Pitfalls in the external validation of FRAX

Race-specific FRAX models are evidence-based and support equitable care: a response to the ASBMR Task Force report on Clinical Algorithms for Fracture Risk

Task Force on 'Clinical Algorithms for Fracture Risk' commissioned by the American Society for Bone and Mineral Research (ASBMR) Professional Practice Committee has recommended that FRAX® models in the US do not include adjustment for race and ethnicity. This position paper finds that an agnostic model would unfairly discriminate against the Black, Asian and Hispanic communities and recommends the retention of ethnic and race-specific FRAX models for the US, preferably with updated data on fracture and death hazards. In contrast, the use of intervention thresholds based on a fixed bone mineral density unfairly discriminates against the Black, Asian and Hispanic communities in the US. This position of the Working Group on Epidemiology and Quality of Life of the International Osteoporosis Foundation (IOF) is endorsed both by the IOF and the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO).

Fracture risk prediction in postmenopausal women with traditional and machine learning models in a nationwide, prospective cohort study in Switzerland with validation in the UK Biobank

Fracture prediction is essential in managing patients with osteoporosis and is an integral component of many fracture prevention guidelines. We aimed to identify the most relevant clinical fracture risk factors in contemporary populations by training and validating short- and long-term fracture risk prediction models in 2 cohorts. We used traditional and machine learning survival models to predict risks of vertebral, hip, and any fractures on the basis of clinical risk factors, T-scores, and treatment history among participants in a nationwide Swiss Osteoporosis Registry (N = 5944 postmenopausal women, median follow-up of 4.1 yr between January 2015 and October 2022; a total of 1190 fractures during follow-up). The independent validation cohort comprised 5474 postmenopausal women from the UK Biobank with 290 incident fractures during follow-up. Uno's C-index and the time-dependent area under the receiver operating characteristics curve were calculated to evaluate the performance of different machine learning models (Random survival forest and eXtreme Gradient Boosting). In the independent validation set, the C-index was 0.74 [0.58, 0.86] for vertebral fractures, 0.83 [0.7, 0.94] for hip fractures, and 0.63 [0.58, 0.69] for any fractures at year 2, and these values further increased for longer estimations of up to 7 yr. In comparison, the 10-yr fracture probability calculated with FRAX Switzerland was 0.60 [0.55, 0.64] for major osteoporotic fractures and 0.62 [0.49, 0.74] for hip fractures. The most important variables identified with Shapley additive explanations values were age, T-scores, and prior fractures, while number of falls was an important predictor of hip fractures. Performances of both traditional and machine learning models showed similar C-indices. We conclude that fracture risk can be improved by including the lumbar spine T-score, trabecular bone score, numbers of falls and recent fractures, and treatment information has a significant impact on fracture prediction.

An electronic health record (EHR)-based risk calculator can predict fractures comparably to FRAX: a proof-of-concept study

Information in the electronic health record (EHR), such as diagnoses, vital signs, utilization, medications, and laboratory values, may predict fractures well without the need to verbally ascertain risk factors. In our study, as a proof of concept, we developed and internally validated a fracture risk calculator using only information in the EHR.

PURPOSE

Fracture risk calculators, such as the Fracture Risk Assessment Tool, or FRAX, typically lie outside the clinician workflow. Conversely, the electronic health record (EHR) is at the center of the clinical workflow, and many variables in the EHR could predict fractures without having to verbally ascertain FRAX risk factors. We sought to evaluate the utility of EHR variables to predict fractures and, as a proof of concept, to create an EHR-based fracture risk model.

METHODS

Routine clinical data from 24,189 subjects presenting to primary care from 2010 to 2018 was utilized. Major osteoporotic fractures (MOFs) were captured by physician diagnosis codes. Data was split into training (n = 18,141) and test sets (n = 6048). We fit Cox regression models for candidate risk factors in the training set, and then created a global model using a backward stepwise approach. We then applied the model to the test set and compared the discrimination and calibration to FRAX.

RESULTS

We found variables related to vital signs, utilization, diagnoses, medications, and laboratory values to be associated with incident MOF. Our final model included 19 variables, including age, BMI, Parkinson's disease, chronic kidney disease, and albumin levels. When applied to the test set, we found the discrimination (AUC 0.73 vs. 0.70, p = 0.08) and calibration were comparable to FRAX.

CONCLUSION

Routinely collected data in EHR systems can generate adequate fracture predictions without the need to verbally ascertain fracture risk factors. In the future, this could allow for automated fracture prediction at the point of care to improve osteoporosis screening and treatment rates.

CT image-based biomarkers for opportunistic screening of osteoporotic fractures: a systematic review and meta-analysis

The use of opportunistic computed tomography (CT) image-based biomarkers may be a low-cost strategy for screening older individuals at high risk for osteoporotic fractures and populations that are not sufficiently targeted. This review aimed to assess the discriminative ability of image-based biomarkers derived from existing clinical routine CT scans for hip, vertebral, and major osteoporotic fracture prediction. A systematic search in PubMed MEDLINE, Embase, Cochrane, and Web of Science was conducted from the earliest indexing date until July 2023. The evaluation of study quality was carried out using a modified Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS-2) checklist. The primary outcome of interest was the area under the curve (AUC) and its corresponding 95% confidence intervals (CIs) obtained for four main categories of biomarkers: areal bone mineral density (BMD), image attenuation, volumetric BMD, and finite element (FE)-derived biomarkers. The meta-analyses were performed using random effects models. Sixty-one studies were included in this review, among which 35 were synthesized in a meta-analysis and the remaining articles were qualitatively synthesized. In comparison to the pooled AUC of areal BMD (0.73 [95% CI 0.71-0.75]), the pooled AUC values for predicting osteoporotic fractures for FE-derived parameters (0.77 [95% CI 0.72-0.81]; p < 0.01) and volumetric BMD (0.76 [95% CI 0.71-0.81]; p < 0.01) were significantly higher, but there was no significant difference with the pooled AUC for image attenuation (0.73 [95% CI 0.66-0.79]; p = 0.93). Compared to areal BMD, volumetric BMD and FE-derived parameters may provide a significant improvement in the discrimination of osteoporotic fractures using opportunistic CT assessments.

Development and internal validation of a clinical prediction model for osteopenia in Chinese middle-aged and elderly men: a prospective cohort study

BACKGROUND

Early identification of patients at risk of osteopenia is an essential step in reducing the population at risk for fractures. We aimed to develop and validate a prediction model for osteopenia in Chinese middle-aged and elderly men that provides individualized risk estimates.

METHODS

In this prospective cohort study, 1109 patients who attend regular physical examinations in the Second Medical Centre of Chinese PLA General Hospital were enrolled from 2015.03 to 2015.09. The baseline risk factors included dietary habits, exercise habits, medical histories and medication records. Osteopenia during follow-up were collected from Electronic Health Records (EHRs) and telephone interviews. Internal validation was conducted using bootstrapping to correct the optimism. The independent sample T-test analysis, Mann_Whitney U test, Chi-Square Test and multivariable Cox regression analysis were utilized to identify predictive factors for osteopenia in Chinese middle-aged and elderly men. A nomogram based on the seven variables was built for clinical use. Concordance index (C-index), receiver operating characteristic curve (ROC), decision curve analysis (DCA) and calibration curve were used to evaluate the efficiency of the nomogram.

RESULTS

The risk factors included in the prediction model were bone mineral density at left femoral neck (LNBMD), hemoglobin (Hb), serum albumin (ALB), postprandial blood glucose (PBG), fatty liver disease (FLD), smoking and tea consumption. The C-index for the risk nomogram was 0.773 in the prediction model, which presented good refinement. The AUC of the risk nomogram at different time points ranged from 0.785 to 0.817, exhibiting good predictive ability and performance. In addition, the DCA showed that the nomogram had a good clinical application value. The nomogram calibration curve indicated that the prediction model was consistent.

CONCLUSIONS

Our study provides a novel nomogram and a web calculator that can effectively predict the 7-year incidence risk of osteopenia in Chinese middle-aged and elderly men. It is convenient for clinicians to prevent fragility fractures in the male population.

