Introduction

Osteoporosis is a disease characterized by low bone mass and microarchitectural deterioration of bone tissue, resulting in bone fragility and propensity to fractures [1]. With the population aging, osteoporosis became increasingly common and is now a major public health concern. Most individuals are at risk of developing osteoporosis at some point during their lifetime [2]. In 2019, there were approximately 25.5 million women and 6.5 million men with osteoporosis in the EU27 + 2 (European Union plus the United Kingdom and Switzerland), about four times more women than men [3].

The most obvious and serious consequence of osteoporosis is fractures, which result in increased morbidity and mortality, and represent a major and growing economic burden on healthcare systems worldwide [4, 5]. In Portugal, it was estimated that the number of new osteoporosis-related fractures was 70,730 in men and women aged 50 years or more, in 2019 [3]. Although the most well studied fragility fractures are vertebral and hip fractures [6,7,8], several studies have shown that non-hip and non-vertebral fractures (e.g., lower leg, wrist, humerus, rib, elbow, or clavicle) account for most osteoporosis-related fractures and are associated with an increased risk of subsequent fracture, higher morbidity, higher mortality, and a significant increase in direct and indirect health costs [9,10,11].

More than three million osteoporosis-related fractures occur annually in the United States, costing an estimated $25.3 billion in 2025 [12]. The average direct costs during the first year after common fractures (hip, vertebral, and forearm) in Iran were estimated to be $3,030, $2,317, and $925, respectively [13]. In the EU27 + 2, the direct costs of incident osteoporosis-related fractures in 2019 were €36.3 billion [3]. Marques et al. highlight that the economic burden attributable to osteoporotic hip fractures in Portugal could be about €216 million in 2011 [14]. Although the cost of hip fractures in Portugal is relatively well-known [14,15,16], information about the costs of non-hip osteoporosis-related fractures is lacking, which is a major gap in knowledge given their prevalence. Quantification of fracture-related costs is needed to efficiently allocate resources, for example for fracture liaison services (FLS). FLS improves healthcare for the patient by reducing their risk of a new fracture, with a subsequent cost reduction for the healthcare system [17].

This study aimed to estimate the costs of incident non-hip osteoporosis-related fractures in postmenopausal women living in Portugal, focusing on direct costs from healthcare consumption in the first year following fracture, and distinguishing between the main types of fracture (vertebral, wrist, and others).

Methods

Type of study, population of interest, and methodology

This is a cost-of-illness type study, where the costs associated with non-hip osteoporosis-related fractures were estimated from a payer perspective for the year following the fracture. This study was focused on postmenopausal women, using the lower cut-off age of 50 years to define “postmenopausal”.

Throughout the study, non-hip fractures were stratified into three groups based on the location of the fracture: vertebral, wrist, and other fractures (clavicle, humerus, elbow, and lower leg). The same stratification was used in the scorecard for osteoporosis in Europe (SCOPE 2021) [3]. We estimated the number of fractures of each type occurring annually among postmenopausal women in Portugal and consulted with an expert panel to identify the healthcare resource consumption pattern of patients suffering each type of fracture, which we multiplied by the unit costs of each resource, to obtain the costs per patient in the year following a fracture. The number of women in each group of fractures was then multiplied by the costs per patient per year to obtain the total annual cost of incident non-hip osteoporosis-related fractures among postmenopausal women for the Portuguese National Health Service (NHS).

Data

Healthcare use

The following healthcare resources were considered: doctor visits, medical exams, hospitalization, outpatient care, and physiotherapy (e.g., etching and balance exercises, cryotherapy, spine manipulation techniques, ice and heat therapy). Essential technical aids (e.g., orthoses) prescribed in each case were also considered, even though they are purchased out-of-pocket by the patients (i.e., fall outside the payer’s perspective). It was not possible to consider nursing home/long-term care or medications because utilization of these is not clearly defined (e.g., nursing home care depends strongly on socioeconomic status and availability of informal care), and we lack individual-level utilization data. Considering that individual level utilization is not available, information on the typical healthcare resources used in the year following each type of fracture was gathered from a panel of experts: two rheumatologists, a physiatrist, and an orthopaedist. Experts discussed typical healthcare use based on their experience and were asked to reach a consensus.

Costs

Medical exams, hospitalization, outpatient care procedures, as well as physiotherapy were valued according to the national tariffs practiced in the NHS: those tariffs are based on Diagnosis Related Groups (DRG), in the case of hospitalization, and fee-for-service, in the case of medical exams, outpatient procedures, and physiotherapy (Portaria nº 254/2018, de 7 de setembro).

