Three-year follow-up of a novel orthopedic ward fracture liaison services (OWFLS) model

Abstract

Objective

We established an orthopedic ward fracture liaison services (OWFLS) model and evaluated its role in improving detection rates of bone metabolic markers, treatment rates, and long-term treatability.

Methods

This observational retrospective cohort study included 120 patients aged >50 years hospitalized for primary osteoporotic fracture from January 2018 to January 2019 (group A: not included in OWFLS). Group B (included in OWFLS) comprised 120 patients from February 2019 to February 2020. We compared rates of bone metabolic index testing, treatment, and adherence; symptomatic improvement; and recurrent fracture between groups.

Results

Rates of bone metabolism index testing (50% vs. 0%) and medication use (94.2% vs. 64.2%) were significantly higher after OWFLS implementation. There was no significant difference in adherence rates at 3 months between groups (97.3% vs. 93.5%). Adherence rates at 1 and 3 years were better in group B than A (73.5% vs. 51.9%; 57.5% vs. 26%, respectively). Recurrence of bone pain at 1 and 3 years was significantly lower in group B than A (20.4% vs. 46.8%; 45.1% vs. 76.6%, respectively).

Conclusions

OWFLS improved the detection rate of bone metabolism indicators, treatment rate, and patient adherence and reduced recurrence of bone pain. OWFLS may be suitable for settings lacking human resources.

Introduction

In the global ageing population, osteoporosis is a relatively common disease. According to statistics, the number of people with osteoporosis exceeds 75 million in Europe, the United States, and Japan. ¹ Osteoporosis is characterized by a loss of bone mass and changes in the microstructure of bone tissue, which inevitably increase the risk of osteoporotic fractures. ² Fractures owing to osteoporosis account for approximately 9 million cases each year, imposing a tremendous economic burden on society in terms of medical costs. ³ A retrospective observational study of more than 70,000 osteoporotic fractures in older adults including data from multiple sources highlighted the clinical, human, social, and financial burden caused by these fractures. ⁴ In 2005, the cost of osteoporotic fractures in the United States was estimated to be USD 19 billion. It is expected that by 2025, the losses will reach approximately USD 25.3 billion.^5,6

Nearly all fractures are associated with an increased risk of future fractures, independent of age, fracture site and baseline bone mineral density (BMD).^7–9 The risk of re-fracture increases by 200% after vertebral fracture and 300% after hip fracture, according to studies. ¹⁰ A large population-based cohort study demonstrated that 1 year after the first osteoporotic fracture, the risk of re-fracture was 2.7 times higher than the risk with no previous fracture. ¹¹ Kanis et al. found that patients who experienced any fracture had an 86% increased risk of other types of fracture in the future. ⁹ Additionally, most re-fractures occur within the initial years after a fracture, meaning that the risk of subsequent fractures in the short term after a fracture is higher than that in the long term, referred to as the “imminent fracture risk.” ¹² Nevertheless, there is a considerable gap in the number of patients with recent fractures who are assessed and treated, with only 10% to 20% of patients worldwide undergoing formal treatment.^13–16 The American Society for Bone and Mineral Research, the International Osteoporosis Foundation, the European League Against Rheumatism, and the European Federation of National Orthopaedic and Trauma Societies have formulated recommendations for the secondary prevention of fractures, referred to as fracture liaison services (FLS).^14,17,18

FLS is a multidisciplinary functional coordination model that allows patients with fracture to quickly initiate secondary prevention plans for fractures.^16,19 In FLS, various medical professionals (orthopedic physicians, endocrinologists, geriatric physicians, radiologists, pain specialists) or specific caregivers work in collaboration. ²⁰ Treatment adherence is increased by 20%, the treatment gap is reduced by 20%, and re-fracture and mortality rates decline by 5% when FLS is fully implemented. ²¹ The FLS system is currently in the initial rollout stage in China, with issues such as a lack of high-quality specialist care and patient compliance. Hence, we introduced the new orthopedic ward fracture liaison services (OWFLS) model. We conducted an observational retrospective study aimed at evaluating the role of OWFLS in improving the detection rate of bone metabolic markers, treatment rates, and long-term treatability among patients.