FRAX predicts cardiovascular risk in women undergoing osteoporosis screening: the Manitoba bone mineral density registry

Osteoporosis and cardiovascular disease (CVD) are highly prevalent in older women, with increasing evidence for shared risk factors and pathogenesis. Although FRAX was developed for the assessment of fracture risk, we hypothesized that it might also provide information on CVD risk. To test the ability of the FRAX tool and FRAX-defined risk factors to predict incident CVD in women undergoing osteoporosis screening with DXA, we performed a retrospective prognostic cohort study which included women aged 50 yr or older with a baseline DXA scan in the Manitoba Bone Mineral Density Registry between March 31, 1999 and March 31, 2018. FRAX scores for major osteoporotic fracture (MOF) were calculated on all participants. Incident MOF and major adverse CV events (MACE; hospitalized acute myocardial infarction [AMI], hospitalized non-hemorrhagic cerebrovascular disease [CVA], or all-cause death) were ascertained from linkage to population-based healthcare data. The study population comprised 59 696 women (mean age 65.7 ± 9.4 yr). Over mean 8.7 yr of observation, 6021 (10.1%) had MOF, 12 277 women (20.6%) had MACE, 2274 (3.8%) had AMI, 2061 (3.5%) had CVA, and 10 253 (17.2%) died. MACE rates per 1000 person-years by FRAX risk categories low (10-yr predicted MOF <10%), moderate (10%-19.9%) and high (≥20%) were 13.5, 34.0, and 64.6, respectively. Although weaker than the association with incident MOF, increasing FRAX quintile was associated with increasing risk for MACE (all P-trend <.001), even after excluding prior CVD and adjusting for age. HR for MACE per SD increase in FRAX was 1.99 (95%CI, 1.96-2.02). All FRAX-defined risk factors (except parental hip fracture and lower BMI) were independently associated with higher non-death CV events. Although FRAX is intended for fracture risk prediction, it has predictive value for cardiovascular risk.

The implications of competing risks and direct treatment disutility in cardiovascular disease and osteoporotic fracture: risk prediction and cost effectiveness analysis

Background

Clinical guidelines commonly recommend preventative treatments for people above a risk threshold. Therefore, decision-makers must have faith in risk prediction tools and model-based cost-effectiveness analyses for people at different levels of risk. Two problems that arise are inadequate handling of competing risks of death and failing to account for direct treatment disutility (i.e. the hassle of taking treatments). We explored these issues using two case studies: primary prevention of cardiovascular disease using statins and osteoporotic fracture using bisphosphonates.

Objectives

Externally validate three risk prediction tools [QRISK®3, QRISK®-Lifetime, QFracture-2012 (ClinRisk Ltd, Leeds, UK)]; derive and internally validate new risk prediction tools for cardiovascular disease [competing mortality risk model with Charlson Comorbidity Index (CRISK-CCI)] and fracture (CFracture), accounting for competing-cause death; quantify direct treatment disutility for statins and bisphosphonates; and examine the effect of competing risks and direct treatment disutility on the cost-effectiveness of preventative treatments.

Design, participants, main outcome measures, data sources

Discrimination and calibration of risk prediction models (Clinical Practice Research Datalink participants: aged 25-84 years for cardiovascular disease and aged 30-99 years for fractures); direct treatment disutility was elicited in online stated-preference surveys (people with/people without experience of statins/bisphosphonates); costs and quality-adjusted life-years were determined from decision-analytic modelling (updated models used in National Institute for Health and Care Excellence decision-making).

Results

CRISK-CCI has excellent discrimination, similar to that of QRISK3 (Harrell's c = 0.864 vs. 0.865, respectively, for women; and 0.819 vs. 0.834, respectively, for men). CRISK-CCI has systematically better calibration, although both models overpredict in high-risk subgroups. People recommended for treatment (10-year risk of ≥ 10%) are younger when using QRISK-Lifetime than when using QRISK3, and have fewer observed events in a 10-year follow-up (4.0% vs. 11.9%, respectively, for women; and 4.3% vs. 10.8%, respectively, for men). QFracture-2012 underpredicts fractures, owing to under-ascertainment of events in its derivation. However, there is major overprediction among people aged 85-99 years and/or with multiple long-term conditions. CFracture is better calibrated, although it also overpredicts among older people. In a time trade-off exercise (n = 879), statins exhibited direct treatment disutility of 0.034; for bisphosphonates, it was greater, at 0.067. Inconvenience also influenced preferences in best-worst scaling (n = 631). Updated cost-effectiveness analysis generates more quality-adjusted life-years among people with below-average cardiovascular risk and fewer among people with above-average risk. If people experience disutility when taking statins, the cardiovascular risk threshold at which benefits outweigh harms rises with age (≥ 8% 10-year risk at 40 years of age; ≥ 38% 10-year risk at 80 years of age). Assuming that everyone experiences population-average direct treatment disutility with oral bisphosphonates, treatment is net harmful at all levels of risk.

Limitations

Treating data as missing at random is a strong assumption in risk prediction model derivation. Disentangling the effect of statins from secular trends in cardiovascular disease in the previous two decades is challenging. Validating lifetime risk prediction is impossible without using very historical data. Respondents to our stated-preference survey may not be representative of the population. There is no consensus on which direct treatment disutilities should be used for cost-effectiveness analyses. Not all the inputs to the cost-effectiveness models could be updated.

Conclusions

Ignoring competing mortality in risk prediction overestimates the risk of cardiovascular events and fracture, especially among older people and those with multimorbidity. Adjustment for competing risk does not meaningfully alter cost-effectiveness of these preventative interventions, but direct treatment disutility is measurable and has the potential to alter the balance of benefits and harms. We argue that this is best addressed in individual-level shared decision-making.

Study registration

This study is registered as PROSPERO CRD42021249959.

Funding

This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: 15/12/22) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 4. See the NIHR Funding and Awards website for further award information.

A Risk Assessment Tool for Predicting Fragility Fractures in People with HIV: Derivation and Internal Validation of the FRESIA Model

People with HIV have a higher risk of fracture than the general population. Because of the low performance of the existing prediction tools, there is controversy surrounding fracture risk estimation in this population. The aim of the study was to develop a model for predicting the long-term risk of fragility fractures in people with HIV. We included 11,899 individuals aged ≥30 years from the Spanish HIV/AIDS research network cohort. We identified incident fragility fractures from medical records, defined as nontraumatic or those occurring after a casual fall, at major osteoporotic sites (hip, clinical spine, forearm, proximal humerus). Our model accounted for the competing risk of death and included 12 candidate predictors to estimate the time to first fragility fracture. We assessed the discrimination and calibration of the model and compared it with the FRAX tool. The incidence rate of fragility fractures was 4.34 (95% CI 3.61 to 5.22) per 1000 person-years. The final prediction model included age, chronic kidney disease, and chronic obstructive pulmonary disease as significant predictors. The model accurately predicted the 5- and 10-year risk of fragility fractures, with an area under the receiving operator characteristic curve of 0.768 (95% CI 0.722 to 0.814) and agreement between the observed and expected probabilities. Furthermore, it demonstrated better discrimination and calibration than the FRAX tool, improving the classification of over 35% of individuals with fragility fractures compared to FRAX. Our prediction model demonstrated accuracy in predicting the long-term risk of fragility fractures. It can assist in making personalized intervention decisions for individuals with HIV and could potentially replace the current tools recommended for fracture risk assessment in this population. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

External validation of the QLifetime cardiovascular risk prediction tool: population cohort study

BACKGROUND

Prediction of lifetime cardiovascular disease (CVD) risk is recommended in many clinical guidelines, but lifetime risk models are rarely externally validated. The aim of this study was to externally validate the QRiskLifetime incident CVD risk prediction tool.

METHODS

Independent external validation of QRiskLifetime using Clinical Practice Research Datalink data, examining discrimination and calibration in the whole population and stratified by age, and reclassification compared to QRISK3. Since lifetime CVD risk is unobservable, performance was evaluated at 10-years' follow-up, and lifetime performance inferred in terms of performance for in the different age-groups from which lifetime predictions are derived.

RESULTS

One million, two hundreds sixty thousand and three hundreds twenty nine women and 1,223,265 men were included in the analysis. Discrimination was excellent in the whole population (Harrell's-C = 0.844 in women, 0.808 in men), but moderate to poor stratified by age-group (Harrell's C in people aged 30-44 0.714 for both men and women, in people aged 75-84 0.578 in women and 0.556 in men). Ten-year CVD risk was under-predicted in the whole population, and in all age-groups except women aged 45-64, with worse under-prediction in older age-groups. Compared to those at highest QRISK3 estimated 10-year risk, those with highest lifetime risk were younger (mean age: women 50.5 vs. 71.3 years; men 46.3 vs. 63.8 years) and had lower systolic blood pressure and prevalence of treated hypertension, but had more family history of premature CVD, and were more commonly minority ethnic. Over 10-years, the estimated number needed to treat (NNT) with a statin to prevent one CVD event in people with QRISK3 ≥ 10% was 34 in women and 37 in men, compared to 99 and 100 for those at highest lifetime risk.

CONCLUSIONS

QRiskLifetime underpredicts 10-year CVD risk in nearly all age-groups, so is likely to also underpredict lifetime risk. Treatment based on lifetime risk has considerably lower medium-term benefit than treatment based on 10-year risk.

General Comorbidity Indicators Contribute to Fracture Risk Independent of FRAX: Registry-Based Cohort Study

CONTEXT

FRAX® estimates 10-year fracture probability from osteoporosis-specific risk factors. Medical comorbidity indicators are associated with fracture risk but whether these are independent from those in FRAX is uncertain.

OBJECTIVE

We hypothesized Johns Hopkins Aggregated Diagnosis Groups (ADG®) score or recent hospitalization number may be independently associated with increased risk for fractures.

