Clinical Publications

Key findings

1. Overall, 64% of fragility fracture patients assessed by this FLS were judged to have an indication for osteoporosis treatment (448/705), including 71% after major osteoporotic fracture (311/438) and 51% after non-major osteoporotic fracture (137/267).
2. The Number Needed to Screen (NNS) was low in both groups, showing that FLS assessment is efficient beyond traditional major fracture categories: 1.41 for major osteoporotic fractures and 1.95 for non-major fractures.
3. Among patients with treatment indication, bone mineral density and fracture risk profiles were similar across fracture categories, suggesting that restricting FLS to major osteoporotic fractures may miss many patients who could benefit from secondary fracture prevention.

Summary

This study examined 705 adults aged 50 years or older who were assessed in a Fracture Liaison Service (FLS) at Skaraborg Hospital in Skövde, Sweden, after a recent fracture. Patients with both major osteoporotic fractures (MOF) and non-major osteoporotic fractures (non-MOF) were included. MOF were defined as fractures of the hip, vertebrae, proximal humerus, wrist or pelvis. Non-MOF were defined as fractures of the elbow, clavicle, rib, knee, ankle, and a residual “other” category. The ICD-10 codes used to classify patients into these two groups are provided in the supplementary information available here (Tables S2A and S2B).

The investigators compared bone mineral density, clinical risk factors, FRAX® probability, vertebral fracture assessment, trabecular bone score and physician-assessed osteoporosis treatment indication to determine whether non-major fractures also identify patients who may benefit from treatment.

Overall, almost two-thirds of patients (64%) assessed through the FLS had an indication for osteoporosis treatment. As expected, treatment indication was more common after major osteoporotic fracture (71%) than after non-major osteoporotic fracture (51%), but it was still present in just over half of the non-MOF group. The Number Needed to Screen (NNS) was low in both groups, and among patients judged to need treatment, bone mineral density and risk profiles were similar across fracture categories. The findings suggest that limiting FLS case-finding to major osteoporotic fractures may miss a substantial number of patients who would benefit from osteoporosis assessment and secondary fracture prevention.

The full publication is available here as an open-access article.

Key findings

1. Overall, 60% of patients who sustained a fragility fracture while already taking bone-active medication were advised to change treatment after specialist review (500/828), while 34% were advised to remain on the same medication (281/828) and 3% to discontinue treatment (22/828).
2. Among those advised to change treatment, most were switched to denosumab (418/500; 84%), with much smaller proportions switched to teriparatide (26/500; 5%) or risedronate (12/500; 2%).
3. Older age was associated with a higher likelihood of treatment change recommendation, whereas prior denosumab use was associated with a much lower likelihood of switching, suggesting that specialist decision-making reflects both clinical judgement and jurisdiction-specific reimbursement constraints.

Summary

This study examined 828 adults aged 50 years or older in Ontario, Canada, who sustained a fragility fracture while already taking bone-active medication and were screened through the Fracture Screening and Prevention Program (FSPP), a large jurisdiction-wide Fracture Liaison Service. Patients were enrolled between August 2017 and March 2024 and were subsequently referred for assessment by a bone health specialist.

The investigators examined patterns of pharmacotherapy change, continuation, or discontinuation after specialist review. The cohort had a mean age of 74.6 years, 94% were female, and the most common index fracture sites were the wrist (356/828; 43%) and hip (161/828; 19%). The most commonly reported baseline medications were risedronate (433/828; 52%), denosumab (225/828; 27%), and alendronate (157/828; 19%).

Overall, 60% of patients were advised to change treatment after fracturing while on therapy, compared with 34% advised to remain on the same medication and 3% advised to discontinue treatment. Most patients who changed treatment were switched to denosumab (418/500; 84%), while only a small minority were switched to anabolic therapy (26/500; 5%). Older patients were more likely to receive a recommendation to change medication, whereas patients already taking denosumab were much less likely to be switched. The findings suggest that fracture on treatment commonly triggers pharmacotherapy reassessment and escalation, but that management decisions are also shaped by access, reimbursement criteria, and other real-world clinical considerations.

The full publication is available here as an open-access article.

