Osteoporosis is a disease in which there is loss of bone tissue (and so loss of bone density), causing bones to become brittle and fracture easily.


Thinning of the bones is a natural part of the aging process. However, women are especially vulnerable to loss of bone density after the menopause, because their ovaries no longer produce oestrogen hormones, which help maintain bone mass.

Other causes of osteoporosis include premature removal of the ovaries; a diet that is deficient in calcium; certain hormonal disorders, such as an overactive thyroid gland (hyperthyroidism); long-term treatment with corticosteroid drugs; and prolonged immobility.

Osteoporosis is most common in heavy smokers and drinkers, and in very thin people.


Osteoporosis may go undiagnosed for many years. The first sign is often a fracture, typically at the wrist or the top of the femur (thigh bone), after what may have been a trivial injury. One or several vertebrae may fracture spontaneously and cause the bones to crumble, leading to progressive height loss and/or pain due to compression of spinal nerve roots.

Diagnosis and treatment

Osteoporosis is confirmed by the use of bone densitometry. Bone loss can be minimized by adequate dietary calcium and regular, sustained exercise to build the bones and maintain their strength. Bisphosphonate drugs may be used to prevent and treat bone loss following the menopause.

Long-term hormone replacement therapy (HRT) is now usually only advised for women who cannot take other treatments or in whom other treatments have been unsuccessful. Women who have gone through a premature menopause before the age of 40 may be advised to have HRT until the age of 50 to protect them against osteoporosis; for these women, the health risks of HRT are thought to be less than normal.

Osteoporosis in detail -technical


Osteoporosis is characterized by a reduction in bone mass and disruption of bone architecture, resulting in increased bone fragility and fracture risk, with fractures of the distal radius (Colles’ fracture), spine and proximal femur being most characteristic. One in two women and one in five men over the age of 50 years will suffer an osteoporotic fracture during their remaining lifetime, with massive cost to health care services.

Pathogenesis—bone mass in later life depends both on (1) peak bone mass achieved in early adulthood—strongly influenced by genetic factors, also sex hormone status, nutrition and physical activity; and (2) rate of age-related bone loss—oestrogen deficiency is a major factor in menopausal bone loss in women.

Diagnosis—dual energy X-ray absorptiometry (DXA) is the best method for measuring bone mineral density (BMD) in the spine and hip, with osteoporosis defined as present when the BMD is 2.5 standard deviations or more below normal peak bone mass (T-score ≤ –2.5).

Risk assessment—an algorithm to estimate 10-year fracture probability (FRAX) uses (1) clinical risk factors—including age, glucocorticoid therapy, a previous history of fracture, a family history of hip fracture, current smoking, alcohol abuse, and certain diseases associated with osteoporosis, e.g. rheumatoid arthritis; with or without (2) BMD measurements. This enables intervention thresholds to be based on absolute risk rather than on BMD T-scores.

Treatment—appropriate levels of exercise should be recommended, and smoking and alcohol abuse discouraged. In postmenopausal women with osteoporosis, reductions of around 30 to 50% in vertebral fracture are seen after 3 years treatment with most drug interventions, with the current consensus being that this should be continued for a minimum of 5 years. (1) First-line treatments—for postmenopausal women these would generally be regarded as alendronate, risedronate, zoledronate (all bisphosphonates) and strontium ranelate, with a particularly strong evidence base for strontium ranelate in very elderly patients. (2) Second-line treatments—raloxifene (a selective oestrogen-receptor modulator) or ibandronate. (3) Other considerations—(a) intravenous zoledronate—the treatment of choice when oral medication cannot be given or will not be absorbed; (b) parathyroid hormone peptides—use limited to women with severe vertebral osteoporosis who are intolerant of or unresponsive to other treatments; (c) hormone replacement therapy—an appropriate option in younger postmenopausal women at high risk of fracture; (d) calcium and vitamin D—should be coprescribed with other treatments; (e) glucocorticoid-induced osteoporosis—primary prevention with a bisphosphonate is recommended for patients committed to any oral dose of prednisolone for> 3 months who are> 65 years or who have sustained a previous fragility fracture. Other patients taking oral glucocorticoids for> 3 months should have their BMD measured, and those with a T-score of—1.5 or lower should be considered for treatment.