METHODS

This retrospective cohort study included women and men age ≥ 40 in the Manitoba BMD Registry (1996-2016) with at least 3 years prior health care data and used linked administrative databases to construct ADG scores along with number of hospitalizations for each individual. Incident Major Osteoporotic Fracture and Hip Fracture was ascertained during average follow-up of 9 years; Cox regression analysis determined the association between increasing ADG score or number of hospitalizations and fractures.

RESULTS

Separately, hospitalizations and ADG score independently increased the hazard ratio for fracture at all levels of comorbidity (hazard range 1.2-1.8, all P < 0.05), irrespective of adjustment for FRAX, BMD, and competing mortality. Taken together, there was still a higher than predicted rate of fracture at all levels of increased comorbidity, independent of FRAX and BMD but attenuated by competing mortality. Using an intervention threshold of major fracture risk >20%, application of the comorbidity hazard ratio multiplier to the patient population FRAX scores would increase the number of treatment candidates from 8.6% to 14.4%.

CONCLUSION

Both complex and simple measures of medical comorbidity may be used to modify FRAX-based risk estimates to capture the increased fracture risk associated with multiple comorbid conditions in older patients.

Improving Recognition of Fracture Risk in People with Human Immunodeficiency Virus: Performance and Model Contribution of Two Common Risk Assessment Tools

Current guidelines recommend screening people with HIV (PWH) for bone disease using predictive tools developed for the general population, although data on PWH are scarce. In this study, we assessed the performance of FRAX and QFracture scoring systems to predict the occurrence of fragility fractures in a prospective cohort of 17,671 adults with human immunodeficiency virus (HIV) included in the HIV/AIDS research network (CoRIS) in Spain. The survival estimates of fragility fractures during follow-up were calculated and FRAX and QFracture scores were computed at cohort inclusion. For both tools, discriminatory measures and the observed-to-expected (O/E) ratios were assessed. During a follow-up time of 42,411.55 person-years, 113 fragility fractures were recorded. Areas under the curve were 0.66 [95% confidence interval (95% CI) 0.61-0.71] for FRAX and 0.67 (95% CI 0.62-0.73) for QFracture for major osteoporotic fractures, and 0.72 (95% CI 0.57-0.88) and 0.81 (95% CI 0.68-0.95) for hip fracture, respectively. The O/E was 1.67 for FRAX and 5.49 for QFracture for major osteoporotic fractures, and 11.23 for FRAX and 4.87 for QFracture for hip fractures. Moreover, O/E raised as the risk increased for both tools and in almost all age groups. When using the recommended assessment thresholds, <6% and 10% of major osteoporotic and hip fractures would have been identified, respectively. In conclusion, FRAX and QFracture displayed acceptable discrimination, although both tools significantly underestimated the risk of fragility fractures in PWH. The recommended assessment thresholds may not be appropriate for this population as they were unable to identify individuals with fragility fractures during follow-up.

The need to distinguish intervention thresholds and diagnostic thresholds in the management of osteoporosis

This position paper of the International Osteoporosis Foundation (IOF) and the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO) addresses the rationale for separate diagnostic and intervention thresholds in osteoporosis. We conclude that the current BMD-based diagnostic criteria for osteoporosis be retained whilst clarity is brought to bear on the distinction between diagnostic and intervention thresholds.

An assessment of intervention thresholds for high fracture risk in Chile

Assessment and treatment pathways using FRAX-based intervention thresholds in Chile can be used to identify patients at high risk of fracture and avoid unnecessary treatment in those at low fracture risk.

PURPOSE

The aim of the present study was to explore treatment paths and characteristics of women eligible for treatment in Chile based on major osteoporotic fracture (MOF) probabilities derived from FRAX®.

METHODS

Intervention and assessment thresholds were derived using methods adopted by the National Osteoporosis Guideline Group for FRAX-based guidelines in the UK but based on the epidemiology of fracture and death in Chile. Age-dependent and hybrid assessment and intervention thresholds were applied to 1998 women and 1122 men age 50 years or more drawn from participants in the National Health Survey 2016-2017.

RESULTS

Approximately 12% of men and women had a prior fragility fracture and would be eligible for treatment for this reason. Using age-dependent thresholds, an additional 2.6% of women (0.3% of men) were eligible for treatment in that MOF probabilities lay above the upper assessment threshold. A BMD test would be recommended in 5% of men and 38% of women. With hybrid thresholds, an additional 13% of women (3.6% of men) were eligible for treatment and BMD recommended in 11% of men and 42% of women.

CONCLUSION

The application of hybrid intervention thresholds ameliorates the disparity in fracture probabilities seen with age-dependent thresholds. Probability-based assessment of fracture risk, including the use of the hybrid intervention thresholds for Chile, is expected to help guide decisions about treatment.

Effect of competing mortality risks on predictive performance of the QFracture risk prediction tool for major osteoporotic fracture and hip fracture: external validation cohort study in a UK primary care population

Objective

To externally evaluate the QFracture risk prediction tool for predicting the risk of major osteoporotic fracture and hip fracture.

Design

External validation cohort study.

Setting

UK primary care population. Linked general practice (Clinical Practice Research Datalink (CPRD) Gold), mortality registration (Office of National Statistics), and hospital inpatient (Hospital Episode Statistics) data, from 1 January 2004 to 31 March 2016.

Participants

2 747 409 women and 2 684 730 men, aged 30-99 years, with up-to-standard linked data that had passed CPRD checks for at least one year.

Main outcome measures

Two outcomes were modelled based on the QFracture: major osteoporotic fracture and hip fracture. Major osteoporotic fracture was defined as any hip, distal forearm, proximal humerus, or vertebral crush fracture, from general practice, hospital discharge, and mortality data. The QFracture 10 year predicted risk of major osteoporotic fracture and hip fracture was calculated, and performance evaluated versus observed 10 year risk of fracture in the whole population, and in subgroups based on age and comorbidity. QFracture calibration was examined accounting for, and not accounting for, competing risk of mortality from causes other than the major osteoporotic fracture.

Results

2 747 409 women with 95 598 major osteoporotic fractures and 36 400 hip fractures, and 2 684 730 men with 34 321 major osteoporotic fractures and 13 379 hip fractures were included in the analysis. The incidence of all fractures was higher than in the QFracture internal derivation. Competing risk of mortality was more common than fracture from middle age onwards. QFracture discrimination in the whole population was excellent or good for major osteoporotic fracture and hip fracture (Harrell's C statistic in women 0.813 and 0.918, and 0.738 and 0.888 in men, respectively), but was poor to moderate in age subgroups (eg, Harrell's C statistic in women and men aged 85-99 years was 0.576 and 0.624 for major osteoporotic fractures, and 0.601 and 0.637 for hip fractures, respectively). Without accounting for competing risks, QFracture systematically under-predicted the risk of fracture in all models, and more so for major osteoporotic fracture than for hip fracture, and more so in older people. Accounting for competing risks, QFracture still under-predicted the risk of fracture in the whole population, but over-prediction was considerable in older age groups and in people with high comorbidities at high risk of fracture.

Conclusions

The QFracture risk prediction tool systematically under-predicted the risk of fracture (because of incomplete determination of fracture rates) and over-predicted the risk in older people and in those with more comorbidities (because of competing mortality). The use of QFracture in its current form needs to be reviewed, particularly in people at high risk of death from other causes.

Update of the fracture risk prediction tool FRAX: a systematic review of potential cohorts and analysis plan

We describe the collection of cohorts together with the analysis plan for an update of the fracture risk prediction tool FRAX with respect to current and novel risk factors. The resource comprises 2,138,428 participants with a follow-up of approximately 20 million person-years and 116,117 documented incident major osteoporotic fractures.

INTRODUCTION

The availability of the fracture risk assessment tool FRAX® has substantially enhanced the targeting of treatment to those at high risk of fracture with FRAX now incorporated into more than 100 clinical osteoporosis guidelines worldwide. The aim of this study is to determine whether the current algorithms can be further optimised with respect to current and novel risk factors.

METHODS

A computerised literature search was performed in PubMed from inception until May 17, 2019, to identify eligible cohorts for updating the FRAX coefficients. Additionally, we searched the abstracts of conference proceedings of the American Society for Bone and Mineral Research, European Calcified Tissue Society and World Congress of Osteoporosis. Prospective cohort studies with data on baseline clinical risk factors and incident fractures were eligible.

RESULTS

Of the 836 records retrieved, 53 were selected for full-text assessment after screening on title and abstract. Twelve cohorts were deemed eligible and of these, 4 novel cohorts were identified. These cohorts, together with 60 previously identified cohorts, will provide the resource for constructing an updated version of FRAX comprising 2,138,428 participants with a follow-up of approximately 20 million person-years and 116,117 documented incident major osteoporotic fractures. For each known and candidate risk factor, multivariate hazard functions for hip fracture, major osteoporotic fracture and death will be tested using extended Poisson regression. Sex- and/or ethnicity-specific differences in the weights of the risk factors will be investigated. After meta-analyses of the cohort-specific beta coefficients for each risk factor, models comprising 10-year probability of hip and major osteoporotic fracture, with or without femoral neck bone mineral density, will be computed.

