Gram Research analysis shows that a new diagnostic model combining bone density scans, fall history, and blood tests for testosterone and vitamin D can correctly identify secondary osteoporosis in older men about 80% of the time. The model uses six factors, age, spine bone density, falls, testosterone, steroid use, and vitamin D, to help doctors spot weak bones caused by underlying health conditions rather than aging alone.
Researchers created a new tool to help doctors identify a serious type of bone disease called secondary osteoporosis in older men living in their communities. The tool combines bone density measurements with fall risk assessment and checks for specific health conditions like low testosterone and vitamin D deficiency. In testing with 843 men over age 65, the tool correctly identified about 80% of men with secondary osteoporosis. While the results are promising, doctors say more testing is needed before using it widely in clinics.
Key Statistics
A 2026 cross-sectional study of 843 community-dwelling men aged 65+ found that a diagnostic model combining bone density, fall risk, and blood markers achieved 80% accuracy in identifying secondary osteoporosis in the validation set.
Among 843 older men studied from June 2023 to December 2025, 22.18% (187 men) were diagnosed with secondary osteoporosis, with the diagnostic model showing an optimism-corrected AUC of 0.731 for moderate discrimination.
The diagnostic model identified six independent predictors of secondary osteoporosis: age, lumbar spine bone density T-score, fall history, testosterone level, glucocorticoid use, and vitamin D level, according to Gram Research analysis of this 2026 study.
In internal validation testing, the diagnostic model showed an apparent AUC of 0.846 in the training set but a more realistic optimism-corrected AUC of 0.731, demonstrating the importance of validation in clinical prediction models.
The Quick Take
- What they studied: Can doctors use a combination of bone density scans, fall history, and blood tests to reliably identify secondary osteoporosis (weak bones caused by other health problems) in older men?
- Who participated: 843 men aged 65 and older from community health centers and senior clinics. About 22% had secondary osteoporosis, while the rest had normal or slightly weak bones without an underlying cause.
- Key finding: A new diagnostic model combining six factors (age, spine bone density, fall history, testosterone level, steroid use, and vitamin D level) correctly identified secondary osteoporosis about 80% of the time in testing.
- What it means for you: If you’re an older man with weak bones, this tool could help your doctor figure out if another health condition is causing the problem. However, doctors need to test it more broadly before using it in regular practice.
The Research Details
Researchers looked back at medical records from 843 older men seen at community health centers and senior clinics between June 2023 and December 2025. They divided the men into two groups: those with secondary osteoporosis (weak bones caused by specific health problems like low testosterone, steroid use, or vitamin D deficiency) and those with normal or mildly weak bones without an underlying cause.
The researchers used advanced statistical methods to identify which measurements and test results were most important for spotting secondary osteoporosis. They tested their new diagnostic model on 70% of the men to build it, then checked how well it worked on the remaining 30% to make sure it was reliable. They also used a technique called bootstrap resampling to get a realistic picture of how the model would perform in real-world situations.
This research approach is important because secondary osteoporosis is different from regular osteoporosis and requires different treatment. By creating a tool that combines multiple pieces of information (bone scans, blood tests, and medical history), doctors can more accurately identify which older men need special investigation for underlying health problems. Testing the model on a separate group of patients ensures the results aren’t just lucky: they show the tool actually works.
The study included a reasonably large sample of 843 men and used rigorous statistical methods. The researchers tested their model on a separate validation group to check reliability. However, the study looked back at existing medical records rather than following people forward over time, which is a limitation. Additionally, some of the predictors used in the model overlap with the diagnostic criteria for secondary osteoporosis itself, which could inflate how good the model appears to be.
What the Results Show
The diagnostic model identified six key factors for spotting secondary osteoporosis: age, lumbar spine bone density, history of falls, testosterone level, glucocorticoid (steroid) use, and vitamin D level. When tested on the initial group of men, the model correctly identified secondary osteoporosis about 85% of the time. When tested on a new group of men it hadn’t seen before, it correctly identified the condition about 80% of the time.
After adjusting for the fact that the model was built on the same data it was tested on, the realistic accuracy dropped to about 73%. This is considered moderate accuracy, good enough to be helpful, but not perfect. The model also showed good calibration, meaning when it said there was a 60% chance someone had secondary osteoporosis, about 60% of those people actually did.
