The interdisciplinary fracture liaison service improves health-related outcomes and survival of older adults after hip fracture surgical repair

Falls and fragility fractures may hurt older adults’ quality of life and their families due to catastrophic loss of independence, morbidity, permanent disability, and death [1, 2]. Hip fractures are a priority among fragility fractures causing the most significant cost burden on healthcare services in Italy and worldwide [3].

In 2000, the incidence of hip fractures was about 1.6 million worldwide. Accounting for the aging population, authors estimate that hip fractures will reach 6.3 million in 2050 [4]. In Italy, hospitalizations due to hip fractures reached 94,525 in 2014, mainly affecting the oldest old subgroup, representing 2.5% of the population [5]. In the Umbria region, hospitalizations due to hip fractures were 1825 in 2014, with a peak of 1885 in 2012. The direct cost estimations reached € 18,566,633 in 2014 and peaked at € 19,177,043 in 2012 [5].

Several institutions strongly recommend interventions to prevent fragility fractures. Balancing vitamin D and calcium deficiency, encouraging physical activity, and a healthy lifestyle are priorities in the general population. Promoting initiation and adherence to appropriate anti-osteoporosis medications over the top of previous interventions is mandatory among high-risk subjects [6‐8]. The cost-effective ratio of anti-osteoporosis treatments has been proved among high-risk patients, especially those with previous hip fractures [9, 10]. However, appropriately managing subjects at high risk of fragility fractures remains an unsolved issue with many barriers identified: uncertainties about who is responsible for osteoporosis care once a fracture has occurred and the lack of coordinated services with the capability to receive, assess, and monitor treatments of an increasing number of high-risk subjects [11].

Fracture liaison services (FLSs) are interdisciplinary or integrated clinical pathways particularly welcomed for providing care to high-risk fracture subjects [12, 13]. The FLSs may grant a synergy between in-hospital and community care services and professionals, overcoming patients’ issues and obstacles to diagnostic and therapeutic initiation. FLS has been identified among cost-effective interventions [13‐16], mainly if an interdisciplinary service provides care [17, 18]. Besides interventions dedicated to patients’ and caregivers’ education, falls and fracture prevention, FLS may support a gaining in QALYs and optimize the country-specific cost per QALY [10]. However, high heterogeneity characterizes available FLSs, which are usually unique to the providers, needs, healthcare systems, and patients they serve. Direct comparisons are not feasible among FLS models. Outcomes mainly depend on the methodology adopted to implement the services, including the availability of dedicated resources, the adoption of performance indicators, and a quality improvement approach [13, 16‐19].

FLSs are not an established pathway of care at the regional and national levels in our country, and further evidence may help make them considered a standard of care. We showed that implementing a FLS improved identification, clinical management, and adherence to anti-osteoporosis treatments in a high-volume trauma center [20]. We implemented the outpatient service to provide an interdisciplinary assessment at 30-day from the patient’s hospital discharge after hip fracture repair. The main novelty of the service is the contextual assessment by the orthopedic surgeon and geriatrician, providing patients with tailored interventions to optimize pharmacological treatments, functional recovery, and appropriate monitoring. We hypothesize that such a service could improve clinical outcomes of older adults who underwent hip repair and positively impact healthcare-related burden.

The primary aim of this study is to confirm the impact of an innovative interdisciplinary FLS care pathway (i.e., FLS-CP) as compared to a usual care pathway (i.e., U-CP) on patients’ adherence to timely prescribed anti-osteoporosis treatments and then to explore the service’s effect on healthcare service utilization, fall and fracture incidence, functional recovery, and mortality as secondary outcomes.

This study confirmed a higher adherence to secondary prevention fragility fractures of older adults with previous hip fractures who entered the FLS-CP than those who followed the U-PC. Although the absolute numbers of people entering and following the innovative pathway of care were lower than that observed in the usual care pathway, we found that about three out of four subjects belonging to the FLS-CP initiated appropriate treatments within 30 days from hip fracture surgery and remained on treatments for up to 1 year. Older adults in the FLS-CP group experienced a lower probability of being multiple fallers, without difference in fracture incidence over 1-year follow-up. However, subjects in the FLS-CP experienced lower health facility admissions and a higher hospital-free survival than those in the U-CP. Furthermore, older adults belonging to the FLS-CP experienced lower 5-year mortality than those in the U-CP group, with evidence for an early survival advantage in the FLS-CP than in the U-CP group.

