Introduction

Fractures as a result of low velocity or low energy falls are referred to as ‘fragility’ or osteoporotic fractures. These fractures are life-changing events for many, resulting in substantial morbidity and mortality as well as high economic costs [1,2,3,4]. Recent European and global reports show the incidence is rising as the planet’s population ages and people live longer lives; European data suggests there will be an almost 25% increase in number by 2030 [3, 4].

Despite a wealth of evidence of the effectiveness of osteoporosis drugs in reducing fracture risk, finding and treating those at the highest risk of fractures remains a challenge. In Ireland, similar to many countries worldwide, the billions of euros spent on fragility fracture care is almost all on the treatment of acute fracture rather than on fracture prevention [3]. This cost was estimated to be close to €400 million in Ireland over a decade ago, projected to be €1billion today, and close to €2billion by 2030 [5,6,7]. After an osteoporotic fracture in those aged over 60 years, a third of survivors will go on to have another fracture within 5 years, the risk being highest in the early post-fracture period [8, 9]. These are a key target group to capture therefore, namely those who have just suffered a first fragility fracture. International evidence shows the majority are neither diagnosed nor treated for their underlying osteoporosis following these sentinel events, and a significant gap remains between best evidence and current practice [10,11,12]. A recent survey of Orthopaedic staff in Ireland showed that while the majority agrees with treating the underlying osteoporosis, they also believe the responsibility rests with others [13]. Analysis of a large primary care dataset showed that only 1 in 5 patients deemed at high risk of fracture is initiated on osteoporosis medication [14]

Fracture liaison services (FLS) have emerged to address these gaps in care, which are evidence-based systems to case-find, assess, treat and monitor patients aged 50 years and over with incidence fragility fractures [15]. These programmes identify patients with recent fractures, address their falls risk and bone health and provide education and appropriate intervention. High-quality FLS are cost-effective and propose or initiate clinically effective interventions shown to reduce fracture numbers in these high-risk populations [15]. FLS exist in many countries today, and in 2012, the International Osteoporosis Foundation (IOF) launched a Best Practice Framework for FLS, entitled ‘Capture the Fracture’ [16]. This initiative outlines clearly in a step-wise manner how a service should operate and the necessary quality standards for efficacy in reducing fracture numbers in the population the FLS serves [17]. In 2020, a complementary set of Key Performance Indicators (KPIs) were launched by the IOF which measure performance at the patient level [18].

A number of FLS have been in existence in the Republic of Ireland (RoI) for some time but details are limited on the number, scope and process. Only six are listed on the IOF ‘Capture The Fracture’ website, the same number as Northern Ireland which has a population of around one-third that of RoI. There are few restrictions on the availability of osteoporosis medications in RoI; practitioners can prescribe all available licenced medications as per their clinical judgement, other than teriparatide and intravenous bisphosphonates which are initiated by hospital specialists only. As part of the establishment of a RoI National Fracture Liaison Service Database (FLSDB), there is a need to better understand existing services. We therefore conducted a facilities audit of all hospitals in Ireland, to enquire re-existence of an FLS and to evaluate their performance against the 13 IOF best practice standards for FLS, similarly to other European countries [19,20,21,22].

Methods

A questionnaire (shown in supplement) was distributed to 16 public trauma hospitals in Ireland that receive acute fracture patients. The questionnaire was mailed to the hip fracture coordinator or fracture liaison service lead in each site, and for hospitals where this was not known, the chief executive officer was asked to delegate its completion to the most appropriate clinician locally.

Firstly, we asked about the existence of an FLS either on-site or in a linked hospital. Sites without an FLS were excluded from further analysis. Where such a service exists, a series of questions were included detailing their setup, staffing, case-finding strategy, treatment protocols and monitoring approaches for fragility fracture patients. We compared our results to the 13 standards of the IOF best practice framework as per Tables 1 and 2 [17]. We summarised data and calculated proportions as this was a descriptive study so no formal statistical analyses were deemed necessary. Agreement to anonymous publication of results was sought and received from each hospital site.

