1. Consider assessment of fracture risk:
    • in all women aged 65 years and over and all men aged 75 years and over
    • in women aged under 65 years and men aged under 75 years in the presence of risk factors, for example:
      • previous fragility fracture
      • current use or frequent recent use of oral or systemic glucocorticoids
      • history of falls
      • family history of hip fracture
      • other causes of secondary osteoporosisg
      • low body mass index (BMI) (less than 18.5 kg/m2)
      • smoking
      • alcohol intake of more than 14 units per week for women and more than 21 units per week for men.
  2. Do not routinely assess fracture risk in people aged under 50 years unless they have major risk factors (for example, current or frequent recent use of oral or systemic glucocorticoids, untreated premature menopause or previous fragility fracture), because they are unlikely to be at high risk.
Relative values of different outcomesThe GDG wished to target opportunistic risk assessment to those individuals who are likely to be at increased risk. Hip fracture and osteoporotic fractures were considered clinically important.
Trade-off between clinical benefits and harmsThe GDG considered there were unlikely to be harms in recommending that healthcare professionals consider risk assessment in groups whose risk is higher than the general population.. Appropriate risk assessment may allow individuals to avail of preventative treatment and some factors such as alcohol and smoking can be modified using lifestyle interventions.
The GDG considered that there it was unlikely to be clinically harmful to assess people at low risk and understood that both they and healthcare professionals may gain reassurance if risk is assessed as low.
Economic considerationsThe GDG considered that absolute risk of fragility fracture differs within the general population; in particular, fracture rates for patients below 50 years are very low. Thus, risk assessment for everyone would incur unnecessary costs and may not provide any additional benefit to patients.
Targeting risk assessment for all women from age 65 and all men from age 75, and for women less than 65 years and men less than 75 years with risk factors, was considered an efficient use of resources because the risk of fracture is likely to be highest in this population.
The recommendations aim at promoting better use of resources by targeting risk assessment to patients more likely to require treatment.
Quality of evidenceAge:, the GDG reviewed the incidence and rates of fracture as reported in both the internal and external validation of QFracture and also a paper reporting fracture rates from Edinburgh, which was used in development of FRAX score for the UK. The GDG considered that hip fractures were likely to be well recorded in GP databases but that vertebral fractures were less likely to be comprehensively diagnosed and recorded.
Previous fracture: The IPD analysis of 11 cohorts (N = 60,161; mean age = 62.9 years) indicated that previous fracture is associated with increased risk of fracture, mostly independent of BMD. The GDG considered that site of fracture and the occurrence of multiple fractures are important when considering fracture risk but these are not reported in the review.
The identified data in all IPD analyses were on author-selected studies, no details on literature searching were reported, therefore there is potential for study-inclusion bias. Statistical analysis was appropriate for all the IPD analyses, limiting the potential for presentation of invalid results. The measurement of the outcome of fracture was adequately valid and reliable to limit misclassification in all the IPD analyses.
Oral or systemic glucocorticoid use: The IPD analysis of 7 cohorts (N = 42,542; mean age ranged from 55 to 70 years in men and 58 to 72 years in women, across cohorts) indicated that ever use of oral glucocorticoid is associated with an increased risk of osteoporotic fracture. The IPD analysis does not make a distinction between past and current use of glucocorticoid, the duration of treatment (continuous/intermittent use) or dose. The IPD analysis included oral glucocorticoid use only and follow up search did not find more up to date analyses that included inhaled glucocorticoids. Oral glucocorticoid use is defined in QFracture as use in the previous 6 months and was associated with increased risk of osteoporotic fracture in women and men. This finding of increased risk is further supported by data reported in the QFracture derivation study.
Smoking: In the IPD analysis of 10 cohorts (N = 59,232; mean age = 63 years), current smoking was associated with a significantly increased risk of any osteoporotic fracture and hip fracture (when relative risk [RR] adjusted for age, age and BMD, age and BMI, age and BMI and BMD). Past smoking was associated with an increased risk compared with no smoking history, but lower than for current smoking. The QFracture studies indicated increased risk with increasing cigarette consumption with people smoking more than 10 cigarettes a day.