The prices of doctor visits in the Portuguese NHS are not defined in the same way as for diagnoses and treatment procedures. Specialty consultations were valued according to the terms defined by the Central Administration of the Health System for 2020, which specify the amounts paid to hospitals for each outpatient visit. The lowest price (€39.00) was adopted. For general practice visits, there is no public information on either cost or reimbursement prices. For this work, general practice visits were valued at half the price of a specialty consultation (€19.50).

The tariffs for X-rays varied by type of fracture. For other fractures, the average price between upper and lower limb fractures X-rays (i.e., hand and foot) was considered. The same rationale was applied to estimate the price of outpatient care and technical aids for other fractures. Technical aids purchased out-of-pocket by the patients were valued at market prices extracted from the website salusa.pt (also 2020 prices). The national tariffs defined in Portaria nº 254/2018 were also still applied in 2020, so all costs are at 2020 prices.

Numbers of fractures per year

The annual numbers of incident wrist, vertebral, and other non-hip fractures among postmenopausal women in Portugal were estimated from a population based Portuguese cohort—Epidemiology of Chronic Diseases (EpiDoC) study. EpiDoC is a longitudinal cohort enrolling 10,661 non-institutionalized individuals representative of the adult Portuguese population living in mainland Portugal, Azores, and Madeira islands [18]. A full description of the EpiDoC cohort is provided elsewhere [18, 19].

The sample used in the present study includes women aged 50 years and above who participated in the baseline assessment (2011–2013) and in the first follow-up wave (2013–2015) of the EpiDoC study (EpiDoc 2).

Incident non-hip osteoporosis-related fractures were defined as any new self-reported low impact non-hip fractures: fractures that resulted from a fall, from standing height or less, or that occurred in the absence of trauma, since the previous wave (EpiDoC 1, conducted in 2011–2013) [20]. The accuracy of self-reported osteoporosis-related fractures was previously shown to be acceptable [21,22,23]. For the purposes of this study, it is important to note that women diagnosed with bone metastasis or other bone metabolic diseases, such as Paget’s disease of bone, were excluded. Fractures of the face, skull, foot, fingers, and toes were also excluded, as they are usually unrelated with osteoporosis.

The numbers of incident fractures of each type were extracted (i.e., sample totals) and multiplied by the sampling weights to obtain population totals. The sampling weights take into account the stratified sampling design of the EpiDoC study and when applied to the sample, give the same distribution of sex, age, and region of residence (Nomenclature of Territorial Units for Statistics, NUTS II – seven regions in Portugal) as is observed for the total adult Portuguese population (according to the 2011 census).

Healthcare services used after a non-hip osteoporosis-related fracture – expert opinion panel

Experts considered that upon suffering an incident osteoporosis-related fracture, patients require X rays. The percentage of cases that require surgery varied with the type of fracture. Wrist fractures require surgery in about 25% of the cases, vertebral fractures in about 3% of the cases, and other fractures in about 5% of the cases. Moreover, about 5% of the cases of vertebral fractures require hospitalization for further study and observation, but no surgery. In the remaining cases, patients are treated on an outpatient basis (e.g., closed reduction, a non-operative procedure to put bone pieces back into their correct position and alignment).

All fractures should be followed by a rehabilitation program (i.e., physiotherapy) for appropriate recovery, although the exact procedures depend on the fracture site. According to best clinical practice [24], patients are meant to attend 36 sessions of physiotherapy (ideal scenario), although most patients attend fewer sessions (12 sessions), for various reasons such as cost or time constraints.

In the year following a fracture episode, patients typically attend five orthopaedics consultations, three physiatry consultations, and two rheumatology consultations (Table 1). Each rheumatology consultation also involves a check-up X-ray of the site of fracture.

Table 1 Standard healthcare services used in Portugal in the year following a non-hip osteoporosis-related fracture, by type of fracture
Full size table

In addition, different fractures require specific technical aids. Wrist fracture patients require wrist orthoses, vertebral fracture patients need support belts, and other fracture patients require corresponding orthoses (e.g., shoulder orthoses, leg orthoses).

Results

A total of 2,762 women aged 50 + were selected (Fig. 1).