Table 1 Performance of each FLS against IOF Best Practice Standards for an FLS
Full size table
Table 2 IOF capture the fracture: best practice framework for fracture liaison services
Full size table

Results

Demographics

Of the 16 trauma hospitals, all 16 responded to the survey. Nine of these sites stated that they have an FLS in place. Four of these covered more than one hospital site. Of the remaining non-trauma acute hospitals in Ireland, there was one which had an established FLS whose patients were not included in their regional trauma hospital FLS, i.e. a standalone service for local fracture patients, so their results were included in the analysis, i.e. 9 FLS at trauma hospitals and 1 FLS within a non-trauma hospital.

Of the 10 sites with an FLS, 4 were established for more than a decade. All had a nursing staff of various grades running the service and regardless of service size; there was only one member of staff in most units, though the proportion of time spent on FLS was not captured in the survey. The consultant supervision for the service was by a geriatrician in 7/10 centres, a rheumatologist in 3, an orthopaedic surgeon in 3 and an endocrinologist in 2, so 5 centres had a combination of more than one specialist. One site had no consultant input or supervision, being entirely nurse-led. Four of 10 sites had some administrative support but none included the task of data entry, which is carried out by the FLS nurse only.

Access to DXA scanning was available for all, 3 of these being on the same hospital site. Seven sites had a waiting time of over 4 months for a DXA, with average waiting times ranging from 4 to 76 weeks. The FLS nurse conducted the scans in 5 of these sites, a radiographer or nurse in the remaining. Seven sites used a fracture risk assessment tool, but only one site used it to determine the need for DXA scanning.

IOF standards

Standard 1: Patient identification

Identification processes varied across sites: 7 use hospital IT systems to identify all patients admitted with fractures or attending fracture clinics. Though we did not ask for technical details, none of these systems appears to be automated and required considerable time to run searches, as per free text comments received on the survey. For admitted fractures, 4 sites screen inpatient fractures on all wards, 3 sites do so only if they are on orthopaedic wards, the remaining 3 do not see patients during their admission with a fracture. One site does not employ a case-finding strategy and instead relies solely on referrals to the service. Insufficient time (5/10) and service still in development (7/10) were the primary reasons for service shortcomings.

Standard 2: Patient evaluation

The proportion and type of identified fracture patients who are evaluated by the FLS varied. All FLS exclude skull and digit fractures. rib, metatarsal and metacarpal fractures are excluded in 7 FLS; 6 FLS exclude patella and 3 exclude ankle fractures. Five FLS do not evaluate hip fractures; instead devolving this role to geriatricians who record their acute stay data on the IHFD (Irish Hip Fracture Database), as per Table 3 [23]. Time and lack of resources were again cited as the primary limitations for not assessing all fracture types.

Table 3 Non-hip fracture numbers per FLS vs. hip fracture numbers as recorded on IHFD in trauma hospitals with an FLS in operation in the calendar year 2019
Full size table

The proportion of non-hip fractures seen by each FLS relative to the number of hip fractures at each centre is shown in Table 3. For the calendar year 2019, n = 3444 non-hip fractures were assessed in eight of the trauma hospital FLS (as one FLS was in development, no data was available for 2019). The number of hip fractures (as per IHFD) in 2019 in these eight sites was 1905 which represents 51% of all hip fractures in Ireland that year [23]. The UK FLS database estimate that for every 1 hip fracture, there are at least an additional 4 other fragility fracture types presenting clinically, though this varies widely in estimates from other countries [3, 22]; if we apply this denominator to an Irish population, these eight FLS sites currently capture about 45% of their target population of non-hip fractures, ranging from 25 to 90% or 1.1- to 3.6-fold the number of hip fractures per site (Table 3).

Standard 3: Timing of assessment

The IOF standard recommends performing the FLS assessment within 8 weeks of fracture occurrence. Five sites reported achieving this standard for the majority of those identified, 3 sites reported this as up to 52 weeks post-fracture, while 2 provided no information.

Standard 4: Vertebral fracture identification

Vertebral fracture (VF) ascertainment is an established step in the programme since they are common, often unrecognised or unreported. Three centres screen radiology reports for new VF, i.e. VF that present clinically, one centre also reviews all spine imaging for unreported VF. Two of these 4 centres also perform VFA (vertebral fracture assessment) screening during DXA scanning, to identify prevalent VF but it was not clear if this was on all patients having DXA or on select groups. Two additional services use VFA to capture VF (i.e. they capture prevalent VF, rather than acute, and only in those referred for DXA). Four other centres only see VF patients in their service if they are referred or if they appear in fracture clinic patient lists, i.e. no case finding of VF by the FLS takes place. As this IOF standard refers to screening for prevalent VF, 4/10 services were compliant with this standard.