Methods

Interventions

The First Affiliated Hospital of Ningbo University approved this study (Ethical Review of New Hospital Technologies and Projects in 2019 Opinion No. 95, March 2019). All procedures implemented in this study involving patients complied with the ethical standards of the institution and the National Research Council. All participants provided written informed consent at the beginning of the study. We de-identified all patient details. The reporting of this study conforms to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ²²

OWFLS model

First, we screened patients with osteoporotic fracture older than age 50 years. OWFLS liaison officers (trained nurses) were responsible for targeted contact with the supervising physician. The supervising physician prescribed osteoporosis-related auxiliary examination with dual-energy X-ray absorptiometry (DXA) and laboratory testing (blood count, urine tests, liver function, kidney function, parathyroid hormone, blood calcium, blood phosphorus, 25-hydroxyvitamin D, and bone turnover markers). Based on the auxiliary examination and laboratory testing results, each patient was assessed as having primary or secondary osteoporosis. If the patient had primary osteoporosis, they were included in the OWFLS diagnosis and treatment system. Otherwise, the patient was excluded and multi-disciplinary treatment (MDT) was initiated to treat the patient’s primary concerns. Patients who met the diagnostic criteria for primary osteoporosis were included in OWFLS and received personalized anti-osteoporosis treatment planning based on the patient’s situation. The diagnostic criteria for primary osteoporosis were as follows: 1) T-score ≤−2.5 for BMD measured via DXA, 2) T-score for BMD of the proximal humerus, pelvis, or distal forearm between −2.5 and −1.0, 3) osteoporotic fracture of the hip or vertebrae. Patients with primary osteoporosis were included in the OWFLS database and received bone health education. In follow-up, the OWFLS liaison officer followed up with patients by telephone every 3 months to check on their treatment, improvement of symptoms, and whether they had experienced any fractures. The OWFLS liaison officer also encouraged patients to take active steps to treat their osteoporosis (Figure 1).

Figure 1.

Processes in the orthopedic ward fracture liaison services (OWFLS) model. MDT, multi-disciplinary treatment.

Patients

OWFLS is a small ward-based precision management model with patients originating from our orthopedic ward. Because OWFLS had not been implemented in our hospital before 2019, patients who were admitted to our orthopedic ward with primary osteoporotic fracture from January 2018 to January 2019 were selected as group A. Patients admitted from February 2019 to February 2020 were selected as group B; these patients were included in the OWFLS model. To control for selection bias, the above patients were randomly selected.

Inclusion criteria were as follows: patients aged 50 to 95 years; primary osteoporotic fracture; vertebral, hip and other (extremity) fracture; and no prior history of anti-osteoporosis medication treatment. The exclusion criteria were traumatic fracture, open fracture, secondary osteoporotic fracture owing to tumor; type 1 diabetes mellitus; adult patients with osteogenesis imperfecta, long-term untreated hyperthyroidism, hypogonadism, or early menopause (age <45 years), chronic malnutrition or malabsorption, or chronic liver disease.

Evaluation of OWFLS

In evaluating the model, we used the bone metabolism index testing rate and treatment rate (use of anti-osteoporosis drugs, i.e., bisphosphonates or denosumab). Patients undergoing anti-osteoporotic treatment were followed up at 3 months, 1 year, and 3 years to assess whether they had continued treatment, whether their symptoms had improved, and whether they had experienced a recurrent fracture.

Statistical analysis

Statistical data analysis was performed using IBM SPSS Version 26 (IBM Corp., Armonk, NY, USA). Welch’s t-tests were used for comparing patient ages. The chi-square test was applied to analyze differences according to patients’ sex, fracture site, and outcome of the OWFLS model. A p value <0.05 was considered statistically significant.