Fig. 1
figure 1

Flowchart EpiDoC Study

Full size image

Costs in the year following a non-hip osteoporosis-related fracture

Blood tests encompass a list of 15 + items, which added up to €57.98. The tariff for a bone DEXA was €19.17.

The treatment of fractures requiring surgery involves pre-surgery routines and procedures such as blood tests, an anesthesiology consultation, a thorax X-ray, and an electrocardiogram, which were valued at €66.50 according to the national tariffs. Surgery prices also varied depending on the type of fracture as depicted in Table 2. Surgery was valued at €1,316.16 for wrist fractures (DRG for hand or wrist procedures), and €644.37 for other fractures (DRG for fractures & dislocations except for femur, pelvis, and back). Vertebral fractures may undergo two different surgical procedures. Among the 3% of the cases of vertebral fractures that require surgery, about 1% involve arthrodesis, valued at €3,205.63 (DRG for dorsal and/or lumbar arthrodesis except for resolution of the vertebral curvature), and 2% require vertebroplasty, valued at €1,573.22 (DRG for other procedures in the musculoskeletal system and/or connective tissue). For the 5% of the cases that require hospitalization for observation, costs were valued at €700.58 (DRG for other diagnoses of the musculoskeletal system and/or connective tissue).

Table 2 Unit costs of each healthcare service/medical equipment by type of fracture
Full size table

Outpatient procedures usually involve the reduction and immobilization of the fracture (and later the removal of the plaster), as well as X-rays. This was valued at €78.80 for wrist fractures, €26.50 for vertebral fractures, and €80.50 for other fractures. Physiotherapy includes a range of procedures that together added up to about €2,000 if all 36 sessions were taken. The price varied depending on the fracture site. Concerning technical aid prices at market values, a wrist orthosis costs €19.95, a support belt costs €84.90, and the average between the cost of a foot orthosis and the cost of a shoulder orthosis was €94.85 (Table 2).

Direct healthcare costs in the year following each type of fracture are reported in Table 3.

Table 3 Costs per patient in the year following fracture, by type of fracture
Full size table

When a patient suffers a fracture, the costs corresponding to doctor consultations were the same regardless of the type of fracture (€429), and include general practice as well as specialty consultations (Orthopaedics, Physiatry, and Rheumatology). Variations in the costs of medical exams reflected the different costs of X-rays, which vary with the type of fracture. Wrist fractures have the lowest costs with medical exams (€159.83), while vertebral fractures have the highest (€263.50).

For hospitalization costs, the percentages of fractures that require hospitalization were considered, as described previously. For this reason, the seemingly low hospitalization costs reflect the fact that, in most cases, wrist, vertebral, and other fractures do not require surgery.

The bulk of annual healthcare costs arose from physiotherapy, ranging between €1,807.20 for vertebral fractures and €2,298.60 for other fractures. These values reflect scenarios where the patient attended 36 physiotherapy sessions (with the exact treatment depending on the type of fracture). However, patients are likely to attend fewer sessions due to out-of-pocket costs, time constraints, lack of transportation, or other barriers. Lastly, technical aids were valued at €19.95 for wrist orthoses, €84.90 for support belts in the case of vertebral fractures, and €94.85 for various kinds of orthoses in the case of other fractures (weighted average).

The total estimated cost in the year following wrist fracture is €2,964.75 per patient, a vertebral fracture costs €2,709.52, and other fractures cost €3,096.35, in the ideal scenario that patients would attend the recommended 36 sessions of physiotherapy. Alternatively, if patients only attend on average 12 physiotherapy sessions, then the total healthcare costs per patient in the year following fracture decrease to €1,663.95 for wrist fractures, €1,504.72 for vertebral fractures, and €1,563.95 for other fractures (Table 4).

Table 4 Total healthcare costs in the year following fracture, by type of fracture and number of physiotherapy sessions
Full size table

Numbers and direct cost of non-hip osteoporosis-related fractures per year in Portugal

The annual number of incident non-hip osteoporosis-related fractures among women aged 50 + are 4,847 wrist, 3,491 vertebral, and 38,881 other fractures (Table 5).

Table 5 Numbers of non-hip osteoporosis-related fractures per year in Portugal
Full size table

Considering the annual costs per patient presented in the previous section and assuming that all 36 physiotherapy sessions are taken, the total cost is estimated at almost €144 million (Table 6). If the 12-session scenario is adopted, which is likely to be closer to actual utilization patterns, the total cost drops to about €74 million.