Osteoporosis is characterized by a reduction in bone mass and disruption of bone architecture, resulting in increased bone fragility and an increase in fracture risk. These fractures are widely recognized as a major health problem in the elderly population, resulting in an estimated annual cost to United Kingdom health services of £1.8 billion. One in two women and one in five men over the age of 50 years will have a fracture due to osteoporosis during their remaining lifetime. Demographic changes during the first half of the 21st century are predicted to lead to at least a doubling in the number of these fractures, largely as a result of increased longevity.


Osteoporotic fractures are termed fragility fractures (defined as occurring after a fall from standing height or less). They may occur at a number of skeletal sites but fractures of the distal radius (Colles’ fracture), spine, and proximal femur are most characteristic. The incidence of osteoporotic fractures increases markedly with age; in women, the median age for Colles’ fracture is 65 years and for hip fracture, 80 years. The age at which incidence of vertebral fractures reaches a peak is less well defined but is thought in women to be between 65 and 80 years. In men, no age-related increase in forearm fractures is seen but incidence of hip fracture rises exponentially after the age of 75 years. The prevalence of vertebral fractures rises with age in men, although less steeply than in women.

Clinical features

Colles’ fractures typically occur after a fall forwards on to the outstretched hand. They cause considerable inconvenience, usually requiring from 4 to 6 weeks in plaster and long-term adverse sequelae are seen in up to one-third of patients. These include pain, sympathetic algodystrophy, deformity, and functional impairment.

Vertebral fractures may occur spontaneously or as a result of normal activities such as lifting, bending, and coughing. A minority of vertebral fractures (possibly around one-third) present with acute and severe pain at the site of the fracture, often radiating around the thorax or abdomen. The natural history of this pain is variable; in general, there is a tendency for improvement with time but resolution is often incomplete. Multiple vertebral fractures result in spinal deformity (kyphosis), height loss, and corresponding alterations in body shape with protuberance of the abdomen and loss of normal body contours. These changes are commonly associated with loss of self-confidence and self-esteem, difficulty with daily activities, and increased social isolation. The clinical impact of vertebral fractures is thus substantial, although often underestimated.

Of all the osteoporotic fractures, hip fractures cause the greatest morbidity and mortality. They almost always follow a fall, either backwards or to the side, and require admission to hospital and surgical treatment. Because hip fractures characteristically affect frail elderly people, postoperative morbidity and mortality are high; at 6 months after fracture, mortality rates from 12 to 20% have been reported. Only a minority of older people who have had falls regain their former level of independence following a hip fracture and up to one-third require institutionalized care.


Peak bone mass is attained in the third decade of life and age-related bone loss is believed to start in both men and women around the beginning of the fifth decade; thereafter bone loss continues throughout life. In women, there is an acceleration of the rate of bone loss around the time of the menopause, the duration of which is poorly characterized but may be from 5 to 10 years.

Bone mass in later life thus depends both on the peak bone mass achieved in early adulthood and on the rate of age-related bone loss. Genetic factors strongly influence peak bone mass, accounting for up to 70 or 80% of its variance. A number of genes are likely to be involved; these include the collagen type I α-1 gene (COL1A1), a polymorphism of which is associated both with low bone mineral density (BMD) and fracture risk. Sex hormone status, nutrition, and physical activity also influence peak bone mass.

In women, oestrogen deficiency is a major pathogenetic factor in menopausal bone loss. In older men, oestrogen status is also significantly related to BMD, whereas the relationship between age-related bone loss and declining testosterone levels is less prominent. In older people, vitamin D insufficiency and secondary hyperparathyroidism are common and contribute to age-related bone loss. Other potential pathogenetic factors include declining levels of physical activity and reduced serum concentrations of insulin-like growth factors.