Standard 5: Assessment guidelines

There is no current approved official national osteoporosis guideline, though the Irish Osteoporosis Society and the Health Service Executive (HSE) have previously published a strategy to prevent falls and fractures [24, 25]. Most FLS reported devising local guidelines drawn from European, US or UK guidelines, with some following a combination of these.

Standard 6: Secondary causes of osteoporosis

Seven sites performed investigations to exclude secondary causes of osteoporosis in all cases, 2 perform investigations only for those with low DXA T-scores, while 1 advises GPs to conduct these investigations as needed.

Standard 7: Falls prevention

Six services reported that they evaluate patients for falls risk, while 3 either refer to a falls service or ask GP to do this, and the remaining site neither assesses nor refers patients for this. The content of the falls risk assessment in the 6 performing sites varied (Table 4), with just 2 covering all elements of a standard multifactorial falls risk assessment [26]. Three refer patients directly for strength and balance exercise programmes, and an additional 5 either asked GPs to refer or patients to self-refer. The 2 remaining were not aware of the availability of such classes.

Table 4 Falls risk factors elicited, as reported by 5 of 10 FLS who screen for falls risk factors
Full size table

Standard 8: Health and lifestyle risk factor assessment

Eight sites provide general education and specific lifestyle advice for all patients as part of their patient assessment.

Standard 9: Medication initiation

A decision to start osteoporosis medication was made by the FLS with specialist consultant input in 90% of sites, while the remaining site delegated this task to primary care. There were variations in the practice of prescribing medication with 7 directing patients to their GP for this, while the other 2 are the initial prescribers.

Standard 10: Medication review

New osteoporosis medication initiation includes reviewing the need for a change of existing medication, though this was not specifically addressed in our survey. Five sites conducted a medication review as part of the falls risk assessment for those medications known to increase falls risk.

Standard 11: Communication

All 10 sites send a copy of the assessment to the GP; 1 site also sends a copy to the managing orthopaedic surgeon as part of the communication standard. Only 2 send a copy to the patient. The content of this report varies but generally includes fracture risk factors, DXA results and treatment recommendations (8 sites), with fewer providing detail on falls risk (6 sites) and lifestyle factors (5 sites).

Standard 12: Long-term management

Five sites report monitoring patients to ensure those needing intervention are maintained on it, mostly by follow-up telephone clinics or by posting a questionnaire to patients. The timing of the first follow-up was at 4 months for 3 locations, at 12 months in 1 other, while the rest were either ad hoc or did not occur. All of these follow-up encounters addressed medication adherence, while just 1 FLS enquired about falls at the follow-up.

Standard 13: Database

One site has no database, while the remaining 9 maintain some information. The nurse coordinator performs all data entry as none had administrative staff support.

Discussion

This first facilities audit of trauma hospitals in The Republic of Ireland (RoI) evaluating fracture liaison services shows that there is marked heterogeneity in the present structure and functioning of these services. FLS services exist in just over half of public trauma hospitals, which are receiving hundreds or thousands of fracture patients each year. This falls short of the standard set by the Irish Trauma society in 2015 to extend this service to all 16 hospitals, and the IOF standard that all fracture patients should be assessed for secondary fracture prevention [17, 27]. While we found that most FLS access hospital electronic data systems to identify fractures, many inpatient fractures are not screened and many fragility fracture types are excluded. Of those FLS in existence, 2 were established in the last year which is a promising development, yet 6 of 10 sites were established more than a decade ago and still cite lack of resources, including time and staff as primary reasons for failing to complete the standards outlined in The International Best Practice Framework [17, 18].