CONCLUSIONS

These assembled cohorts and described models will provide the framework for an updated FRAX tool enabling enhanced assessment of fracture risk (PROSPERO (CRD42021227266)).

Is it time to consider population screening for fracture risk in postmenopausal women? A position paper from the International Osteoporosis Foundation Epidemiology/Quality of Life Working Group

The IOF Epidemiology and Quality of Life Working Group has reviewed the potential role of population screening for high hip fracture risk against well-established criteria. The report concludes that such an approach should strongly be considered in many health care systems to reduce the burden of hip fractures.

INTRODUCTION

The burden of long-term osteoporosis management falls on primary care in most healthcare systems. However, a wide and stable treatment gap exists in many such settings; most of which appears to be secondary to a lack of awareness of fracture risk. Screening is a public health measure for the purpose of identifying individuals who are likely to benefit from further investigations and/or treatment to reduce the risk of a disease or its complications. The purpose of this report was to review the evidence for a potential screening programme to identify postmenopausal women at increased risk of hip fracture.

METHODS

The approach took well-established criteria for the development of a screening program, adapted by the UK National Screening Committee, and sought the opinion of 20 members of the International Osteoporosis Foundation's Working Group on Epidemiology and Quality of Life as to whether each criterion was met (yes, partial or no). For each criterion, the evidence base was then reviewed and summarized.

RESULTS AND CONCLUSION

The report concludes that evidence supports the proposal that screening for high fracture risk in primary care should strongly be considered for incorporation into many health care systems to reduce the burden of fractures, particularly hip fractures. The key remaining hurdles to overcome are engagement with primary care healthcare professionals, and the implementation of systems that facilitate and maintain the screening program.

Reverse engineering the FRAX algorithm: Clinical insights and systematic analysis of fracture risk

The Fracture Risk Assessment Tool (FRAX) is a computational tool developed to predict the 10-year probability of hip fracture and major osteoporotic fracture based on inputs of patient characteristics, bone mineral density (BMD), and a set of seven clinical risk factors. While the FRAX tool is widely available and clinically validated, its underlying algorithm is not public. The relative contribution and necessity of each input parameter to the final FRAX score is unknown. We systematically collected hip fracture risk scores from the online FRAX calculator for osteopenic Caucasian women across 473,088 unique inputs. This dataset was used to dissect the FRAX algorithm and construct a reverse-engineered fracture risk model to assess the relative contribution of each input variable. Within the reverse-engineered model, age and T-Score were the strongest contributors to hip fracture risk, while BMI had marginal contribution. Of the clinical risk factors, parent history of fracture and ongoing glucocorticoid treatment had the largest additive effect on risk score. A generalized linear model largely recapitulated the FRAX tool with an R2 of 0.91. Observed effect sizes were then compared to a true patient population by creating a logistic regression model of the Study of Osteoporotic Fractures (SOF) cohort, which closely paralleled the effect sizes seen in the reverse-engineered fracture risk model. Analysis identified several clinically relevant observations of interest to FRAX users. The role of major osteoporotic fracture risk prediction in contributing to an indication of treatment need is very narrow, as the hip fracture risk prediction accounted for 98% of treatment indications for the SOF cohort. Removing any risk factor from the model substantially decreased its accuracy and confirmed that more parsimonious models are not ideal for fracture prediction. For women 65 years and older with a previous fracture, 98% of FRAX combinations exceeded the treatment threshold, regardless of T-score or other factors. For women age 70+ with a parent history of fracture, 99% of FRAX combinations exceed the treatment threshold. Based on these analyses, we re-affirm the efficacy of the FRAX as the best tool for fracture risk assessment and provide deep insight into the interplay between risk factors.

Predictive Performance of the FRAX Tool Calibrated for Spain vs. an Age and Sex Model: Prospective Cohort Study with 9082 Women and Men Followed for up to 8 Years

In Spain, the Fracture Risk Assessment Tool (FRAX) was adapted using studies with a small number of patients, and there are only a few external validation studies that present limitations. In this prospective cohort study, we compared the performance of FRAX and a simple age and sex model. We used data from the ESOSVAL cohort, a cohort composed of a Mediterranean population of 11,035 women and men aged 50 years and over, followed for up to 8 years, to compare the discrimination, calibration, and reclassification of FRAX calibrated for Spain and a logistic model including only age and sex as variables. We found virtually identical AUC, 83.55% for FRAX (CI 95%: 80.46, 86.63) and 84.10% for the age and sex model (CI 95%: 80.91, 87.29), and there were similar observed-to-predicted ratios. In the reclassification analyses, patients with a hip fracture that were reclassified correctly as high risk by FRAX, compared to the age and sex model, were −2.86%, using either the 3% threshold or the observed incidence, 1.54% (95%CI: −8.44, 2.72 for the 3% threshold; 95%CI: −7.68, 1.97 for the incidence threshold). Remarkably simple and inexpensive tools that are easily transferable into electronic medical record environments may offer a comparable predictive ability to that of FRAX.

Prediction of imminent fracture risk in Canadian women and men aged 45 years or older: external validation of the Fracture Risk Evaluation Model (FREM)

The Fracture Risk Evaluation Model (FREM) identifies individuals at high imminent risk of major osteoporotic fractures. We validated FREM on 74,828 individuals from Manitoba, Canada, and found significant fracture risk stratification for all FREM scores. FREM performed better than age alone but not as well as FRAX® with BMD.

INTRODUCTION

The FREM is a tool developed from Danish public health registers (hospital diagnoses) to identify individuals over age 45 years at high imminent risk of major osteoporotic fractures (MOF) and hip fracture (HF). In this study, our aim was to examine the ability of FREM to identify individuals at high imminent fracture risk in women and men from Manitoba, Canada.

METHODS

We used the population-based Manitoba Bone Mineral Density (BMD) Program registry, and identified women and men aged 45 years or older undergoing baseline BMD assessment with 2 years of follow-up data. From linked population-based data sources, we constructed FREM scores using up to 10 years of prior healthcare information.

RESULTS

The study population comprised 74,828 subjects, and during the 2 years of observation, 1612 incident MOF and 299 incident HF occurred. We found significant fracture risk stratification for all FREM scores, with AUC estimates of 0.63-0.66 for MOF for both sexes and 0.84 for women and 0.65-0.67 for men for HF. FREM performed better than age alone but not as well as FRAX® with BMD. The inclusion of physician claims data gave slightly better performance than hospitalization data alone. Overall calibration for 1-year MOF prediction was reasonable, but HF prediction was overestimated.

CONCLUSION

In conclusion, the FREM algorithm shows significant fracture risk stratification when applied to an independent clinical population from Manitoba, Canada. Overall calibration for MOF prediction was good, but hip fracture risk was systematically overestimated indicating the need for recalibration.

Bibliometric analysis of global research trends on male osteoporosis: a neglected field deserves more attention

We analyzed the knowledge structure, current status, and future directions of 3243 publications on male osteoporosis by employing bibliometric analysis. Our results indicated that Osteoporosis International was the most influential journal in this field. And the study of epidemiology and risk factors has been recognized as a hot research topic in recent years. This study also calls for more attention to be given on male osteoporosis research.

INTRODUCTION

Male osteoporosis is increasing as a serious health problem worldwide with the aging of population. However, a comprehensive understanding of the current status and future trends in this field is lacking to date. The goal of the present study was to summarize and visualize the knowledge framework, research hotspots, and emerging trends of male osteoporosis research based on the bibliometric method.

METHODS

Scientific publications regarding male osteoporosis from 1998 to 2020 were downloaded from the SCIE database. VOSviewer, CiteSpace, and online bibliometric website were used for this study. The main analyses include cooperative relationships between countries/institutions/authors, co-citation analysis of authors/journals, and co-occurrence analysis of keywords/subject categories, as well as analyses on keyword/reference bursts.

RESULTS

A total of 3243 publications with 128,751 citations were identified. Despite experiencing a period of increase in the number of publications, incentives for conducting male osteoporosis research seem to have decreased during recent years. The USA has the most prominent contributions, as reflected by most publications and the highest H-index value. Oregon Health and Science University was the most prolific institution within this domain. The most influential academic journal was Osteoporosis International. Keywords were categorized into four clusters: basic research, epidemiology and risk factors, diagnostic studies, treatment and fracture prevention. Burst keyword detection suggested that the following research directions including "obesity," "zoledronic acid," "DXA," "inflammation," "fall," "microarchitecture," and "sarcopenia" remain research hotspots in the near future and deserve our further attention.

CONCLUSIONS

This is the first bibliometric analysis that provides a comprehensive overview of male osteoporosis research, which may provide helpful references for investigators to further explore hot issues in this field.

Impact of population-based or targeted BMD interventions on fracture incidence

In a simulated population of older women, we demonstrate that an upward shift in the population distribution of BMD by approximately 0.3SD may decrease the risk of incident fractures to the same extent as an intervention targeted to those with T-score less than -2.5.