The mechanical competence of the skeleton is maintained by the process of bone remodelling, in which a quantum of bone is removed by osteoclasts followed by the formation of new bone, in the cavity so created, by osteoblasts. Under normal circumstances, resorption always occurs before formation and the amounts of bone resorbed and formed within each bone-remodelling unit are similar.

In menopausal bone loss, there is an increase in the number of bone remodelling units on the bone surface (along with an increased remodelling rate), resulting in a higher number than normal of remodelling units undergoing resorption at any one time. In addition, within each of these units, less bone is formed than is resorbed, leading to a negative remodelling imbalance. It is believed that one of the early, and probably transient, effects of oestrogen deficiency is to increase the activity of osteoclasts, at least in part by suppressing apoptosis. Increased osteoclastic activity causes an increase in the depth of erosion of bone by these cells, contributing to the trabecular penetration and disruption of bone architecture that characterizes postmenopausal osteoporosis.

Although bone mass and architecture are important determinants of bone strength and fracture risk, other aspects of bone composition and structure also contribute. These include composition of the bone matrix, mineral constituents, bone size, and bone geometry. In addition, increased bone turnover per se contributes to bone fragility independently of its effects on bone mass.

The pathophysiology of other forms of osteoporosis remains to be fully defined. In glucocorticoid-induced osteoporosis, reduced bone formation and low bone turnover predominate in those being treated for the long term, but there is evidence that in the early stages of treatment there is an increase in bone turnover and osteoclast activity. The alterations in bone remodelling responsible for osteoporosis in men have not been established, but the lesser degree of structural disruption of cancellous bone during ageing suggests that reduced bone formation plays a greater role in age-related bone loss in men than women. Whether this applies to men with osteoporosis, however, is uncertain.

In recent years, a number of signalling pathways central to the regulation of bone remodelling have been defined. These include the receptor activator of NF-κB ligand/osteoprotegerin pathway, which plays a major role in the regulation of osteoclast development and activity and is being exploited in the development of a human monoclonal antibody to the receptor activator of NF-κB ligand for the treatment of osteoporosis and other diseases associated with excessive bone resorption. Another is the Wnt signalling pathway, which regulates bone formation. Inactivating mutations of sclerostin, which inhibits the pathway and activating mutations of LRP5, which is a coreceptor in the pathway, are associated with high bone mass and increased bone strength.

Diagnosis and risk assessment

Measurement of BMD

Bone mass can be assessed by a number of techniques, of which dual-energy X-ray absorptiometry (DXA) is the gold standard and provides measurements of BMD in the spine and hip. According to the World Health Organization (WHO) classification, osteoporosis is present when the BMD is 2.5 standard deviations or more below normal peak bone mass (i.e. a T-score of –2.5 or less). Established osteoporosis is defined as a T-score of –2.5 or less in association with a previous fragility fracture.

Other approaches to assessment of bone mass include broad-band ultrasound attenuation and peripheral DXA. The T-scores generated by these methods differ according to the device used and so cannot be used to diagnose osteoporosis in the same way as central DXA. Nevertheless, low values obtained using these methods indicate increased fracture risk and are regarded as an indication for DXA measurements of the hip and spine.

Clinical risk factors

In clinical practice, BMD values are used to predict fracture risk in much the same way that blood pressure is used to predict stroke. Other clinical risk factors can also be used to improve prediction of fracture risk, since some of these act independently of BMD. These include age, glucocorticoid therapy, a previous history of fracture, a family history of hip fracture, current smoking practice, alcohol abuse, and certain diseases associated with osteoporosis, e.g. rheumatoid arthritis (Table 1). A World Health Organization-supported algorithm that uses these risk factors, with or without measurements of BMD, to estimate a 10-year fracture probability has been developed (FRAX; http://www.shef.ac.uk/FRAX) and enables intervention thresholds to be based on absolute risk rather than on T-scores of BMD.

