matters most in bone density testing.
What a T-Score Tells You
Your T-score compares your bone mineral density to that of a healthy young adult at peak bone mass. The reference group used by most DXA machines is white females aged 20 to 29, drawn from a large national health survey called NHANES III. The World Health Organization recommends this same reference population for both women and men, which means a T-score of -2.5 represents the same actual bone density regardless of the patient’s sex.
The WHO classifies T-scores into four categories:
- Normal: -1.0 and above
- Osteopenia (mild bone loss): between -1.0 and -2.5
- Osteoporosis: -2.5 or lower
- Severe osteoporosis: -2.5 or lower with at least one fragility fracture
So if your T-score is -1.8, your bones are thinner than those of a typical 25-year-old, but you haven’t crossed into osteoporosis territory. A T-score of -3.0 means your bone density is three standard deviations below that young-adult average, placing you firmly in the osteoporosis range.
One thing to keep in mind: certain conditions can throw off your T-score. Arthritis in the spine, for example, can make the reading appear higher (better) than your actual bone strength. Previous fractures or bone spurs can also distort results. A qualified physician needs to interpret the number in context, not just read it off the report.
What a Z-Score Tells You
A Z-score answers a different question: how does your bone density compare to other people who are the same age, sex, and ethnicity as you? Instead of measuring you against a 25-year-old at peak bone mass, it measures you against your peers. This distinction matters because everyone loses some bone density with age, and comparing a 70-year-old to a 25-year-old will almost always show a gap. The Z-score filters out that expected, age-related loss.
A Z-score of -2.0 or lower is considered “below the expected range for age.” When your bones are significantly weaker than those of your peers, it suggests something beyond normal aging is at work. Your doctor may look for secondary causes of bone loss, things like vitamin D deficiency, thyroid disorders, celiac disease, long-term steroid use, or other medical conditions that accelerate bone breakdown.
When Each Score Is Used
T-scores are the standard for diagnosing osteoporosis in postmenopausal women and men over 50. These are the groups at highest risk for fractures, and the WHO diagnostic categories (normal, osteopenia, osteoporosis) were designed specifically for them.
Z-scores are the standard for nearly everyone else. The International Society for Clinical Densitometry recommends using Z-scores for premenopausal women, men under 50, and children. In these younger groups, applying T-score categories can be misleading. A premenopausal woman might have a T-score in the osteopenia range simply because she hasn’t yet reached peak bone mass, or because her natural bone density is slightly below the population average. That doesn’t carry the same fracture risk it would for a 65-year-old. Even though T-scores and Z-scores are numerically similar in young adults (since their peers and the reference group are roughly the same age), the clinical interpretation differs enough that the diagnostic label of “osteopenia” should not be applied to premenopausal women based on T-scores alone.
For younger patients, a Z-score at or below -2.0 is described as “below expected range for age” rather than labeled as osteoporosis. This language matters because the treatment approach and clinical significance are different from what they would be in an older adult.
Reading Your DXA Results
A DXA scan report typically lists both scores for each body site measured, usually the hip (femoral neck) and the lumbar spine. You might see a T-score of -1.4 at the hip and -2.1 at the spine on the same report. When scores differ between sites, the lowest T-score is generally the one used for diagnosis in postmenopausal women and older men.
The formula behind both scores is identical in structure: take your bone density measurement, subtract the average of the comparison group, and divide by the standard deviation of that group. The only difference is which comparison group gets plugged in. For T-scores, it’s the young-adult reference population. For Z-scores, it’s your age-matched peers.
A few practical points worth noting. Your scores can change over time, and repeat scans are typically done on the same machine to keep comparisons consistent. Small changes between scans (a fraction of a standard deviation) may fall within the machine’s margin of error, so a shift from -1.3 to -1.5 over two years isn’t necessarily meaningful. Larger shifts, or scores that cross a diagnostic threshold, are what prompt changes in treatment.
If you’re postmenopausal or over 50, your T-score is the number that drives clinical decisions. If you’re younger, pay closer attention to your Z-score. Either way, the number is a starting point for a conversation about fracture risk, not a diagnosis on its own. Factors like your age, fracture history, family history, body weight, and medication use all play into the full picture.