INTRODUCTION

To investigate the impact of population level or targeted alterations to BMD on the incidence of fractures.

METHODS

We used a simulated cohort of 49,242 women with age and body mass index distribution from the UK, and prevalence of other clinical risk factors based on European FRAX® cohorts. Using FRAX probabilities of major osteoporotic fracture (MOF: hip, clinical vertebral, distal forearm, proximal humerus) and hip fracture, calculated with femoral neck BMD, we determined the expected number of fractures over 10 years, stratified by 10-year age band from 50 years. We then investigated the effect of (i) uplifting all individuals with T-score below -2.5 to be exactly -2.5 (high-risk strategy) and (ii) shifting the entire BMD distribution upwards (population strategy).

RESULTS

Overall, the high-risk strategy prevented 573 MOF including 465 hip fractures. Moving the BMD T-score distribution upward by 0.27SD gave an equivalent reduction in numbers of MOF; for hip fractures prevented, this was 0.35SD. A global upward 0.25SD BMD shift prevented 524 MOF including 354 hip fractures, with corresponding figures for an increase of 0.5SD being 973 MOF prevented and 640 hip fractures prevented. The ratio of hip fracture to MOF prevented differed by the two approaches, such that for the high-risk strategy, the ratio was 0.81, and for the population strategy was 0.68 (0.25SD BMD uplift) and 0.66 (0.5SD BMD uplift). The numbers of fractures prevented by the high-risk strategy increased with age. In contrast, the age-related increase in numbers of fractures prevented with the population strategy rose with age, but peaked in the 70-79-year age band and declined thereafter.

CONCLUSIONS

Both strategies reduced the numbers of expected incident fractures, with contrasting relative impacts by age and fracture site. Whilst the current analysis used UK/European anthropometric/risk factor distributions, further analyses calibrated to the distributions in other settings globally may be readily undertaken. Overall, these findings support the investigation of both population level interventions and those targeted at high fracture risk groups.

Development and validation of the fracture risk scale home care (FRS-HC) that predicts one-year incident fracture: an electronic record-linked longitudinal cohort study

BACKGROUND

Fractures have dire consequences including pain, immobility, and death. People receiving home care are at higher risk for fractures than the general population. Yet, current fracture risk assessment tools require additional testing and assume a 10-year survival rate, when many die within one year. Our objectives were to develop and validate a scale that predicts one-year incident hip fracture using the home care resident assessment instrument (RAI-HC).

METHODS

This is a retrospective cohort study of linked population data. People receiving home care in Ontario, Canada between April 1st, 2011 and March 31st, 2015 were included. Clinical data were obtained from the RAI-HC which was linked to the Discharge Abstract Database and National Ambulatory Care Reporting System to capture one-year incident hip fractures. Seventy-five percent (n = 238,011) of the sample were randomly assigned to a derivation and 25% (n = 79,610) to a validation sample. A decision tree was created with the derivation sample using known fracture risk factors. The final nodes of the decision tree were collapsed into 8 risk levels and logistic regression was performed to determine odds of having a fracture for each level. c-Statistics were calculated to compare the discriminative properties of the full, derivation, and validation samples.

RESULTS

Approximately 60% of the sample were women and 53% were 80 years and older. A total of 11,526 (3.6%) fractures were captured over the 1-year time period. Of these, 5057 (43.9%) were hip fractures. The proportion who experienced a hip fracture in the next year ranged from 0.3% in the lowest risk level to 5.2% in the highest risk level. People in the highest risk level had 18.8 times higher odds (95% confidence interval, 14.6 to 24.3) of experiencing a hip fracture within one year than those in the lowest. c-Statistics were similar for the full (0.658), derivation (0.662), and validation (0.645) samples.

CONCLUSIONS

The FRS-HC predicts hip fracture over one year and should be used to guide clinical care planning for home care recipients at high risk for fracture. Our next steps are to develop a fracture risk clinical assessment protocol to link treatment recommendations with identified fracture risk.

New Developments in Fracture Risk Assessment for Current Osteoporosis Reports

PURPOSE OF REVIEW

Identifying individuals at high fracture risk can be used to target those likely to derive the greatest benefit from treatment. This narrative review examines recent developments in using specific risk factors used to assess fracture risk, with a focus on publications in the last 3 years.

RECENT FINDINGS

There is expanding evidence for the recognition of individual clinical risk factors and clinical use of composite scores in the general population. Unfortunately, enthusiasm is dampened by three pragmatic randomized trials that raise questions about the effectiveness of widespread population screening using clinical fracture prediction tools given suboptimal participation and adherence. There have been refinements in risk assessment in special populations: men, patients with diabetes, and secondary causes of osteoporosis. New evidence supports the value of vertebral fracture assessment (VFA), high resolution peripheral quantitative CT (HR-pQCT), opportunistic screening using CT, skeletal strength assessment with finite element analysis (FEA), and trabecular bone score (TBS). The last 3 years have seen important developments in the area of fracture risk assessment, both in the research setting and translation to clinical practice. The next challenge will be incorporating these advances into routine work flows that can improve the identification of high risk individuals at the population level and meaningfully impact the ongoing crisis in osteoporosis management.

Evaluation of a Panel of MicroRNAs that Predicts Fragility Fracture Risk: A Pilot Study

The assessment of fragility fracture risk based on bone densitometry and FRAX°, although commonly used, has shown some limitations. MicroRNAs (miRNAs) are promising biomarkers known to regulate post-transcriptional gene expression. Many studies have already shown that microRNAs are involved in bone homeostasis by modulating osteoblast and osteoclast gene expression. In this pilot study, we investigated the ability of an miRNA panel (namely, the OsteomiR° score) to predict fragility fracture risk in older people. miRNAs were extracted from the sera of 17 persons who developed a fracture within 3 years of collecting the serum and 16 persons who did not experience fractures in the same period. Nineteen miRNAs known to be involved in bone homeostasis were assessed, and 10 miRNAs were employed to calculate the OsteomiR° score. We found a trend towards higher OsteomiR° scores in individuals who experienced fractures compared to control subjects. The most suitable cut-off that maximized sensitivity and specificity was determined by ROC curve analysis, and a positive predictive value of 68% and a sensitivity of 76% were obtained. The OsteomiR° score was higher in osteopenic and osteoporotic subjects compared to subjects with a normal T score. Additionally, the OsteomiR° score predicted more fracture events than the recommended "need-to-treat" thresholds based on FRAX° 10-year probability. miRNAs reflect impairments in bone homeostasis several years before the occurrence of a fracture. The OsteomiR° score seems to be a promising miRNA panel for fragility fracture risk prediction and might have added value compared to FRAX°. Given the limited cohort size, further studies should be dedicated to validating the OsteomiR° score.

Performance of FRAX in Predicting Fractures in US Postmenopausal Women with Varied Race and Genetic Profiles

BACKGROUND

Whether the Fracture Risk Assessment Tool (FRAX) performed differently in estimating the 10-year fracture probability in women of different genetic profiling and race remained unclear.

METHODS

The genomic data in the Women's Health Initiative (WHI) study was analyzed (n = 23,981). The genetic risk score (GRS) was calculated from 14 fracture-associated single nucleotide polymorphisms (SNPs) for each participant. FRAX without bone mineral density (BMD) was used to estimate fracture probability.

RESULTS

FRAX significantly overestimated the risk of major osteoporotic fracture (MOF) in the WHI study. The most significant overestimation was observed in women with low GRS (predicted/observed ratio (POR): 1.61, 95% CI: 1.45-1.79) specifically Asian women (POR: 3.5, 95% CI 2.48-4.81) and in African American women (POR: 2.59, 95% CI: 2.33-2.87). Compared to the low GRS group, the 10-year probability of MOF adjusted for the FRAX score was 21% and 30% higher in the median GRS group and high GRS group, respectively. Asian, African American, and Hispanic women respectively had a 78%, 76%, and 56% lower hazard than Caucasian women after the FRAX score was adjusted. The results were similar for hip fractures.

CONCLUSIONS

Our study suggested the FRAX performance varies significantly by both genetic profile and race in postmenopausal women.

How well do the FRAX (Australia) and Garvan calculators predict incident fractures? Data from the Geelong Osteoporosis Study

This study reports that both FRAX and Garvan calculators underestimated fractures in Australian men and women, particularly in those with osteopenia or osteoporosis. Major osteoporotic fractures were poorly predicted, while both calculators performed acceptably well for hip fractures.

INTRODUCTION

This study assessed the ability of the FRAX (Australia) and Garvan calculators to predict fractures in Australian women and men.

METHODS

Women (n = 809) and men (n = 821) aged 50-90 years, enrolled in the Geelong Osteoporosis Study, were included. Fracture risk was estimated using FRAX and Garvan calculators with and without femoral neck bone mineral density (BMD) (FRAX_BMD, FRAX_noBMD, Garvan_BMD, Garvan_noBMD). Incident major osteoporotic (MOF), fragility, and hip fractures over the following 10 years were verified radiologically. Differences between observed and predicted numbers of fractures were assessed using a chi-squared test. Diagnostics indexes were calculated.