Decision curve analysis showed that using this model would provide real clinical benefit compared to not using any tool at all. The model was particularly good at identifying men who definitely had secondary osteoporosis and those who definitely didn’t, making it useful for narrowing down which patients need further investigation.
This is one of the first studies to create a specific diagnostic tool for secondary osteoporosis in older men living in the community. Previous research has identified individual risk factors, but this study combined them into a single practical tool. The accuracy level (73% after adjustment) is reasonable for a preliminary model and similar to other bone disease prediction tools in development.
The biggest limitation is that three of the six predictors used in the model (testosterone, vitamin D, and glucocorticoid use) are also part of the definition of secondary osteoporosis itself. This means the model might appear better at prediction than it actually is. The study only looked at men from specific clinics in one region, so results may not apply to all older men. The researchers only tested the model on data from the same clinics where it was developed, it needs testing in completely different hospitals and communities before doctors should use it in regular practice. Additionally, the study was retrospective, meaning it relied on existing medical records rather than carefully collected new data.
The Bottom Line
This diagnostic model shows promise for helping doctors identify secondary osteoporosis in older men, but it’s not ready for widespread clinical use yet. Doctors should continue using standard bone density scans and blood tests to evaluate older men with weak bones. If you’re an older man diagnosed with osteoporosis, ask your doctor to check for underlying causes like low testosterone, vitamin D deficiency, and steroid use: this model may eventually help guide that investigation.
This research is most relevant to older men (65+) with diagnosed osteoporosis or weak bones, their doctors, and geriatric specialists. It’s also important for researchers developing better diagnostic tools. This doesn’t apply to younger people or women at this time.
If this model is validated in future studies and approved for clinical use, it could help doctors make diagnoses faster. However, realistic timeline for clinical implementation is 2-3 years minimum, pending external validation studies.
Frequently Asked Questions
What is secondary osteoporosis and how is it different from regular osteoporosis?
Secondary osteoporosis is weak bones caused by another health condition like low testosterone, vitamin D deficiency, or steroid use. Regular osteoporosis happens mainly from aging. This matters because secondary osteoporosis requires treating the underlying cause, not just bone-strengthening drugs.
Can this new diagnostic tool replace bone density scans for older men?
No. This tool is designed to work alongside bone density scans, not replace them. It helps doctors figure out whether weak bones are caused by an underlying health problem. More testing is needed before doctors can use it in regular practice.
What blood tests does this diagnostic model use to identify secondary osteoporosis?
The model uses testosterone level, vitamin D level, and considers glucocorticoid (steroid) use. These three factors, combined with bone density scans and fall history, help identify secondary osteoporosis in older men.
How accurate is this new diagnostic model for identifying secondary osteoporosis?
The model correctly identified secondary osteoporosis about 80% of the time when tested on new patients. After adjusting for realistic performance, accuracy was about 73%, which is considered moderate and useful for helping doctors decide which patients need further investigation.
When will doctors be able to use this diagnostic model in their offices?
The model needs external validation in different hospitals and communities before clinical use. Realistic timeline is 2-3 years minimum. Currently, it’s a research tool showing promise but not yet approved for routine clinical practice.
Want to Apply This Research?
- Track your fall incidents monthly, including date, location, and circumstances. Also log any bone density test results and blood work showing vitamin D and testosterone levels when available.
- If you’re an older man with weak bones, use the app to record your fall history and medication list (especially steroids). Share this summary with your doctor to help them assess whether secondary osteoporosis might be a concern.
- Set quarterly reminders to update your fall history, vitamin D supplementation, and any changes in steroid medications. Compare trends over time to see if fall frequency or bone health markers are improving or worsening.
This research describes a preliminary diagnostic model that has not yet been validated for clinical use outside the original study setting. This model should not be used to diagnose or treat secondary osteoporosis without consultation with a qualified healthcare provider. If you are an older man concerned about bone health or have been diagnosed with osteoporosis, speak with your doctor about appropriate testing and treatment options. The findings presented here are based on a single retrospective study and require external validation before widespread clinical application. Always consult with your healthcare provider before making any medical decisions based on research findings.
This research translation is published by Gram Research, the science division of Gram, an AI-powered nutrition tracking app.