Identifying interventions driven by the CGA is an essential pillar of the FLS-CP. Our service is a physician-led FLS offering patients’ evaluation by orthopedic surgeons and geriatricians and providing integrated notes to manage bone fragility and frailty of older adults with hip fracture repair [25, 26]. The integrated multidisciplinary approach delivers tailored interventions for fractures and fall prevention, taking into account optimization of comorbidity and polypharmacy, frailty, functional and social-behavioral aspects, patients’ preferences, and care goals [24‐26]. The interdisciplinary approach may strengthen adherence to treatments, improve the awareness of the causal relationship between osteoporosis and fragility fractures by patients and caregivers, and even support the GPs in fostering persistence in the secondary prevention of fragility fractures [27‐29]. However, there is a need for systematic and standardized implementation of such services especially among high-risk patients [13, 16]. We acknowledge the issues associated with entering the service based on patients’ care goals and preferences, perhaps under-powered the possible impact of the service on maximizing adherence to treatments among patients highly deserving them.

Similar to previous study, our FLS care pathway profits from a local network of available services, including an administrative assistant in charge of scheduling patients’ exams and visits [20, 30]. The administrative assistant resulted particularly helpful to overcome patients’ non-clinical, educational, and financial barriers, improving patients’ timing for entering the care pathway and reducing dispersion through the follow-up. Building on the local network of services, high-risk patients received lab exams and X-ray evaluation of repaired hips. One-third of FLS-CP recipients had BMD testing within 30 days from surgery, improving previous timing and performance [20]. However, DXA scans cannot be considered a marker of how often patients were worked-up for fracture risk. In most subjects, the secondary prevention of fragility fractures started independently of a T-score threshold [31].

A unique feature of our interdisciplinary FLS-CP is the capability to capture more than a half of patients deserving secondary prevention of fragility fractures in a high-volume trauma center. Although our service is still evolving to fulfill the leading key performance indicators [28], our performance is close to that of a French FLS service [32] and higher than those from previous other experiences [20, 30‐36]. Both structural and process features might ensure the FLS satisfies the healthcare demands of high-risk patients [16, 17, 36] and contribute to the effectiveness of increasing patients’ adherence to timely initiated treatments. Over 70% of patients in the FLS-CP were on complete anti-fracture therapy 3 months later the post-surgical visit, and 80% persisted on treatments over the follow-up up to 1 year. In this context, it is conceivable that a higher rate of drug prescription or initiation of treatments close to the index event may increase medication adherence and persistence [33, 35]. We argue that interdisciplinary management might have been crucial to overcoming previous results showing that almost half of hip-fracture patients receive anti-osteoporosis drug prescriptions within 3 months from hip fracture repair, and 39% assured 1-year drug adherence [20].

Contrary to previous studies, we failed to demonstrate that FLS-CP reduces the rate of new fractures and falls within 1 year [37‐39]. Gomez et al. showed that an integrated multidisciplinary care model reduces falls and fractures over 6 months [37]. Inderjeeth et al. proved that FLS minimizes the rate of new fractures [38]. González-Quevedo et al. explained the apparent failure of FLS in reducing the secondary fractures over 1 year since hip fracture based on the underdiagnosis of vertebral fractures, which are the most common in this subgroup [39]. Findings from our study support that vertebral fractures are frequently underdiagnosed due to patients’ lack or atypical symptoms [39, 40].

Falls are strongly associated with hip fractures. One out of two patients develops them within 1 year, and one in ten patients experiences further fractures [41]. Growing evidence supports bone-muscle interaction leading to sarcopenia and osteoporosis complex, high risk of falls, mobility disability, and fractures [42]. According to previous studies, a high vitamin D initiation and adherence rate may significantly prevent falls [43]. Combining vitamin D, tailored exercise, diet, and pharmacological advice might have impacted risk factors for falls, possibly reducing the proportion of multiple fallers, but there was no effect on falls in our study.

We add to the literature by showing that FLS reduces the rate of health facility admissions, including ER and re-hospitalization due to adverse events. Similarly, McLellan et al. reported saving 266 hospital bed days for FLS over usual care in a hypothetical cohort of 1000 patients [15].