Other risk factors that are associated with low BMD include untreated premature menopause, other causes of hypogonadism including treatment with aromatase inhibitors or gonadotropin-releasing hormone analogues, low body mass index, hyperthyroidism, and malabsorption. Recently, proton-pump inhibitors, thiazolidinediones, and selective serotonin receptor uptake inhibitors have been associated with increased fracture risk, although it is uncertain whether their effects are mediated solely through reduced BMD.

Risk factors for falling are major determinants of fracture risk, particularly for hip fractures in older people. Their recognition is important since many are modifiable. They include poor visual acuity, neuromuscular weakness and incoordination, reduced mobility, cognitive impairment, and the use of sedatives, tranquillizers, and alcohol. There are also many environmental hazards that increase the risk of falling, such as uneven paving stones, poor lighting, and loose carpets and wires.

Table 1  Risk factors for osteoporosis
Independent of bone mineral density Dependent on bone mineral density
Age Untreated hypogonadism
Previous fragility fracture Malabsorption
Maternal history of hip fracture Endocrine disease
Oral glucocorticoid therapy Chronic renal disease
Smoking Chronic liver disease
Alcohol intake ≥ 3 units/day Chronic obstructive pulmonary disease
Rheumatoid arthritis Immobility
Body mass index ≤ 19 kg/m2 Drugs, e.g. aromatase inhibitors, androgen deprivation therapy
Table 2  Pharmacological interventions used in the prevention of osteoporotic fractures
Intervention Dosing regimen Route of administration
  • 70 mg once weekly
  • 5 or 10 mg once daily
Etidronate 400 mg daily for 2 weeks every 3 months Oral
Ibandronatea 150 mg once monthly Oral
Ibandronateb 3 mg once every 3 months Intravenous injection
  • 35 mg once weekly
  • 5 mg once daily
Zoledronate 5 mg once yearly Intravenous infusion
Raloxifene 60 mg once daily Oral
Strontium ranelate 2 gm once daily Oral
Teriparatide 20 µg once daily Subcutaneous injection
Preotact 100 µg once daily Subcutaneous injection


Radiology plays an important role in the diagnosis of osteoporosis, particularly in the case of vertebral fractures. Since only approximately 20 or 30% of these fractures come to medical attention, lateral radiographs of the spine may be the only means of diagnosis. Even though vertebral fractures may be asymptomatic in some individuals, their diagnosis is important because of the high risk of future fractures, both in the spine and elsewhere, and the consequent need for treatment.

Table 3  Spectrum of antifracture efficacy of pharmacological interventions for osteoporosis
  Vertebral fracture Nonvertebral fracture Hip fracture
Alendronate + + +
Etidronate + nae nae
Hormone replacement therapy + + +
Ibandronate  + +* nae
Preotact  + nae nae
Raloxifene + nae nae
Risedronate + + +
Strontium ranelate + + +a
Teriparatide + + nae
Zoledronate + + +

nae, not adequately evaluated.

a Post hoc analysis in subset of patients.

Biochemical markers of bone turnover

Biochemical markers of bone resorption (such as urinary deoxypyridinoline, pyridinoline, and N-terminal and C-terminal cross-linked telopeptides of type I collagen) and formation (such as osteocalcin, bone-specific alkaline phosphatase, and C-terminal propeptide of type I procollagen) have been shown to be useful in the prediction of fracture risk, particularly when combined with measurements of BMD, and have potential use in the monitoring of response to treatment. However, their role in clinical practice has not been firmly established.

Differential diagnosis

Secondary causes of osteoporosis should be excluded where appropriate. A full blood count, liver function tests, serum calcium and phosphate levels, thyroid function tests, plasma immunoelectrophoresis, and Bence–Jones protein determination should be performed in the first instance, with further investigation if indicated. In men, in whom secondary causes are more common, determination of serum testosterone, gonadotropin, and prolactin concentrations and tests for 24-h urinary cortisol and/or dexamethasone suppression should also be performed.

