Although doctors focus on bone density to diagnose osteoporosis and risk of bone fracture, it is known that the strength of bone is more complicated than that. An amorphous term bone quality is used to get at the resistance to breaks and overall strength. It is not quantified in any accepted manner. One researcher called it "the geometric and material factors that contribute to fracture resistance"
Everything other than density that affects bone strength can be lumped into bone quality. Bone strength is a function of quantity of boney material (density) and quality.
Bone quality is affected by the architecture of the bone at a microscopic level – meaning the patterns of bony and spongy material, shape of the bony structures, and position of collagen and osteoblasts and osteoclasts. Small fractures at this level affect bone quality, as does the length, density, and thickness of the collagen strands. The size and shape of the hydroxylapatite crystals influence how fast the minerals deposit and are reabsorbed are arguably part of bone quality.
The statistical T score is (or should be) a surrogate marker for decline in bone quality and lowering of the bone’s Youngs Modulus. It is well known that the T-score has limits and is not a perfect proxy for bone strength.
Bone strength does not rise directly with bone density, but there is a relationship. It is not linear and it is not general across all people so it is not necessarily true that increasing your bones’ density increases their strength. Indeed one side-ways barb against bisphosphonates is that they are better than hardening bone than strengthening bone.
Even two bones with the same density may have markedly different qualities and strength. Young bone is generally stronger than old bone, even at the same density. Genetics makes people in their 20s have different bone densities, but the risk for fracture may be more tied to changes in density than to absolute density. It would be ideal if each person had their densities recorded at age 25 or 30 and could refer to them later in life, but the health care system today does not support that.
Tensile strength, elastic modulus, and compressive strength are metrics scientists and engineers use to describe material. Physiologist may use these terms to talk about bones, but as a practical matter these properties are not measured in the clinic with regard to diagnosis or treatment of bone diseases.
Contrary to what you might think, bones are not hard or brittle or lifeless. This might be a perception from seeing bones outside the body. Loving bone is as alive as any other organ. It uses energy (calories), oxygen, and has blood vessels coursing through it.
There are imaging and diagnostic tools that can help get at the microarchitecture of bone in an attempt to elucidate quality. The more advanced computed tomography (CT) and high-resolution magnetic resonance imaging (MRI) scans and more powerful mathematical models may allow future doctors to get a handle on quality without taking biopsies. This has to be worked out, however, and the norms and standards for each established so that the results of the tests can be translated to a risk of bone fracture and osteoporosis.
Everyone agrees that bone density is at best a surrogate for bone quality and maybe accounts for 70% of the predictive value at best.
A study of a tribe in Ecuador found earlier age at at menarche were correlated with higher bone densities in adults. There was weak evidence that the age the woman was when she first gave birth would have a correlation – later age was a little correlated with higher density in old age. Another idea is that by nursing babies, mothers put themselves at increased risk for osteoporosis – the calcium in the breast milk being mined from the mother’s bones. The Equador study found weak evidence for this idea, but not enough to confirm it. The length of time between childbirths was not found to have any correlation with bone density.
The concern about DEXA scans is that they are not accurate. Different machines can produce different readings, resulting in interpreted bone densities of up t0 10%. Given that the standard deviations in bone densities are only about 10%, this is a huge potential factor in misclassifying people as having osteopenia, osteoporosis, or healthy bones. Further, if a patient has a DEXA scan two years in a row on two different machines, the results could suggest a rapid loss or gain in bone density, giving a misleading picture about the efficacy of treatment and state of skeletal health. To be accurate, the patient should be measured in the same DEXA scan every time, and that is not feasible.
The range of bone densities used to establish T-scores is reference set taken by someone. Who? The American Medical Association? The International Society for Clinical Densitometry? We can’t find the answer to that question.
In any case, this reference set is said to be the densities of bones (hip, spine, wrist, etc.) for people at age 30. (It used to be age 25, but now 30 is more often mentioned.) There are separate reference sets for men and women. There do not appear to be separate reference sets based on race or area of the world where the patient resides, although some people feel these should be factored into the T-score.
The International Society for Clinical Densitometry claims scans are a better predictor of bone fractures than hypertension is of strokes or high cholesterol is of heart attacks.
Of course the ISCD is a professional organization of technologists that do these scans, which may inform their statements.
A recent study in Australia of patients who take bisphosphonates shows that although the drugs work, regular bone density scans in the years following initiation of treatment are not cost effective. This is due to the inaccuracy of the scans and the low variability in response to the medicine.
The International Society for Clinical Densitometry, a professional organization with a vested interest, responded that the Australian authors did not account for all possible biases in the clinical studies they considered. The criticism sounds rather weak to us.