Table 6 Direct cost of non-hip osteoporosis-related fractures per year in Portugal
Full size table

Discussion

This study estimated the direct costs of non-hip osteoporosis-related fractures in postmenopausal women in Portugal. Nearly 5,000 wrist, 3,500 vertebral, and 39,000 other-site osteoporosis-related fractures occur per year among women aged 50 + in Portugal. Of these, other-site fractures account for the highest direct healthcare costs per patient (~ €3,096). Vertebral fractures represent the lowest annual direct healthcare costs at €2,710 per patient. Evidence of high costs of non-hip osteoporosis-related fractures is currently available for other countries worldwide, such as the United States of America [25, 26], China [27], South Korea [28] and Canada [29], and are in line with this study. The bulk of healthcare costs per patient per year were related to physiotherapy, as found in other developed countries [29]. Yet, in a recent study published by the same Canadian group, hospitalization accounted for a large portion of annual direct healthcare costs instead of physiotherapy [30]. The same was found by Tran et al. [31] and Rajabi et al. [13]. A possible explanation for the lower relative importance of physiotherapy in these studies is the lower coverage of physiotherapy by health insurance. In our case, physiotherapy was supported by NHS. In our conservative scenario, where patients undergo on average 12 physiotherapy sessions (rather than the 36 sessions in the ideal scenario), this study estimated that non-hip osteoporosis-related fractures among 50 + aged women cost about €74 million per year in direct healthcare costs alone (€144 million in the ideal scenario). Compared with the costs of other non-communicable diseases in Portugal, Gouveia et al. estimated a cost of €152 million associated with heart failure for the same demographic group (women 50 +) [32]. Thus, in line with other non-communicable diseases, osteoporosis has a major economic burden on society.

Comparisons with other studies are difficult due to differences in methodology and data sources. To the best of our knowledge, this is the first study in Portugal that has quantified the direct cost of non-hip osteoporosis fractures from a payer perspective.

One of the limitations of this study is the fact that the authors can only estimate costs based on available self-reported data on incident fractures in the years 2013–2015 from the EpiDoC cohort, combined with typical (not actual) utilization based on opinions of an informal panel of experts. Given that osteoporosis prevalence has shown an increasing trend, this is likely to result in the underestimation of the number of fractures, which again translates into conservative estimates of the costs of non-hip osteoporosis-related fractures in our study. Using EpiDoC data to estimate the numbers of non-hip osteoporosis-related fractures is still a major strength of our study because EpiDoC includes a large sample representative of the Portuguese population, and the accuracy of self-reported osteoporosis-related fractures has been shown to be acceptable [21,22,23]. Another source of underestimation is attrition, i.e., women in the EpiDoC cohort lost to follow-up, due to death or other reasons (Fig. 1).

This study calculated only the direct medical costs due to healthcare use and only in the year following non-hip osteoporosis-related fractures, again contributing to conservative estimates. Resources not considered include medication, other elective technical aids (e.g., bath aids, walking stick), social care (e.g., sickness benefits, disability pension, dependency supplement, formal caregiver support), and transportation. For these resources, it is difficult to draw a typical utilization profile without actual utilization data. Depending on the socioeconomic background, some individuals will purchase other equipment to make recovery more comfortable, and even hire home care services, while others may rely exclusively on informal care. The loss of productivity at work associated with absenteeism or caregiver burden due to disease loses relevance as this study focused on women close to retirement age, and a significant portion do not participate in the labour market. As for costs for informal caregivers, they are highly dependent on the underlying assumptions (e.g., the value of one hour of informal care) and may represent relatively small portions of the total [33, 34].

Lastly, the existence of comorbidities (and other covariates) was not considered, again because resource use data was obtained from an expert panel. This may further contribute to the underestimation of the costs associated with fragility fractures in Portuguese women because we have not taken into account the costs of possible decompensation of chronic diseases that could occur in a patient after suffering a fragility fracture. A recent Canadian study showed that in patients with fragility fractures cost increased with the number of comorbidities [30].

Conclusion

The results of this study point towards the relevance of preventing non-hip osteoporosis-related fractures among postmenopausal women, as they represent about €74 million per year in direct healthcare costs alone in Portugal. This reveals a substantial impact of treating non-hip osteoporosis-related fractures in the healthcare budget. These data should alert government entities to the importance of preventing first and subsequent osteoporotic fractures, for example through national campaigns to increase health literacy in osteoporosis and promote the implementation of various FLS in the Portuguese territory.