Results

A total of 240 patients were included in this study, 120 in group A (not included in OWFLS) and 120 in group B (included in OWFLS). There were no apparent differences in mean age (72.03 ± 9.51 years vs. 73.68 ± 8.00 years), sex (59 women vs. 64 women), or fracture site between group A and group B. The rates of bone metabolism index testing (50% [60/120] vs. 0%; p < 0.0001) and medication use (94.2% [113/120] vs. 64.2% [77/120]; p < 0.0001) were significantly higher after OWFLS implementation (Table 1; Figures 2, 3).

Table 1.

Comparison of characteristics, bone metabolism index testing rates, and treatment rates between groups.

Index	Group A (N = 120)	Group B (N = 120)	p value
Age	72.03 ± 9.51	73.68 ± 8.00	0.144
Sex (woman)	59	64	0.606
Fracture site
Vertebra	41	28	0.174
Hip	68	79
Other	11	13
Bone metabolism indicator detection rate	0 (0%)	60 (50.0%)	<0.0001
Treatment rate	77 (64.2%)	113 (94.2%)	<0.0001

Figure 2.

Comparison of bone metabolism index testing rates between groups.

Figure 3.

Comparison of treatment rates between groups.

There was no significant difference in the treatment adherence rate at 3 months between groups B and A (97.3% [110/113] vs. 93.5% [72/77]). However, the adherence rate at 1 and 3 years was better among patients included in OWFLS than among patients in group A (73.5% [83/113] vs. 51.9% [40/77], p = 0.003 and 57.5% [65/113] vs. 26% [20/77], p < 0.0001, respectively). There was no significant difference in the recurrence rate of bone pain between the two groups after 3 months. However, this rate was significantly lower among patients included in OWFLS than among patients in group A at 1 and 3 years (20.4% vs. 46.8%; 45.1% vs. 76.6%, respectively; p < 0.0001) (Table 2; Figures 4, 5).

Table 2.

Comparison of treatment adherence after 3 months, 1 year, and 3 years

Index		Group A (N = 77)	Group B (N = 113)	p value
3 months	Consistent treatment	72 (93.5%)	110 (97.3%)	0.273
	Bone pain recurrence	1 (1.3%)	1 (0.9%)	>0.99
	Fracture recurrence	0	0	–
12 months	Consistent treatment	40 (51.9%)	83 (73.5%)	0.003
	Bone pain recurrence	36 (46.8%)	23 (20.4%)	<0.0001
	Fracture recurrence	3 (3.9%)	3 (2.7%)	0.688
36 months	Consistent treatment	20 (26%)	65 (57.5%)	<0.0001
	Bone pain recurrence	59 (76.6%)	51 (45.1%)	<0.0001
	Fracture recurrence	8 (10.4%)	11 (9.7%)	>0.99

Figure 4.

Comparison of treatment adherence rates after 3 months, 1 year, and 3 years.

Figure 5.

Comparison of bone pain recurrence rates after 3 months, 1 year, and 3 years.

Discussion

The FLS model identifies disease and allows for patient assessment, stratification, and management using evidence-based guidelines.^14,16 FLS has been rolled out in many countries worldwide and has been proven effective in reducing mortality rates and substantially lowering associated medical costs. ²³ In China, there is inadequate management of patients with osteoporotic fractures and critically low treatment rates. The focus of Chinese orthopedic surgeons is often on surgery and postoperative rehabilitation; awareness about and attention to osteoporosis remain relatively poor. To address these issues, we proposed the OWFLS model.

The main objectives of OWFLS are to increase the rate of osteoporosis-related screening and treatment in patients with osteoporotic fracture, improve long-term medication adherence by establishing a patient database for regular follow-up after discharge, provide regular bone health education, and ultimately, to reduce the incidence of bone pain recurrence and re-fracture.