Key findings

1. Over 2.5 years, the number of recognised FLS centres in Belgium increased fourfold, from 5 to 20, with 14 of the 15 new centres established in the two target provinces of East-Flanders and West-Flanders.
2. Implementation was driven by a structured regional quality improvement approach that included regional meetings, webinars, coaching clinics, workshops, and peer-to-peer collaboration between hospitals and clinical leaders.
3. The initiative showed that FLS implementation can be accelerated without direct government financial incentives, but long-term sustainability will require stronger policy support, reliable data systems, and dedicated staffing.

Summary

This article describes the implementation of Fracture Liaison Services (FLS) in Belgium through a regional healthcare quality improvement initiative conducted over 2.5 years, from June 2023 to December 2025. The initiative focused on East-Flanders and West-Flanders and was aligned with the International Osteoporosis Foundation’s Capture the Fracture® framework and the European FLS Academy and Network (FAN), established by UCB Pharma s.a. and endorsed by the global Fragility Fracture Network.

At the outset, Belgium had only five recognised FLS centres despite a substantial osteoporosis care gap, with most fracture patients not receiving osteoporosis treatment. Over the course of the initiative, the number of recognised centres increased to 20, including 14 new centres in the two target provinces. This progress was supported by regional meetings, peer-to-peer engagement, webinars, coaching clinics, workshops, and ongoing collaboration with local clinical champions.

The initiative also identified important barriers to implementation, particularly the lack of structured fracture data and insufficient funding for staff to manage case-finding and data systems. Hospitals responded by investing in FLS coordinators and strengthening use of electronic health records. The authors conclude that this regional model could provide a practical blueprint for other settings, while also emphasising that long-term sustainability and independence will require government support and policy commitment.

For more details, visit PubMed.

Key findings

1. Pharmacological treatment after hip fracture was associated with substantially lower refracture rates, with treated patients showing a median refracture rate of 4% compared with 10% in untreated patients.
2. Fracture Liaison Services (FLS) improved secondary prevention care by increasing treatment initiation rates by an average of 44% and medication adherence by 33% over 12 months.
3. Despite the effectiveness of treatment and coordinated care models, important gaps remain: fewer than half of patients were initiated on treatment after hip fracture, adherence was often poor, and follow-up bone mineral density testing was reported inconsistently.

Summary

This systematic review synthesised evidence on secondary prevention interventions and care approaches for osteoporosis management in older people with hip fracture. The review examined pharmacological therapies including bisphosphonates, anabolic agents, denosumab, and other antiresorptive treatments, as well as service delivery models such as Fracture Liaison Services (FLS) and Orthogeriatric Services. The aim was to assess their effectiveness in reducing refracture rates, improving bone mineral density, increasing treatment adherence, and strengthening post-fracture care.

The review found that pharmacological treatment was associated with a meaningful reduction in refracture risk. Bisphosphonates were the most frequently studied therapies and remain the main first-line treatment option in many settings, while anabolic agents and denosumab were also important options for selected high-risk patients. Reduction in refracture rate after initiation of osteoporosis treatment was reported in 23 (38%) studies. The median refracture rate was 10% among patients who did not receive osteoporosis treatment, compared with 4% among older adults who were treated, a difference that was statistically significant in 13 of the 15 studies that assessed this outcome.

The review also found that FLS are associated with fewer refractures, reduced mortality and improved treatment adherence. In four of the five studies that assessed the impact of FLS, the introduction of an FLS increased treatment initiation rates by an average of 44%. Twelve months after starting treatment, the average medication adherence rate was 39%, with FLS contributing to an average of 33% improvement over this period in five studies.

However, important care gaps remain. In this review, fewer than one-third of studies reported findings related to BMD, and the overall rate of BMD testing in hip fracture patients across studies was less than 20%. Non-adherence to prescribed medications continues to pose a substantial barrier to the effective secondary prevention of osteoporosis. The rate of medication adherence in reviewed studies was found to be below 50%.

The authors conclude that early treatment initiation, structured follow-up, and locally adaptable care pathways are essential to optimise osteoporosis management after hip fracture.

For more details, visit PubMed.

Key findings

1. In this hospital-wide CT screening programme, 14% of adults aged 50 years or older had AI-flagged, radiologist-confirmed vertebral fractures (566/3,971), and 49% of eligible patients with confirmed vertebral fracture required further osteoporosis evaluation or treatment after FLS triage.
2. Among all patients who underwent CT during the study period, 7% were identified as having a vertebral fracture that required further clinical follow-up, equivalent to around 2-3 patients per day in this healthcare system.
3. Routine radiology reporting alone identified only a minority of patients who ultimately required follow-up: only 42% of all vertebral fractures were documented in original radiology reports, and among patients needing action, only about one in three were identified in routine reports.