RESULTS

In women, 115 MOF, 184 fragility, and 42 hip fractures occurred. For men, there were 73, 109, and 17 fractures, respectively. FRAX underestimated MOFs, regardless of sex or inclusion of BMD. FRAX accurately predicted hip fractures, except in women with BMD (20 predicted, p = 0.004). Garvan underestimated fragility fractures except in men using BMD (88 predicted, p = 0.109). Garvan accurately predicted hip fractures except for women without BMD (12 predicted, p < 0.001). Fractures were underestimated primarily in the osteopenia and osteoporosis groups; MOFs in the normal BMD group were only underestimated by FRAX_BMD and fragility fractures by Garvan_noBMD, both in men. AUROCs were not different between scores with and without BMD, except for fragility fractures predicted by Garvan in women (0.696, 95% CI 0.652-0.739 and 0.668, 0.623-0.712, respectively, p = 0.008) and men, which almost reached significance (0.683, 0.631-0.734, and 0.667, 0.615-0.719, respectively, p = 0.051). Analyses of sensitivity and specificity showed overall that MOFs and fragility fractures were poorly predicted by both FRAX and Garvan, while hip fractures were acceptably predicted.

CONCLUSIONS

Overall, the FRAX and Garvan calculators underestimated MOF and fragility fractures, particularly in individuals with osteopenia or osteoporosis. Hip fractures were predicted better by both calculators. AUROC analyses suggest that Garvan_BMD performed better than Garvan_noBMD for prediction of fragility fractures.

European guidance for the diagnosis and management of osteoporosis in postmenopausal women

Guidance is provided in a European setting on the assessment and treatment of postmenopausal women at risk from fractures due to osteoporosis.

INTRODUCTION

The International Osteoporosis Foundation and European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis published guidance for the diagnosis and management of osteoporosis in 2013. This manuscript updates these in a European setting.

METHODS

Systematic reviews were updated.

RESULTS

The following areas are reviewed: the role of bone mineral density measurement for the diagnosis of osteoporosis and assessment of fracture risk; general and pharmacological management of osteoporosis; monitoring of treatment; assessment of fracture risk; case-finding strategies; investigation of patients; health economics of treatment. The update includes new information on the evaluation of bone microstructure evaluation in facture risk assessment, the role of FRAX® and Fracture Liaison Services in secondary fracture prevention, long-term effects on fracture risk of dietary intakes, and increased fracture risk on stopping drug treatment.

CONCLUSIONS

A platform is provided on which specific guidelines can be developed for national use.

A brief history of FRAX

This paper reviews the research programme that went into the development of FRAX® and its impact in the 10 years since its release in 2008.

INTRODUCTION

Osteoporosis is defined on the measurement of bone mineral density though the clinical consequence is fracture. The sensitivity of bone mineral density measurements for fracture prediction is low, leading to the development of FRAX to better calculate the likelihood of fracture and target anti-osteoporosis treatments.

METHODS

The method used in this paper is literature review.

RESULTS

FRAX, developed over an 8-year period, was launched in 2008. Since the launch of FRAX, models have been made available for 64 countries and in 31 languages covering more than 80% of the world population.

CONCLUSION

FRAX provides an advance in fracture risk assessment and a reference technology platform for future improvements in performance characteristics.

Falls Predict Fractures Independently of FRAX Probability: A Meta-Analysis of the Osteoporotic Fractures in Men (MrOS) Study

Although prior falls are a well-established predictor of future fracture, there is currently limited evidence regarding the specific value of falls history in fracture risk assessment relative to that of other clinical risk factors and bone mineral density (BMD) measurement. We therefore investigated, across the three Osteoporotic Fractures in Men (MrOS) Study cohorts, whether past falls predicted future fracture independently of FRAX and whether these associations varied with age and follow-up time. Elderly men were recruited from MrOS Sweden, Hong Kong, and USA. Baseline data included falls history (over the preceding 12 months), clinical risk factors, BMD at femoral neck, and calculated FRAX probabilities. An extension of Poisson regression was used to investigate the associations between falls, FRAX probability, and incident fracture, adjusting for age, time since baseline, and cohort in base models; further models were used to investigate interactions with age and follow-up time. Random-effects meta-analysis was used to synthesize the individual country associations. Information on falls and FRAX probability was available for 4365 men in USA (mean age 73.5 years; mean follow-up 10.8 years), 1823 men in Sweden (mean age 75.4 years; mean follow-up 8.7 years), and 1669 men in Hong Kong (mean age 72.4 years; mean follow-up 9.8 years). Rates of past falls were similar at 20%, 16%, and 15%, respectively. Across all cohorts, past falls predicted incident fracture at any site (hazard ratio [HR] = 1.69; 95% confidence interval [CI] 1.49, 1.90), major osteoporotic fracture (MOF) (HR = 1.56; 95% CI 1.33, 1.83), and hip fracture (HR = 1.61; 95% CI 1.27, 2.05). Relationships between past falls and incident fracture remained robust after adjustment for FRAX probability: adjusted HR (95% CI) any fracture: 1.63 (1.45, 1.83); MOF: 1.51 (1.32, 1.73); and hip: 1.54 (1.21, 1.95). In conclusion, past falls predicted incident fracture independently of FRAX probability, confirming the potential value of falls history in fracture risk assessment. © 2017 The Authors. Journal of Bone and Mineral Research Published by Wiley Periodicals Inc.

FRAX Tool in Brazil: an integrative literature review following validation

Abstract The present article is an integrative review the objective of which was to assess research carried out with the FRAX tool in Brazil following its validation, and describe the conclusions drawn. Two databases were used to select the articles (the Capes Portal and the Virtual Health Library), and the sample of this review was the only four articles published in Brazil relating to the FRAX tool following its validation in May 2013. After analyzing the articles, the results demonstrated that despite some limitations the FRAX Tool can be used to reduce the prevalence of fractures due to its simplicity of use, with an emphasis on prediction and orientation, allowing early and safe therapeutic decision-making.

Putting the Brakes on Breaks: Osteoporosis Screening and Fracture Prevention

Introduction

This prospective study sought to implement a screening tool to identify and risk stratify at-risk patients for osteoporosis and evaluate patient knowledge of osteoporosis and fragility fractures in an orthopedic trauma clinic affiliated with a level 1 trauma academic center.

Methods

Of 297 eligible patients, 291 were screened and risk stratified. Patients completed an osteoporosis screening questionnaire and were risk stratified. Lifestyle advice was given to patients at low fracture risk. A dual-energy X-ray absorptiometry scan was ordered for patients at intermediate fracture risk. A referral was initiated for treatment to a bone health specialist in high fracture risk patients. Twenty patients completed a knowledge-based pretest/posttest.

Results

A total of 291 patients were screened, which represented 97.7% of patients over the age of 50. Of those patients, 165 (56.7%) patients met criteria for further osteoporosis evaluation as they were considered either intermediate or high risk for future fractures. One hundred thirty-six (82.4%) patients were referred for bone mineral density evaluation. For the knowledge-based evaluation portion, patients had a 33% gain in knowledge (P = .0004). The largest knowledge deficit identified pertained to osteoporosis risk factors and lifestyle management.

Discussion

The use of an osteoporosis screening questionnaire in the orthopedic trauma clinic produced clinically significant improvement in identification of at-risk patients. A lack of knowledge regarding osteoporosis and fragility fractures was found to exist among these patients.

Conclusion

The implementation of an osteoporosis screening tool to identify, risk stratify, and treat patients with osteoporosis and related fragility fractures can be successfully integrated into a busy clinical practice.

Validation of FRAX and the impact of self-reported falls among elderly in a general population: the HUNT study, Norway

Fracture Risk Assessment Tool (FRAX) without bone mineral density (BMD) for hip fracture prediction was validated in a Norwegian population 50-90 years. Fracture risk increased with higher FRAX score, and the observed number of hip fractures agreed well with the predicted number, except for the youngest and oldest men. Self-reported fall was an independent risk factor for fracture in women.

INTRODUCTION

The primary aim was to validate FRAX without BMD for hip fracture prediction in a Norwegian population of men and women 50-90 years. Secondary, to study whether information of falls could improve prediction of fractures in the subgroup aged 70-90 years.

METHODS

Data were obtained from the third survey of the Nord-Trøndelag Health Study (HUNT3), the fracture registry in Nord-Trøndelag, and the Norwegian Prescription Database (NorPD), including 15,432 women and 13,585 men. FRAX hip without BMD was calculated, and hip fractures were registered for a median follow-up of 5.2 years. The number of estimated and observed fractures was assessed, ROC curves with area under the curve (AUC), and Cox regression analyses. For the group aged 70-90 years, self-reported falls the last year before HUNT3 were included in the Cox regression model.