Alcohol intake: The IPD analysis of three cohorts (N = 16,971; mean age = 65 years), indicated that alcohol intake of ≥ 2 units/day is associated with an increased risk of fracture. The magnitude of association did not change when BMD was adjusted in the model. In the QFracture derivation study, moderate to very heavy alcohol intake is associated with increasing fracture risk. The GDG had some concerns about the accuracy of alcohol consumption reported both in IPD analyses (only three cohorts included alcohol consumption and none of these were in the UK) and QFracture cohorts because the prevalence of heavy drinking was very low.
Family history of fracture: The IPD analysis of seven cohorts (N = 34,928; mean age = 65 years) indicated that family history of fracture is associated with increased risk of fracture, independent of BMD. Family history of any fractures and family history of hip fractures are both associated with increased fracture risk, but the effect of family history of hip fracture is stronger. This finding of increased risk is further supported by data reported in the QFracture derivation study.
BMI: The IPD analysis of 12 cohorts (N = 59,644; mean age = 63 years) indicated that relative risk for any fracture was 0.98 per unit increase in BMI. However, the proportion of individuals in the high BMI category was small (3.4% in ≥ 35 kg/m2 category), therefore the data may not be robust enough to detect any association. This finding of increased risk is supported by data reported in the QFracture derivation study.
History of falls: 21 studies included in the clinical evidence review indicated that a history of falls is associated with an increased risk of fracture. Eight of the 21 cohort studies that were identified were of high quality, hence they were of sufficient quality to limit bias to the results because of the following: the study population was appropriate, loss to follow-up was adequately addressed, the measurement of falls and the outcome of fracture was appropriate, confounders were accounted for and the appropriate statistical methods were used. Four cohorts were of moderate quality because the only identified bias was that it was unclear if the self-reporting of falls had been validated. One cohort was of moderate quality because loss to follow-up was not addressed. The remaining seven cohort studies were of poor quality because there was potential for bias in two or more of the following: loss to follow-up, measurement of falls, measurement of fracture, measurement of confounding factors.
Definitions of history of falls were heterogeneous across studies: a fall in the past 12 months (five studies), a fall in the past 6 months or 90 days (three studies), a fall in the past 1 month (one study), history of falls (yes vs no) (three studies), greater than two falls in the last year of follow up (one study), fall rate (two studies), and incident falls (≥ 3 vs one fall) during follow up (one study). Many studies were of large sample size and were representative of the population of interest. Most of them were non-UK studies. This finding of increased risk is further supported by data reported in the QFracture derivation study. The GDG considered that falls were likely to be poorly recorded in GP databases and the prevalence reported in GP datasets is low compared with epidemiological data.
Causes of secondary osteoporosis: The decision to include people with secondary osteoporosis in this recommendation and which groups to consider was based on GDG consensus.
No economic evidence was found on risk factors.
Other considerationsFracture rates, particularly fractures associated with osteoporosis, increase with age. The GDG was aware that the recommendations would be used by healthcare professionals without extensive knowledge of fragility fracture risk, and wished to guide practitioners both on who to target and when the use of tool was not necessary.
The GDG used the available data on age and fracture incidence, knowledge of the available risk tools and consensus to develop the recommendations. The review indicated that hip fracture rates begin to increase at age 65 in women and age 75 in men. Risk prediction should occur at around the time that intervention will be needed, so the GDG agreed that risk assessment should be considered from age 65 for women without any risk factors and from age 75 for men without any risk factors because fracture rates increase from these ages. The GDG considered whether these ages might appear as relatively high but this recommendation applies to people who do not have any clinical risk factors and is unlikely to miss those at risk as most people at risk will have additional clinical risk factors.
The GDG wanted to make a recommendation specific for the younger population (below 50 years of age), because they were aware that, without clinical risk factors, the result of the risk assessment would be low. The GDG also wanted to highlight the importance of carrying out a risk assessment in this younger population if they have a major risk factor, and the risk tools can be used within their specified age range (see recommendation 4 for further details).