Summary

This study evaluated the real-world implementation of an AI tool for opportunistic vertebral fracture detection in routine CT scans, aligned with a local Fracture Liaison Service (FLS) pathway at Linköping University Hospital in Sweden. Over a four-month period, the AI algorithm was applied to all non-skeletal thoracic and/or abdominal CT scans performed in adults aged 50 years or older. AI-positive cases were reviewed by experienced radiologists, and confirmed vertebral fractures were then triaged by the FLS team for follow-up.

During the study period, 3,971 unique patients underwent 5,147 CT examinations. The AI system identified vertebral fractures in 566 patients, representing 14% of the screened population. Of the 530 patients eligible for FLS triage, 49% were judged to require further clinical action because the fracture had been previously unrecognised, untreated, or inadequately managed. Overall, this meant that approximately 7% of all patients screened through routine CT had a vertebral fracture requiring follow-up, equivalent to around 2.2 patients per day.

The study also showed that implementation was feasible in routine practice. The AI system was successfully integrated into the hospital IT and PACS environment, radiologist review of AI-positive cases required around 2 to 4 hours per week, and diagnostic performance was strong, with sensitivity of 91% and specificity of 92% in the real-world validation sample. The authors conclude that AI-supported vertebral fracture detection, when linked to an FLS pathway, can uncover a substantial burden of previously missed vertebral fractures and strengthen secondary fracture prevention.

The full publication is available here as an open-access article.

Key findings

1. In this population-based cohort of 88,653 adults aged 40 years or older undergoing first DXA, FRAX® underestimated major osteoporotic fracture risk after non-recent vertebral fracture and non-recent multiple fractures, suggesting that the standard FRAX® prior-fracture input does not fully capture the additional risk associated with fracture site and multiple prior fractures.
2. Simple modifiers improved prediction accuracy: multiplying FRAX® major osteoporotic fracture probability by 1.3 for prior non-recent vertebral or multiple fractures, and multiplying hip fracture probability by 1.7 for prior non-recent multiple fractures, produced good calibration in the validation cohort.
3. Applying these modifiers reclassified 13.3% of individuals with prior non-recent high-risk fracture into the high-risk major osteoporotic fracture category and 3.5% into the high-risk hip fracture category, improving identification of patients who may warrant treatment.

Summary

This large Canadian study examined whether FRAX® adequately captures long-term fracture risk in people with a history of non-recent high-risk fracture, defined as hip, vertebral, or multiple fractures that occurred more than two years before assessment. The investigators used the Manitoba Bone Mineral Density Registry linked to provincial administrative health data and included 88,653 adults aged 40 years or older undergoing their first DXA scan between 1996 and 2018. Individuals with a more recent prior fracture, occurring within two years of the index date, were excluded.

The study found that FRAX® continued to stratify fracture risk in these patients, but that the standard FRAX® prior-fracture variable did not fully account for the residual risk associated with some older high-risk fractures. After adjustment for baseline FRAX® probability, prior non-recent vertebral fracture and prior non-recent multiple fractures were each associated with a 41% higher risk of major osteoporotic fracture, while prior non-recent multiple fractures were also associated with a 41% higher risk of hip fracture. In contrast, prior non-recent hip fracture did not show significant residual risk after baseline FRAX® adjustment.

To improve prediction accuracy, the investigators derived simple modifiers from observed-to-predicted fracture ratios in the derivation cohort. A multiplier of 1.3 improved calibration for major osteoporotic fracture after prior non-recent vertebral or multiple fractures, while a multiplier of 1.7 improved calibration for hip fracture after prior non-recent multiple fractures. When applied in the validation cohort, these adjustments reclassified 13.3% of individuals with prior non-recent high-risk fracture into the high-risk major osteoporotic fracture category and 3.5% into the high-risk hip fracture category. The findings suggest that a simple refinement to FRAX® may improve identification of patients whose long-term fracture risk remains higher than standard FRAX® estimates indicate.

The full publication is available here as an open-access article.