In the sites with an FLS, none are capturing all fracture patients but overall approximately 45% of their target fractures numbers are being captured. This target is based on extrapolation of the denominator figure used by the UK FLS database for case identification, which is an additional 4 non-hip fragility fractures per hip fracture or 5 times for total fractures [22]. This multiple resulted from an assessment of different methods of estimating fracture numbers from hospital coding in the UK, though has not been assessed in an Irish population. Such an approach is pragmatic because the incidence of fragility fractures varies substantially between studies, reflecting the difficulties in reliable ascertainment, whereas estimating hip fracture numbers is around 90% accurate as almost all are admitted to hospital and inpatient coding is more reliable [2, 6]. Though this case ascertainment figure for the 8 sites in Ireland is commendable at 45% vs. the UK FLS national database figure of 49% [28], it represents just 19% of expected fracture numbers in Ireland annually (there were 3701 hip fractures in 2019). This is a conservative estimate: national admitted patient data in Ireland suggests that hip fractures represent between 25 and 35% of inpatient fractures already [6]; adding those non-hip fragility fractures who are not admitted to hospital could give a hip: all fracture ratio of far higher than 1:5. Neither does this figure include patients managed at private facilities or patients managed solely at a primary care level. Further studies to better describe the true incidence of fragility fractures are required in Ireland and the forthcoming Irish FLSDB will help answer this.

FLS are both clinically and cost-effective systems of care for secondary fracture prevention [1, 15]. When fully implemented and the necessary quality standards are met, they will reduce fracture numbers locally and nationally, reducing the suffering for patients, and the illness burden for the healthcare system. In Ireland, fracture admissions to public hospitals increased by 30% from 2000 to 2014, and due to our ageing, demographics are expected to double over the next two decades [6]. International evidence and a national strategy for falls and fracture prevention have long established the urgent need for a national programme to systematically address this. The establishment of the IHFD resulted in annual audits of standards of care, which appears to have resulted in significant improvements in the care of patients aged 60 years and older with hip fractures [29]. A similar rollout of standards of care for all fracture patients could have similar benefits and help reduce the inexorable rise in fracture numbers we have witnessed in the past decade.

Though all sites report being under-resourced, it is reassuring to see the great efforts of many to attain many elements of a high-quality FLS, including a multifactorial falls risk assessment. Some of the falls risk assessments are being carried out by FLS, and strength and balance classes are promoted in the majority of services. We do not know the proportions of patients nor fractures in which these elements are covered however, as this was not a patient-level audit, nor do we know how effective or satisfactory they are for patients and the service. Data from other registries such as the UK FLSDB and the RCP national falls audits show sometimes large mismatches in facilities and patient-level audit data [21, 28, 30].

Initiating FLS patients onto osteoporosis drug treatment in a timely manner is a fundamental standard. Concerningly while we found that most FLS either prescribe or request GPs to initiate, only 50% of sites were able to say that patients who receive a first prescription within 4 months of a fracture. Early treatment is extremely important because of the imminent fracture risk in the months following a first fracture, and the fact that it takes at least 12 months for treatment efficacy to be observed [8, 9, 31].

Osteoporosis is a chronic disease in which patient engagement for self-management can increase successful treatment rates. We found that lifestyle advice and patient education are given by nearly all FLS, but as only 2 centres send the patient a copy of the assessment/GP letter, this is an easy target to address that might improve treatment adherence rates.

Our study has important limitations. Firstly, the data are all self-reported and we are relying on colleagues countrywide to provide the most accurate data available. Since not all have a database and some have limited support it is likely that many key metrics are either incompletely captured, or not at all. Secondly, the data are around facilities and services, rather than patient-level data, which precludes inferences about the quality or outcomes of each service. Although the international evidence shows that fracture liaisons services are both clinically and cost-effective, operationalising a national policy and the required resources will take time. Until all centres have an appropriate level of staffing, training and standards in place, the real and substantial benefits of FLS will not be realised. Despite the significant deficiencies at a local and national level, there are clearly pockets of expertise, goodwill and a strong professional culture to do the right thing, which can help guide the establishment of the national programme and support a national policy to implement best practice to benefit our citizens.

In summary, we have performed the first audit of FLS in Ireland ahead of the establishment of a national FLS database. Although there is a promising activity in this regard, there are large and important deficiencies, which must be addressed in addition to establishing a national framework, policy and strategy. Considerable investment is needed to ensure that all individuals at high risk of fragility fracture are appropriately assessed and treated. Failure to implement the evidence base for FLS will result in an alarming rise in avoidable fractures ahead. The launch of a forthcoming national database is an important step towards addressing this gap.