RESULTS

The risk of fracture increased with higher FRAX score. When FRAX groups were categorized in a 10-year percentage risk for hip fracture as follows, <4, 4-7.9, 8-11.9, and ≥12%, the hazard ratio (HR) for hip fracture between the lowest and the highest group was 17.80 (95% CI: 12.86-24.65) among women and 23.40 (13.93-39.30) in men. Observed number of hip fractures agreed quite well with the predicted number, except for the youngest and oldest men. AUC was 0.81 (0.78-0.83) for women and 0.79 (0.76-0.83) for men. Self-reported fall was an independent risk factor for fracture in women (HR 1.64, 1.20-2.24), and among men, this was not significant (1.09, 0.65-1.83).

CONCLUSIONS

FRAX without BMD predicted hip fracture reasonably well. In the age group 70-90 years, falls seemed to imply an additional risk among women.

Do we need bone mineral density to estimate osteoporotic fracture risk? A 10-year prospective multicentre validation study

Objective

Evaluate the performance of FRAX®, with and without bone mineral densitometry (BMD), in predicting the occurrence of fragility fractures over 10 years.

Methods

Participants aged ≥40 years at baseline, with a complete set of data and a minimum of 8.5 years of follow-up were identified from three cohorts (n=2626). Ten-year fracture risk at baseline were estimated with FRAX® and assessed by comparison with observed fractures and receiver operating characteristic analysis.

Results

During a mean (SD) follow-up of 9.12 (1.5) years, 178 participants suffered a major osteoporotic (MOP) fracture and 28 sustained a hip fracture. The predictive performance of FRAX® was superior to that of BMD alone for both MOP and hip fractures. The area under the curve (AUC) of FRAX® without BMD was 0.76 (95% CI 0.72 to 0.79) for MOP fractures and 0.78 (95% CI 0.69 to 0.86) for hip fractures. No significant improvements were found when BMD was added to clinical variables to predict either MOP (0.78, 95% CI 0.74 to 0.82, p=0.25) or hip fractures (0.79, 95% CI 0.69 to 0.89, p=0.72).

AUCs for FRAX® (with and without BMD) were greater for men than for women. FRAX®, with and without BMD, tended to underestimate the number of MOP fractures and to overestimate the number of hip fractures in females. In men, the number of observed fractures were within the 95% CI of the number predicted, both with and without BMD.

Conclusion

FRAX® without BMD provided good fracture prediction. Adding BMD to FRAX® did not improve the performance of the tool in the general population.

Clinical Utility of Using Lumbar Spine Trabecular Bone Score to Adjust Fracture Probability: The Manitoba BMD Cohort

Decreased lumbar spine trabecular bone score (TBS), a dual-energy X-ray absorptiometry (DXA)-derived image texture measurement, is a risk factor for major osteoporotic fracture (MOF) and hip fracture (HF) independent of 10-year fracture probability estimated using FRAX. We determined how often applying the TBS adjustment to fracture probability altered treatment qualification. Using a population-based registry containing all clinical DXA results for Manitoba, Canada, we identified 34,316 women with baseline spine and hip DXA, FRAX-based fracture probability measurements (computed with femoral neck bone mineral density), lumbar spine TBS, and minimum 5 years of observation (mean 8.7 years). Population-based health services data were used to identify incident non-traumatic MOF and HF in 3503 and 945 women, respectively. Baseline MOF and HF probabilities were estimated using FRAX before and after applying the TBS adjustment. Risk recategorization was assessed using net reclassification improvement (NRI) for individual FRAX-based intervention criteria and three national clinical practice guidelines (CPGs) (US National Osteoporosis Foundation, Osteoporosis Canada, and UK National Osteoporosis Guideline Group). Overall, proportions of women reclassified with the TBS adjustment to FRAX were small (less than 5%) with more than 90% of the reclassification occurring close to the intervention threshold. For women close to an intervention cut-off reclassification, rates ranged from 9.0% to 17.9% and were <1% otherwise. There was a small but significant improvement in overall NRI for all individual FRAX-based intervention criteria (range 0.007 to 0.018) and all three national CPGs (range 0.008 to 0.011). NRI was larger in women below age 65 years (up to 0.056 for hip fracture). In summary, a small but significant improvement in MOF and HF risk assessment was found by using lumbar spine TBS to adjust FRAX probability. An improvement in risk reclassification was observed for CPGs from three different countries, with almost all of the benefit found in individuals close to an intervention threshold. © 2017 American Society for Bone and Mineral Research. © 2017 American Society for Bone and Mineral Research.

Overview of Fracture Prediction Tools

The characterization of risk factors for fracture that contribute significantly to fracture risk, over and above that provided by the bone mineral density, has stimulated the development of risk assessment tools. The more adequately evaluated tools, all available online, include the FRAX® tool, the Garvan fracture risk calculator and, in the United Kingdom only, QFracture®. Differences in the input variables, output, and model construct give rise to marked differences in the computed risks from each calculator. Reasons for the differences include the derivation of fracture probability (FRAX) rather than incidence (Garvan and QFracture), limited calibration (Garvan), and inappropriate source information (QFracture). These differences need to be taken into account in the evaluation of assessment guidelines.

Worldwide Fracture Prediction

The substantial increase in the burden of non-communicable diseases in general and osteoporosis in particular, necessitates the establishment of efficient and targeted diagnosis and treatment strategies. This chapter reviews and compares different tools for osteoporosis screening and diagnosis; it also provides an overview of different treatment guidelines adopted by countries worldwide. While access to dual-energy X-ray absorptiometry to measure bone mineral density (BMD) is limited in most areas in the world, the introduction of risk calculators that combine risk factors, with or without BMD, have resulted in a paradigm shift in osteoporosis screening and management. To-date, forty eight risk assessment tools that allow risk stratification of patients are available, however only few are externally validated and tested in a population-based setting. These include Osteoporosis Self-Assessment Tool; Osteoporosis Risk Assessment Instrument; Simple Calculated Osteoporosis Risk Estimation; Canadian Association of Radiologists and Osteoporosis Canada calculator; Fracture Risk Assessment Calculator (FRAX); Garvan; and QFracture. These tools vary in the number of risk factors incorporated. We present a detailed analysis of the development, characteristics, validation, performance, advantages and limitations of these tools. The World Health Organization proposes a dual-energy X-ray absorptiometry-BMD T-score ≤ -2.5 as an operational diagnostic threshold for osteoporosis, and many countries have also adopted this cut-off as an intervention threshold in their treatment guidelines. With the introduction of the new fracture assessment calculators, many countries chose to include fracture risk as one of the major criteria to initiate osteoporosis treatment. Of the 52 national guidelines identified in 36 countries, 30 included FRAX derived risk in their intervention threshold and 22 were non-FRAX based.  No universal tool or guideline approach will address the needs of all countries worldwide.  Osteoporosis screening and management guidelines are best tailored according to the needs and resources of individual counties. While few countries have succeeded in generating valuable epidemiological data on osteoporotic fractures, to validate their risk calculators and base their guidelines, many have yet to find the resources to assess variations and secular trends in fractures, the performance of various calculators, and ultimately adopt the most convenient care pathway algorithms.

Fracture Prediction With Modified-FRAX in Older HIV-Infected and Uninfected Men

BACKGROUND

FRAX is a validated, computer-based clinical fracture risk calculator that estimates the 10-year risk of major osteoporotic (clinical spine, forearm, hip, or shoulder) fracture, and hip fracture alone. It is widely used for decision making in fracture prevention, but it may underestimate the risk in HIV-infected individuals. Some experts recommend considering HIV as a cause of secondary osteoporosis when calculating FRAX in HIV-infected individuals.

METHODS

From the Veterans Aging Cohort Study Virtual Cohort, we included 24,451 HIV-infected and uninfected men aged 50-70 years with complete data in the year 2000 to approximate all but 2 factors (ie, history of secondary osteoporosis and parental hip fracture) for modified-FRAX calculation without bone density and 10-year observational data for incident fragility fracture. The accuracy of the modified-FRAX calculation was compared by the observed/estimated (O/E) ratios of fracture by HIV status.

RESULTS

The accuracy of modified-FRAX was less for HIV-infected [O/E = 1.62, 95% confidence interval (CI) 1.45 to 1.81] than uninfected men (O/E = 1.29, 95% CI: 1.19 to 1.40), but improved when HIV was included as a cause of secondary osteoporosis (O/E = 1.20, 95% CI: 1.08 to 1.34). However, only 3%-6% of men with incident fractures were correctly identified by the modified-FRAX using accepted FRAX thresholds for pharmacologic therapy.

CONCLUSIONS

Modified-FRAX underestimated the fracture rates more in older HIV-infected than in otherwise similar uninfected men. The accuracy improved when HIV was included as a cause of secondary osteoporosis, but it still performed poorly for case finding. Further studies are necessary to determine how to use FRAX or define an HIV-specific index to risk stratify for screening and treatment in older HIV-infected individuals.