Pharmacological interventions: general considerations

Interventions that are approved for the prevention and treatment of osteoporosis are shown in Table 2. Most of these are approved only for the treatment of postmenopausal osteoporosis, but alendronate, etidronate, risedronate, zoledronate and teriparatide also have licences for the prevention and/or treatment of glucocorticoid-induced osteoporosis, and alendronate, risedronate, zoledronate and teriparatide are approved for treatment of osteoporosis in men.

Calcitonin and calcitriol are also approved for osteoporosis in postmenopausal women but are little used and will not be considered further.

Positioning of treatments

Since there have been no head-to-head studies of these interventions in which fracture has been a primary end point, no direct comparisons can be made of the magnitude of fracture reduction between drugs. However, in the case of vertebral fracture, reductions from about 30 to 50% are seen in postmenopausal women with osteoporosis after 3 years of treatment with most interventions. The evidence base for antifracture efficacy at nonvertebral sites does, however, differ between interventions, as shown in Table 3. Thus only alendronate, risedronate, zoledronate, and strontium ranelate have been shown to reduce vertebral and nonvertebral fractures, including hip fractures. This distinction is important because, once a fracture occurs, the risk of a subsequent fracture at any site is increased independently of BMD and, hence, an intervention that covers all major fracture sites is preferable.

Because of their broader spectrum of antifracture efficacy, alendronate, risedronate, zoledronate, and strontium ranelate are generally regarded as front-line options in the prevention of fractures in postmenopausal women. Strontium ranelate has a particularly strong evidence base in women aged 80 years or more and is the treatment of choice in frail individuals who are unable to comply with the dosing instructions for bisphosphonates.

Since a reduction in hip fracture risk has not been shown for raloxifene or ibandronate, these drugs are generally considered as second-line options. Where intravenous therapy is required, e.g. in patients with malabsorption, intravenous zoledronate is now the treatment of choice because it has a strong evidence base and requires only once-yearly infusion. Finally, the use of parathyroid-hormone peptides is limited by their cost to women with severe vertebral osteoporosis who are intolerant to or appear to be unresponsive to other treatments.

Reduction in fracture risk has been shown to occur within 1 year of treatment for bisphosphonates and strontium ranelate. This is particularly important in the case of vertebral fractures because, after an incident vertebral fracture, there is a 20% risk of a further fracture occurring within the next 12 months, which emphasizes the importance of prompt treatment once a fracture has occurred.

Duration of therapy

The optimum duration of treatment is uncertain. There are potential concerns that long-term treatment with potent antiresorptives may increase bone microdamage and suppress its repair, possibly resulting in increased bone fragility. However, this concern has to be counterbalanced against the possibility that increased bone turnover and bone loss after withdrawal of therapy may result in increased fracture risk. The current consensus is that treatment should be continued for a minimum of 5 years; in those who remain at high risk (based on BMD and/or incident fractures during treatment), longer treatment periods may be indicated.

Compliance and persistence

Compliance and persistence with treatment for osteoporosis are poor; approximately 50% of patients do not follow their prescribed treatment regimen and/or discontinue treatment within 1 year. Patient education is important in this respect and nurse-led monitoring early in the course of treatment has been shown to improve compliance. Whether monitoring of BMD or biochemical markers of bone turnover provides additional benefits has not been established.

Current pharmacological therapeutic options for osteoporosis


The bisphosphonates are synthetic analogues of the naturally occurring compound pyrophosphate and inhibit bone resorption.

Oral bisphosphonates are generally well tolerated. Upper gastrointestinal side effects may occur with nitrogen-containing bisphosphonates (alendronate, risedronate, and ibandronate), particularly if the dosing regimen is not adhered to. It is therefore important that patients take the drug according to the instructions, specifically in the morning with a full glass of water, 30 min before food, drink, or other medications, and remaining upright for about 30 to 60 min after the dose.

