10-Year Fracture Risk Calculator

What a FRAX style score estimates

In 2008 the WHO Collaborating Centre for Metabolic Bone Diseases at the University of Sheffield released FRAX, a calculator that turns a short list of questions into a 10-year probability of breaking a hip or a major osteoporotic bone. A major osteoporotic fracture means the hip, the spine, the forearm or the upper arm, and the model was built from fracture studies covering roughly 60,000 people in 12 countries. It is now the reference tool behind osteoporosis practice in the United States, with country specific versions that match local fracture rates.

The official FRAX algorithm is proprietary, so this page does not attempt to reproduce its percentages. What it does instead is score the same eleven inputs into a low, moderate or high band, and then tell you whether the result points toward a bone density test. Think of it as a triage screen built on the same logic a clinician walks through before ordering a DXA scan: age, sex, body size and the seven yes or no history questions that FRAX asks.

Why settle for a band instead of a precise number? Because the precise number only does real work once a bone density result is on the table. US guidelines attach treatment decisions to FRAX scores that already include bone mineral density, so a screen like this one answers a different question: is it time to get measured at all? (University of Sheffield, tool released 2008)

The inputs, question by question

Every question below is one the official tool asks in some form. The points column shows how much weight this screen gives each answer.

The definitions matter more than they look. The tool counts glucocorticoid exposure only at that 5 mg, 3 month line, so a two-week prednisone burst does not count, and it counts alcohol as a daily average of 3 or more units rather than a weekend binge. Answer with those definitions in mind, not with worst case memory.

How the scoring works

The screen adds your points into a total between 0 and 17, then reads it against three bands. Under 3 points is low, 3 to 5 is moderate, and 6 or more is high. The weights copy the ranking that large fracture studies keep finding: your own fracture history is the strongest single signal at 3 points, a parent who broke a hip comes next at 2, and the lifestyle factors carry 1 point each.

Age and sex do the quiet work in the background. A 45-year-old man with one risk factor can score the same 1 as a 55-year-old woman with none, but they are not the same bet, which is why the age ladder runs 0 to 4 and women carry an extra point. Low body weight works the same way: the less bone bank you carry, the less a fall has to take to snap something.

One caution about the bands: they are ordinal, not arithmetic. The step from 5 to 6 is a bigger clinical jump than the step from 2 to 3, because 6 is where risk factor profiles start to overlap with the patients that DXA scans and fracture treatment trials enrolled.

Three profiles, three answers

Three worked profiles show how different doors lead to the same room.

Profile A, the quiet moderate. A 67-year-old woman stands 5 ft 4 in and weighs 122 pounds. Her BMI runs 703 x 122 / 4,096 = 20.9, which adds 1 point for body size. Age 60 to 69 adds 2 and being female adds 1, so 2 + 1 + 1 = 4 points, the middle of the moderate band. At 65 or older she clears the screening line on age alone, so the DXA suggestion in the panel reads yes before the score is even totaled.

Profile B, the behavioral moderate. A 58-year-old man stands 5 ft 10 in and weighs 209 pounds, a BMI of 30.0, which scores 0. Age adds 1, and his two yes answers, current smoking and a father who broke a hip, add 1 + 2 = 3 points. The total is 1 + 3 = 4, moderate again, but through a different door: his risk rides on behavior and family history rather than body size.

Profile C, the obvious high. A 74-year-old woman takes prednisone 7.5 mg for two years and broke a wrist stepping off a curb. At 5 ft 5 in and 127 pounds her BMI is 21.1, so body size adds 1. Age adds 3, being female adds 1, the wrist fracture adds 3 and the steroids add 1: 3 + 1 + 1 + 3 + 1 = 9, high band. This is the profile where a clinician usually stops debating and orders the scan.

Reading the bands, and the bone density question

The panel answers the bone density question with two rules, and age goes first. The USPSTF advises screening all women 65 and older, and postmenopausal women under 65 who carry at least one risk factor, which is why a 67-year-old woman gets flagged even with a moderate score (USPSTF, 2025). Score comes second: a moderate or high band makes the test worth discussing at any age, because a moderate profile matches the kind of patient who turns out to have low bone density often enough to justify the scan.

Once a DXA result exists, the numbers sharpen. The Bone Health and Osteoporosis Foundation Clinician's Guide sets the treatment line at a 10-year probability of 20% or more for a major osteoporotic fracture, or 3% or more for a hip fracture, in someone whose scan shows osteopenia (BHOF, 2022). For context, a 65-year-old white woman at BMI 25 with no risk factors carries a 10-year major fracture probability of about 9% before any scan, so the treatment line sits above the average healthy 65-year-old rather than below her (USPSTF, 2018).

What the score leaves out

Falls are the biggest blind spot. More than 95% of hip fractures follow a fall, usually sideways, and each year more than 300,000 people 65 and older are hospitalized for hip fractures in the United States (CDC). A 78-year-old with sturdy bones and a shaggy staircase can break a hip before any score says so, which is why fall proofing the home and reviewing balance affecting medications belongs in the same conversation.

Bone density is the second gap. The official tool runs with or without a femoral neck density value, and its predictions get sharper when the scan is included, so this screen and FRAX itself are working from history alone until you have one. Dose is a third gap: the model treats a pack a day and three cigarettes a day identically, and the same flattening applies to steroids and alcohol, so heavy exposure means the band is more likely to understate things than to overstate them.

Finally, the screen is not built for everyone. It assumes no known osteoporosis and no fragility fracture already diagnosed. If you already carry a T-score of -2.5 or lower, or you have already broken a bone after a fall from standing height, you are past screening and into treatment decisions with your clinician.

Lowering the number before it becomes a fracture

Three of the point sources respond to change. Quitting smoking removes the smoking point and slows the bone loss that smoking drives. Keeping alcohol under 3 units a day removes the alcohol point and trims fall odds at the same time. Strength and balance work, walking plus resistance training twice a week, does double duty by slowing bone loss and cutting the falls that turn fragile bones into fractures.

The calcium and vitamin D question is simpler than supplement advertising suggests. Food first is the standing advice from the National Institute of Arthritis and Musculoskeletal and Skin Diseases: dairy, fortified plant milks, canned fish with bones and leafy greens cover most adults, and clinicians usually reserve supplements for people who cannot reach those targets (NIAMS). Steroid exposure deserves its own conversation, not self tapering: if a condition needs prednisone, the fix is the lowest effective dose worked out with the prescribing doctor, not stopping on your own.

Recheck the score when something changes. A new parent hip fracture, a new long term medication, a broken wrist or a five year birthday moves the inputs, and the whole point of a 0 to 17 scale is that it moves with you. Bring your list of yes answers to the appointment, because those seven questions are exactly what the clinician will enter into the official tool.