External validation and comparison of three prediction tools for risk of osteoporotic fractures using data from population based electronic health records: retrospective cohort study

OBJECTIVE

 To directly compare the performance and externally validate the three most studied prediction tools for osteoporotic fractures-QFracture, FRAX, and Garvan-using data from electronic health records.

DESIGN

 Retrospective cohort study.

SETTING

 Payer provider healthcare organisation in Israel.

PARTICIPANTS

 1 054 815 members aged 50 to 90 years for comparison between tools and cohorts of different age ranges, corresponding to those in each tools' development study, for tool specific external validation.

MAIN OUTCOME MEASURE

 First diagnosis of a major osteoporotic fracture (for QFracture and FRAX tools) and hip fractures (for all three tools) recorded in electronic health records from 2010 to 2014. Observed fracture rates were compared to probabilities predicted retrospectively as of 2010.

RESULTS

 The observed five year hip fracture rate was 2.7% and the rate for major osteoporotic fractures was 7.7%. The areas under the receiver operating curve (AUC) for hip fracture prediction were 82.7% for QFracture, 81.5% for FRAX, and 77.8% for Garvan. For major osteoporotic fractures, AUCs were 71.2% for QFracture and 71.4% for FRAX. All the tools underestimated the fracture risk, but the average observed to predicted ratios and the calibration slopes of FRAX were closest to 1. Tool specific validation analyses yielded hip fracture prediction AUCs of 88.0% for QFracture (among those aged 30-100 years), 81.5% for FRAX (50-90 years), and 71.2% for Garvan (60-95 years).

CONCLUSIONS

 Both QFracture and FRAX had high discriminatory power for hip fracture prediction, with QFracture performing slightly better. This performance gap was more pronounced in previous studies, likely because of broader age inclusion criteria for QFracture validations. The simpler FRAX performed almost as well as QFracture for hip fracture prediction, and may have advantages if some of the input data required for QFracture are not available. However, both tools require calibration before implementation.

A systematic review of intervention thresholds based on FRAX : A report prepared for the National Osteoporosis Guideline Group and the International Osteoporosis Foundation

This systematic review identified assessment guidelines for osteoporosis that incorporate FRAX. The rationale for intervention thresholds is given in a minority of papers. Intervention thresholds (fixed or age-dependent) need to be country-specific.

INTRODUCTION

In most assessment guidelines, treatment for osteoporosis is recommended in individuals with prior fragility fractures, especially fractures at spine and hip. However, for those without prior fractures, the intervention thresholds can be derived using different methods. The aim of this report was to undertake a systematic review of the available information on the use of FRAX® in assessment guidelines, in particular the setting of thresholds and their validation.

METHODS

We identified 120 guidelines or academic papers that incorporated FRAX of which 38 provided no clear statement on how the fracture probabilities derived are to be used in decision-making in clinical practice. The remainder recommended a fixed intervention threshold (n = 58), most commonly as a component of more complex guidance (e.g. bone mineral density (BMD) thresholds) or an age-dependent threshold (n = 22). Two guidelines have adopted both age-dependent and fixed thresholds.

RESULTS

Fixed probability thresholds have ranged from 4 to 20 % for a major fracture and 1.3-5 % for hip fracture. More than one half (39) of the 58 publications identified utilised a threshold probability of 20 % for a major osteoporotic fracture, many of which also mention a hip fracture probability of 3 % as an alternative intervention threshold. In nearly all instances, no rationale is provided other than that this was the threshold used by the National Osteoporosis Foundation of the USA. Where undertaken, fixed probability thresholds have been determined from tests of discrimination (Hong Kong), health economic assessment (USA, Switzerland), to match the prevalence of osteoporosis (China) or to align with pre-existing guidelines or reimbursement criteria (Japan, Poland). Age-dependent intervention thresholds, first developed by the National Osteoporosis Guideline Group (NOGG), are based on the rationale that if a woman with a prior fragility fracture is eligible for treatment, then, at any given age, a man or woman with the same fracture probability but in the absence of a previous fracture (i.e. at the 'fracture threshold') should also be eligible. Under current NOGG guidelines, based on age-dependent probability thresholds, inequalities in access to therapy arise especially at older ages (≥70 years) depending on the presence or absence of a prior fracture. An alternative threshold using a hybrid model reduces this disparity.

CONCLUSION

The use of FRAX (fixed or age-dependent thresholds) as the gateway to assessment identifies individuals at high risk more effectively than the use of BMD. However, the setting of intervention thresholds needs to be country-specific.

SIGN Guidelines for Scotland: BMD Versus FRAX Versus QFracture

Scottish Intercollegiate Guidelines Network (SIGN) recently issued guidance on the management of osteoporosis and the prevention of fragility fractures. The aim of this paper was to critically review the guidance. The SIGN guidance utilises risk factors for fracture as an initial step for assessment, but recommends treatment only in individuals with a T-score of -2.5. There are many problems with the sole use of BMD as the sole gateway to treatment. Moreover, the assessment tools to determine risk (FRAX or QFracture) are not designed to detect osteoporosis but rather fracture risk. Whereas SIGN assumes that FRAX overestimates fracture probability, there are compelling reasons to believe that the disparity is related to the inadequate calibration of QFracture. The disparities make the use of a single threshold for BMD testing problematic. The SIGN guidance for men at high risk of fracture provides a set of confused and inconsistent recommendations that are in direct conflict with regulatory authorizations and is likely to increase further the large treatment gap in men. For women, the number of women eligible for treatment (i.e. with osteoporosis) is 81,700 with the use of FRAX but only 12,300 with QFracture representing 8.2 and 1.2 % of the total population at risk, respectively. We conclude that serious problems with the SIGN guidance preclude its implementation.

A Meta-Analysis of Trabecular Bone Score in Fracture Risk Prediction and Its Relationship to FRAX

Trabecular bone score (TBS) is a gray-level textural index of bone microarchitecture derived from lumbar spine dual-energy X-ray absorptiometry (DXA) images. TBS is a bone mineral density (BMD)-independent predictor of fracture risk. The objective of this meta-analysis was to determine whether TBS predicted fracture risk independently of FRAX probability and to examine their combined performance by adjusting the FRAX probability for TBS. We utilized individual-level data from 17,809 men and women in 14 prospective population-based cohorts. Baseline evaluation included TBS and the FRAX risk variables, and outcomes during follow-up (mean 6.7 years) comprised major osteoporotic fractures. The association between TBS, FRAX probabilities, and the risk of fracture was examined using an extension of the Poisson regression model in each cohort and for each sex and expressed as the gradient of risk (GR; hazard ratio per 1 SD change in risk variable in direction of increased risk). FRAX probabilities were adjusted for TBS using an adjustment factor derived from an independent cohort (the Manitoba Bone Density Cohort). Overall, the GR of TBS for major osteoporotic fracture was 1.44 (95% confidence interval [CI] 1.35-1.53) when adjusted for age and time since baseline and was similar in men and women (p > 0.10). When additionally adjusted for FRAX 10-year probability of major osteoporotic fracture, TBS remained a significant, independent predictor for fracture (GR = 1.32, 95% CI 1.24-1.41). The adjustment of FRAX probability for TBS resulted in a small increase in the GR (1.76, 95% CI 1.65-1.87 versus 1.70, 95% CI 1.60-1.81). A smaller change in GR for hip fracture was observed (FRAX hip fracture probability GR 2.25 vs. 2.22). TBS is a significant predictor of fracture risk independently of FRAX. The findings support the use of TBS as a potential adjustment for FRAX probability, though the impact of the adjustment remains to be determined in the context of clinical assessment guidelines. © 2015 American Society for Bone and Mineral Research.

The accuracy of osteoporotic fracture risk prediction tools: a systematic review and meta-analysis

OBJECTIVES

To identify and synthesise the best available evidence on the accuracy of the currently available tools for predicting fracture risk.

METHODS

We systematically searched PubMed MEDLINE, Embase and Cochrane databases to 2014. Two reviewers independently selected articles, collected data from studies, and carried out a hand search of the references of the included studies. The Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS) checklist was used, and the primary outcome was the area under the curve (AUC) and 95% CIs, obtained from receiver operating characteristic (ROC) analyses. We excluded tools if they had not been externally validated or were designed for specific disease populations. Random effects meta-analyses were performed with the selected tools.

RESULTS

Forty-five studies met inclusion criteria, corresponding to 13 different tools. Only three tools had been tested more than once in a population-based setting: FRAX (26 studies in 9 countries), GARVAN (6 studies in 3 countries) and QFracture (3 studies in the UK, 1 also including Irish participants). Twenty studies with these three tools were included in a total of 17 meta-analyses (for hip or major osteoporotic fractures; men or women; with or without bone mineral density).

CONCLUSIONS

Most of the 13 tools are feasible in clinical practice. FRAX has the largest number of externally validated and independent studies. The overall accuracy of the different tools is satisfactory (>0.70), with QFracture reaching 0.89 (95% CI 0.88 to 0.89). Significant methodological limitations were observed in many studies, suggesting caution when comparing tools based solely on the AUC.