The GDG considered that the aim of the recommendations was to trigger assessment and recognised that each factor identified might not be represented in the algorithm or may be expressed in the algorithm in a different way.
Previous fracture: The GDG clarified that both the site of fracture and a history of multiple fractures are important but these are not reported in the IPD review. Previous fragility fracture includes asymptomatic vertebral fractures detected on X-ray or DXA imaging. The GDG considered that height loss of 4 cm or more and/or kyphosis are possible signs of vertebral fracture, and these signs should prompt the healthcare professional to carry out further investigations, such as imaging, to correctly diagnose a vertebral fracture. This should then lead to appropriate risk assessment.
Oral or systemic glucocorticoid use: The GDG considered that there are multiple variations in how glucocorticoids are used (for example, lower doses over longer time periods for polymyalgia or short high doses for exacerbations of asthma). Individuals may also have had one-off treatment courses (for example, for Bell’s palsy). The GDG considered that the risk is associated with current and/or frequent past use of oral glucocorticoids and the GDG was aware that there is a dose effect relationship between steroid use and fracture risk75. Clinical judgement is required in assessing patients receiving glucocorticoid doses exceeding for example 7.5 mg per day for prolonged periods.
The GDG considered it was not possible or desirable to develop precise estimates of risk for all scenarios because risk is also influenced by other characteristics of the patient.
Alcohol intake: The trend in the QFracture dataset was for increased fracture risk in people as alcohol intake increased. Because these factors are not intended to be precise indicators of fracture risk but prompts to risk assessment, the GDG decided to use standard hazardous alcohol levels as a trigger to assessment. These levels are more than 21 units per week in men and more than 14 units per week in women and are consistent with the evidence on fracture risk.
Family history of fracture: The GDG noted that for younger people their parents may not yet be old enough to have sustained a hip or osteoporotic fracture, or may have died before they sustained one. Both FRAX and QFracture consider parental history of hip fracture as risk factor. In QFracture, the risk factor is either parental history of hip fracture or parental history of osteoporosis. Parental history of hip fracture is generally more clearly remembered and/or recorded, however, patients who report clear history of family vertebral or other osteoporotic fracture should also have opportunistic risk assessment performed.
BMI: Low BMI is a risk factor for osteoporosis and high BMI has been considered to be protective. The GDG was aware of increasing evidence that fragility fractures occur in women with high BMI, questioning the presumption that these women are not at risk. As a guide, the GDG agreed BMI lower than 18.5kg/m2 should definitely prompt assessment.
History of falls: The GDG considered that falls and/or a history of falling should prompt the assessment of fracture risk, with falls becoming increasingly important for older, frailer people.
Causes of secondary osteoporosis: Bone metabolism and BMD are affected by other diseases: these effects can include a direct effect on bone, effects on absorption and the effect of treatments (for example, steroids). The GDG did not consider it possible to offer an exhaustive list but suggested that the following list includes the most common causes that need consideration:

Causes of secondary osteoporosis include endocrine (hypogonadism in either sex including untreated premature menopause and treatment with aromatase inhibitors or androgen deprivation therapy; hyperthyroidism; hyperparathyroidism; hyperprolactinaemia; Cushing’s disease; diabetes), gastrointestinal (coeliac disease; inflammatory bowel disease; chronic liver disease; chronic pancreatitis; other causes of malabsorption), rheumatological (rheumatoid arthritis; other inflammatory arthropathies), haematological (multiple myeloma; haemoglobinopathies; systemic mastocytosis), respiratory (cystic fibrosis; chronic obstructive pulmonary disease), metabolic (homocystinuria), chronic renal disease and immobility (due for example to neurological injury or disease).

From: 3, Who needs formal risk assessment?

Cover of Osteoporosis: Fragility Fracture Risk
Osteoporosis: Fragility Fracture Risk: Osteoporosis: Assessing the Risk of Fragility Fracture.
NICE Clinical Guidelines, No. 146.
National Clinical Guideline Centre (UK).
Copyright © 2012, National Clinical Guideline Centre.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of National Clinical Guideline Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.