Regarding mortality, this study adds to the literature by showing that FLS may improve survival up to 5 years from hip fracture surgery, with the highest protective effect during the first year. In our study, survival curves divide in the early phase and remain almost parallel up to 36 months from baseline, with a favorable trend up to 5 years of follow-up. In a pooled data from 15 studies, Wu et al. found that the FLS significantly reduced the risk of mortality by 3% [12], and then González-Quevedo et al. found a reduction of mortality rate by 5.6% using pre- and post-intervention design [37]. In our study, average mortality rates in the FLS group are similar to those found in the previous research. In contrast, a higher mortality risk was found in the control arm and persisted over the years. Focusing on hip fracture survivors, one out of two functionally independent patients before fracture remains dependent on daily living activities up to 1 year from surgery and never regain their pre-fracture functional level. Compared with matched controls, hip fracture subjects already showed lower independence after 1–2 years in previous studies [44, 45].

This study has several limitations, firstly due to the observational nature of the study design. However, a randomized trial may be considered inappropriate given the poor evidence of equipoise about effectiveness from policy, clinical, and patient perspectives [15]. In the absence of strong evidence about the superiority of FLS-CP, all participants were offered to access the multidisciplinary service; however, they voluntarily chose the more suitable pathway of care according to their preferences and goals. We consider it ethically acceptable based on the principles of CGA and taking into account service’s sustainability issues in a real-world healthcare setting.

Third, we acknowledge the limitations of missing data concerning comorbidities and polypharmacy in U-CP. This reflects the usual orthopedic management mainly focused on surgical outcomes after hip fracture, leaving other care aspects to patients’ GPs. Then, we may have selected patients with different characteristics beyond those measured. Patients able or willing to enter the FLS-CP might be healthier and better educated than those who attend U-CP service, then affecting main outcomes through an healthy user effect bias or unmeasured confoundings, including social, economic, or cultural aspects. However, on one side, the slightly higher rates of delayed surgery (88%) and delayed weight-bearing (23.3%) in the U-CP versus the FLS-CP (78.6% and 8.3%, respectively) could suggest a higher burden of comorbidity. Still, the higher IADL independence in the U-CP compared to FLS-CP before index fracture may substantially attenuate the unhealthy user effect bias hypothesis. These aspects may open the discussion about technical options for hip fracture repair, the severity of osteoporosis, and reasons for delayed weight-bearing, suggesting that people who receive delayed weight-bearing should be considered a different clinical subgroup and deserving a specific evaluation. Fourth, the healthcare facility use may depend on the local service availability and organization to fulfill health-related needs. Therefore, our findings cannot be easily compared with other countries. However, it is conceivable that an orthogeriatric FLS may solve patients’ needs associated with frailty, otherwise undetected in the usual care pathway. Fifth, information bias may be present in the U-CP group about BADL and IADL measures over the follow-up telephone calls. The FLS group was already interviewed about those, while the U-CP group only receives the questionnaire during the telephone follow-up. However, two trained geriatricians performed patients’ and the caregivers’ interviews, reducing potentially over- and underestimating or identifying fallacious memory problems. Fifth, transversal statistical analyses do not allow exploring the causal relationship among variables and outcomes and the 1-year follow-up prevents conclusions about falls and fracture incidence. Although we originally planned to continue the telephone follow-up, after 15 months, we suspended given the decline in patients’ and caregivers’ cooperation. However, the survival analysis was not affected because of the length of follow-up time was assured by accessing the regional administrative registry. The 5-year follow-up time for survival may be a strength of our study. We also recognize that information from CGA was not available in the U-CP group, and functional status was not available in deceased patients over the first-year of follow-up. Finally, the low sample size prevents us from making inferences other than adjustments for age and sex.

In conclusion, patients who enter an interdisciplinary FLS care pathway experience better outcomes regarding adherence to anti-osteoporosis treatments timely initiated after hip fracture repair and healthcare facility admissions owing to adverse events. Older subjects who entered the FLS care pathway tended to be less likely multiple fallers and experienced lower mortality rates starting at the early phase of follow-up. We support the favorable impact of an interdisciplinary FLS programs to bridge the gap in secondary prevention of fragility fracture in high-risk patients.