Key findings

1. In this cohort of 952 community-dwelling adults aged 65 years or older, 23% experienced at least one injurious fall requiring emergency presentation during follow-up, equivalent to an incidence rate of 19.3 per 1,000 person-years.
2. Lower balance confidence, measured using the Modified Falls Efficacy Scale (MFES), was independently associated with a higher risk of incident injurious falls: each one-point decrease in MFES score was associated with a 1.9% increase in the sub-distribution hazard of injurious falls.
3. Balance confidence appeared to be clinically relevant but not sufficiently accurate for stand-alone prediction. Although an MFES cut point of 55 was identified, it had extremely high sensitivity (99.1%) but very low specificity (1.3%), with an area under the receiver operating characteristic curve (AUROC) of 0.508.

Summary

This longitudinal study examined whether balance confidence predicts future injurious falls in older adults and whether it helps explain the link between previous and subsequent falls. The investigators analysed data from 952 community-dwelling adults aged 65 years or older participating in the Geelong Osteoporosis Study in Australia. Balance confidence was measured using the 14-item Modified Falls Efficacy Scale (MFES), and the outcome was time to first emergency department presentation for an injurious fall, identified through linkage with the Victorian Emergency Minimum Dataset. Participants were followed for a median of 11.5 years.

Over follow-up, 219 participants (23%) experienced at least one injurious fall requiring emergency presentation. Lower MFES scores were associated with a significantly higher risk of incident injurious falls, with each one-point decrease in score corresponding to a 1.9% increase in the sub-distribution hazard. The study also found that lower balance confidence explained 14.9% of the relationship between previous self-reported falls and subsequent injurious falls, suggesting that reduced confidence may be one pathway through which a prior fall increases future falls risk.

Although the authors identified an MFES cut point of 55, its predictive performance was poor, with very low specificity and an area under the receiver operating characteristic curve (AUROC) close to chance. The findings therefore support the use of balance confidence assessment as one component of a broader multifactorial falls risk assessment, rather than as a stand-alone screening test. This is consistent with the World Falls Guidelines that recommend attention to balance confidence alongside gait, balance, and other modifiable risk factors.

The full publication is available here as an open-access article.

Key findings

1. The authors propose that the osteoporosis “care gap” should be defined more broadly than the traditional treatment gap, as “the gap between care offered to people with, or at increased risk of, osteoporotic fractures, and best practice, person-centred care.”
2. The paper argues that focusing only on the treatment gap risks a disease-focused and paternalistic approach, and that osteoporosis care should instead be person-centred, participatory, equitable, holistic, multidisciplinary, and respectful of patient autonomy.
3. The authors identify multiple drivers of the care gap, including low public awareness, health inequalities, insufficient policy prioritisation, fragmented care pathways, inadequate follow-up, and limited support for shared decision-making, and call for coordinated action across policy, services, and professional practice.

Summary

This position statement presents the work of the REFRAME group, which brought together 32 representatives from six countries, including clinicians, researchers, public contributors, and charity representatives, to define the osteoporosis care gap and develop a framework for addressing it. The group used an online workshop followed by a three-round modified Delphi process to reach consensus on a definition and a broader call to action.

The paper argues that, although the osteoporosis treatment gap remains important, the prevailing narrative has become too narrowly focused on pharmacological treatment. The authors note that the treatment gap in Europe has been estimated at 74.6%, but suggest that concentrating only on treatment risks overlooking other important shortcomings in care. In response, they define the osteoporosis care gap as the gap between care currently offered to people with, or at increased risk of, osteoporotic fractures and best-practice, person-centred care.

The statement identifies determinants of the care gap at multiple levels. These include low public awareness and health inequalities at societal level; insufficient prioritisation, diagnostic confusion, and lack of incentivisation at policy level; fragmented pathways, unclear professional roles, poor communication, inadequate follow-up, and limited support for shared decision-making at service level; and unmet needs for understandable, participatory, equitable, holistic, and multidisciplinary care at individual level. The authors call for systematic case finding, timely diagnosis, access to both pharmacological and non-pharmacological management, better communication using plain language, stronger support for shared decision-making, and closer integration across primary, secondary, tertiary, imaging, and falls services.

Overall, the paper provides an important conceptual shift for clinicians, services, and policymakers. Rather than focusing only on the proportion of eligible patients who receive medication, it argues for a broader and more patient-centred view of quality osteoporosis care. The authors conclude that osteoporosis care should be tailored, inclusive, evidence-based, and multidisciplinary, with the person placed firmly at the centre of decision-making and service design.

The full publication is available here as an open-access article.