Ibandronate is available as an oral or intravenous formulation. The latter is given every 3 months as an injection lasting from 15 to 30 s. Zoledronate is given once yearly in a dose of 5 mg by intravenous infusion over a minimum of 15 min. An acute-phase reaction may occur with intravenous bisphosphonate administration, particularly with the first injection, resulting in flu-like symptoms for 1 or 2 days; the severity and frequency of this can be reduced by administration of paracetamol on the day of the infusion and the subsequent 1 or 2 days.

Strontium ranelate

Strontium ranelate is composed of two atoms of stable strontium with ranelic acid as a carrier. Although its mechanism of action remains to be fully defined, there is evidence that it increases bone strength by altering bone material properties. Treatment is associated with a substantial increase in BMD in the spine and hip, although part of this increase is artefactual and due to incorporation of strontium into bone.

Strontium ranelate is taken as a single daily dose and is generally well tolerated. There is a small increase in the frequency of diarrhoea, nausea, and headache.


Raloxifene is a selective oestrogen receptor modulator that has oestrogenic (antiresorptive) effects in the skeleton without the unwanted effects of oestrogen in the breast and endometrium. It is taken orally as a single daily dose. Adverse effects include leg oedema, leg cramps, hot flushes, and a two- to threefold increase in the risk of venous thromboembolism. Its use is associated with a significant decrease in the risk of breast cancer.

Parathyroid-hormone peptides

Teriparatide (recombinant human 1-34 parathyroid-hormone peptide) and Preotact (recombinant human 1-84 parathyroid-hormone peptide) are administered by subcutaneous injection in daily doses of 20 µg and 100 µg, respectively. They have anabolic effects on bone, increasing bone formation and producing large increases in BMD in the spine. Side effects include nausea, headache, and dizziness; in addition, transient hypercalcaemia and hypercalciuria may occur.

Hormone replacement therapy

Because the risk–benefit balance of hormone replacement therapy is generally unfavourable in older postmenopausal women, it is regarded as a second-line treatment option. However, it is an appropriate option in younger postmenopausal women with a high risk of fracture, particularly those with vasomotor symptoms.

Calcium and vitamin D

Available evidence does not support a role for calcium and vitamin D alone in the prevention of osteoporotic fractures except in the institutionalized elderly population. However, calcium and vitamin D supplements should be coprescribed with other treatments for osteoporosis since the evidence base for their antifracture efficacy is derived from studies in which calcium and vitamin D were routinely administered.

Nonpharmacological interventions

Falls have an important role in the pathogenesis of fragility fractures, particularly in frail and older people. Multiple medical and environmental factors increase the risk of falling and many of these are modifiable. Multifaceted interventions have been shown to reduce the frequency of falling, although a reduction in fractures has not been shown.

A number of lifestyle measures improve bone health, including adequate dietary calcium intake and maintenance of a normal vitamin D status. Appropriate levels of exercise should be recommended and smoking and alcohol abuse discouraged. Physiotherapy and pain relief play important roles in the management of fractures.

Glucocorticoid-induced osteoporosis

Osteoporosis is a common complication of oral glucocorticoid therapy. Bone loss is most rapid during the first few months of therapy, during which there is also a rapid increase in fracture rate. Observational data indicate that increased fracture risk is seen at all doses of oral prednisolone, even those below 5 mg daily; however, higher doses are associated with more rapid bone loss and higher fracture risk.

The effects of inhaled glucocorticoids on bone are less certain but are potentially of great importance given their high level of use in the population. Cross-sectional data indicate that adverse effects on BMD may occur, particularly when high doses are administered on a long-term basis. In both adults and children, a small increase in relative risk of fracture has been demonstrated with inhaled glucocorticoid use, but because similar increases are seen in those using only bronchodilators, it is likely that the underlying illness rather than the glucocorticoids per se is responsible for the observed increase.

In the context of glucocorticoid-induced osteoporosis, the term ‘primary prevention’ is used to denote initiation of bone protective therapy at the time that glucocorticoids are initiated, whereas ‘secondary prevention’ implies that bone protection is started later in the course of glucocorticoid therapy. This distinction is important because of the rapid onset of bone loss and increase in fracture risk after glucocorticoid initiation, providing a strong rationale for early intervention in high-risk individuals.