Our study demonstrated that the rate of bone metabolism index testing (50.0%) and the treatment rate (94.2%) were significantly improved after the implementation of OWFLS (p < 0.0001). Additionally, patient adherence was effectively improved and bone pain recurrence reduced (p < 0.0001). Before the implementation of OWFLS, our institution screened patients with osteoporosis by assessing BMD, but bone metabolism index tests were not conducted. After the implementation of OWFLS, bone metabolism index tests were carried out to evaluate the effect of drug treatment.

Most patients who start treatment will remain in treatment up to 5 years, according to retrospective administrative data collected on 888 patients. ²⁴ In our study, the probability of adherence to 3 months of treatment with bisphosphonates or denosumab was 93.5% and 97.3% in groups A and B, respectively, which was not significantly different (p = 0.273); the 1-year adherence rates were 51.9% and 73.5%, and the 3-year adherence rates were 26% and 57.5%, respectively. Thus, in comparison with those who did not receive OWFLS, adherence was better in patients included in OWFLS (p = 0.003, p < 0.0001, respectively). A large population-based study including 42,249 patients who did not receive FLS revealed that only 48% adhered to bisphosphonate treatment for 1 year, ²⁵ which is similar to the 1-year adherence rate among patients who did not receive OWFLS in our study (51.9%). Naranjo et al. discovered that the 1-year adherence rate for FLS recipients was 72%. ²⁶ Another study in Australia indicated that 74% of FLS recipients had adhered to bisphosphonate therapy at 1 year, ²⁷ which is comparable with the 1-year adherence rate among patients included in OWFLS in our study (73.5%). Remarkably, a study of FLS in France reported a 1-year adherence rate of less than 50% among patients aged 70 years or older; ²⁸ this may be owing to differences in study design. The 3-year adherence rate in an observational study of patients who did not receive FLS was 25%, ²⁵ which is comparable to the rate in our study (26%). Naranjo et al. observed that approximately 57% of patients adhered to treatment for 3 years, ²⁶ similar to our findings (57.5%).

The higher rate of adherence among patients included in OWFLS in this study is partly owing to bone health education provided to these patients. One study demonstrated the critical role of FLS staff in helping patients recognize the potential risks of osteoporotic fractures and encouraging adherence to medical advice. ²⁹ Furthermore, pre-treatment bone health education and follow-up visits also improved treatment compliance. ²⁷ There was no significant difference in the fracture recurrence rate between the groups in this study, which may be related to the small sample size. Our findings should be confirmed in studies with a larger sample size.

The advantage of OWFLS lies in that it is a small and precise management model based on the orthopedic ward of a single hospital, which is suitable for settings lacking human resources. In contrast, the minimal FLS model implemented by Lüthje et al. has a low cost, only 1.3% (EUR 85,726) of the total annual municipal expenditure, which makes it suitable for under-resourced regions. ³⁰ However, OWFLS enables orthopedic surgeons to emphasize the anti-osteoporotic management of patients with osteoporotic fractures. The above study proved that this model is feasible.

There are some limitations in this study. Currently, the bone metabolism indexes in our hospital are tested by other institutions, and it is difficult to obtain the results of these tests. Therefore, comparison of these indexes before and after the implementation of OWFLS was not performed. Second, the study was conducted in the orthopedic wards of our hospital, and the sample size was very small. Third, this was a retrospective study with loss to follow-up. Fourth, no patients included in this study had previously undergone anti-osteoporosis treatment. It is unknown whether previously treated patients would benefit from the FLS model. Fifth, mortality was not analyzed in this study. Finally, the administration route of anti-osteoporosis treatment for patients in this study was not standardized. Most patients received oral treatment, with only three patients using injection treatments.

Conclusions

OWFLS is a small-scale precision management model for identifying, assessing, treating, and following patients with osteoporotic fractures. In this study, the model improved the detection rate of bone metabolism indicators, the treatment rate, and patient adherence and reduced the recurrence rate of bone pain. OWFLS may be suitable for implementation in settings lacking human resources.