Although a number of interventions have been evaluated in the prevention and treatment of glucocorticoid-induced osteoporosis, the evidence base is much less robust than that which exists in postmenopausal women. Nevertheless, there is reasonable evidence that alendronate, risedronate, etidronate, zoledronate and teriparatide are effective and these are approved for this indication.

Guidelines for the management of glucocorticoid-induced osteoporosis have been produced by the Royal College of Physicians. Primary prevention with a bisphosphonate is recommended in men and women committed to any oral dose of prednisolone for 3 months or more who are over the age of 65 years or who have sustained a previous fragility fracture. Bone densitometry is not required in such individuals. In others taking oral glucocorticoids for 3 months or more, bone densitometry is recommended and those with a T-score of –1.5 or less should be considered for treatment. In addition, treatment should be advised for any individuals who sustain a fragility fracture during treatment.

Further reading  

Avenell A, et al. (2005). Vitamin D and vitamin D analogues for preventing fractures associated with involutional and post-menopausal osteoporosis. Cochrane Database Syst Rev, CD000227.
Bischoff-Ferrari HA, et al. (2005). Fracture prevention with vitamin D supplementation: a meta-analysis of randomized controlled trials. JAMA, 293, 2257–64.
Boonen S, et al. (2005). Effect of osteoporosis treatments on risk of non-vertebral fractures: review and meta-analysis of intention-to-treat studies. Osteoporos Int, 16, 1291–8.
Compston JE, Seeman E. (2006). Compliance with osteoporosis therapy is the weakest link. Lancet, 368, 973–4.
Cranney A, et al. (2006). Clinical Guidelines Committee of Osteoporosis Canada. Parathyroid hormone for the treatment of osteoporosis: a systematic review. CMAJ, 175, 52–9.
Cummings SR, Melton III LJ. (2002). Epidemiology and outcomes of osteoporotic fractures. Lancet, 359, 1761–7.
Guidelines Working Group for the Bone and Tooth Society, National Osteoporosis Society, Royal College of Physicians. (2002). Glucocorticoid-induced osteoporosis: guidelines for prevention and treatment. Royal College of Physicians, London.
Johnell O, Kanis JA. (2006). An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int, 17, 1726–33.
Kanis JA. (2002). Diagnosis of osteoporosis and assessment of fracture risk. Lancet, 359, 1929–36. 
Nguyen ND, Eisman JA, Nguyen TV. (2006). Anti-hip fracture efficacy of bisphosphonates: a Bayesian analysis of clinical trials. J Bone Miner Res, 21, 340–9.
Lock CA, et al. (2006). Lifestyle interventions to prevent osteoporotic fractures: a systematic review. Osteoporos Int, 17, 20–8.
O’Donnell S, et al. (2006). Strontium ranelate for preventing and treating postmenopausal osteoporosis. Cochrane Database Syst Rev, CD005326.
Pennisi P, Trombetti A, Rizzoli R. (2006). Glucocorticoid-induced osteoporosis and its treatment. Clin Orthop Relat Res, 443, 39–47.
Poole KE, Compston JE. (2006). Osteoporosis and its management. Br Med J, 333, 1251–6.
Rosen CJ. (2005). Clinical practice. Postmenopausal osteoporosis. N Engl J Med, 353, 595–603.
Seeman E, et al. (2006). Anti-vertebral fracture efficacy of raloxifene: a meta-analysis. Osteoporos Int, 17, 313–16. 
Siris ES, et al. (2004). Bone mineral density thresholds for pharmacological interventions to prevent fractures. AMA Arch Intern Med, 164, 1108–12.
Stevenson M, et al. (2005). A systematic review and economic evaluation of alendronate, etidronate, risedronate, raloxifene and teriparatide for the prevention and treatment of postmenopausal osteoporosis. Health Technol Assess, 9, 1–160.