Fracture rates and economic outcomes in patients with osteoporosis prescribed risedronate gastro-resistant versus other oral bisphosphonates: a claims data analysis

In Europe, in 2010, it was estimated that 22 million women and 5.5 million men aged between 50 and 84 years old were living with osteoporosis [1]. In the USA, approximately 12.3 million individuals older than 50 years were estimated to have osteoporosis in 2020 [2]. These numbers are expected to continue to increase over the next decades with the aging of the population [1]. Men and women with osteoporosis have an increased risk of fractures or bone breaks which can substantially impact their quality of life [3]. Osteoporosis fracture may lead to chronic pain and disability and higher risk of early mortality and pose a substantial financial burden on patients, their family, and the health care systems [4‐7].

Oral bisphosphonates are the recommended first-line treatment for patients with osteoporosis in the USA and in Europe [8, 9]. Acting as antiresorptive agents, the use of oral bisphosphonate treatments has been associated with a significant reduction of the risk of fractures in patients with osteoporosis [7, 8, 10]. However, oral bisphosphonates are poorly absorbed, and lack of adherence to complex dosing instructions involving fasting and not lying down after taking the medication can further limit its absorption [11]. Real-world evidence indicates that more than 50% of patients do not follow all dosing instructions of the immediate-release formulation [12]. A poor absorption of bisphosphonates can result in suboptimal effectiveness and increased risk of fractures [13‐15], leading to significant utilization of health care resources and costs [4, 16, 17].

Risedronate gastro-resistant (GR) was developed with the goal of providing more convenient dosing instructions and reducing the risk of gastrointestinal (GI) side effects by eliminating the need for fasting, without impacting the overall bioavailability and efficacy of the treatment [11, 18]. The GR formulation has the advantages of a better absorption independent of food intake, which is expected to translate into improved efficacy and a reduction in fracture rates compared to other immediate-release oral bisphosphonates, including alendronate sodium, ibandronate sodium, and risedronate sodium. However, there is limited real-world data comparing fracture rates between patients treated with risedronate GR versus other oral bisphosphonates.

The main objective of the study was to quantify and compare fracture rates among women with osteoporosis treated with risedronate GR versus other oral bisphosphonates. Health care resource utilization (HCRU), health care costs (HCC), and medication persistence and adherence were also compared between the two treatment groups.

Results from the current study indicate that over an average observation period of more than 50 months, patients prescribed risedronate GR had lower rates of fracture and in particular a lower rate of spine fractures compared to patients prescribed other bisphosphonates. Of note, although the proportion of patients who had experienced a fracture at any site during the 6-month baseline period was the same in the two treatment cohorts, a slightly higher proportion of patients prescribed risedronate GR experienced a wrist/arm fracture, perhaps indicative of a higher risk of subsequent fractures.

As expected, the incidence of hip fracture, the most expensive fracture type [17], was the lowest among patients under 65 years old and increased sharply in older patients. Overall, the incidence of fractures more than doubled in patients ≥ 65 versus < 65 years old. In general, the lower rates of fractures observed in the present study compared to previously published results from observational studies are consistent with the relatively young age of the studied population [24‐26].

Indeed, in a study using claims data from commercially insured patients (all patients < 65 years old; mean age: 56.0 years old), the authors estimated incidence rates of fractures at any site between 15.4 and 16.1 per 1,000 patient-years for bisphosphonates [25]. In a similar study using claims from Medicare insured patients (all patients ≥ 65 years old; mean age, 78.2 years old), the authors estimated at 56.3 per 1,000 patient-years the incidence rate of fractures at any site [24].

Compared to patients prescribed other bisphosphonates, those who were prescribed risedronate GR were associated with a higher rate of treatment discontinuation, especially during the first few months after the index date. A potential explanation for this observation could be the 4 to 6 times higher out-of-pocket costs observed for risedronate GR than that for other bisphosphonates (mean [median] out-of-pocket costs per dispensing, 28–30-day supply, GR, $42.39 [$37.95] vs. other, $10.72 [$6.10]; 84–90-day supply, GR, $83.69 [$69.71] vs. other, $21.13 [$11.88]). This observation is further supported by a well-documented body of literature on the detrimental impact of increased out-of-pocket costs on medication compliance and persistence, in particular for chronic conditions [27‐32].

This study offers a unique perspective on the fracture rates of patients prescribed risedronate GR versus other oral bisphosphonates in real-world settings. Several randomized clinical trials have assessed the efficacy of bisphosphonates, and the body of evidence from these studies suggests no differential effect on fracture rates between bisphosphonates [33]. However, in real-world settings, patients’ behavior towards their medication may impact the ability of bisphosphonates to prevent fractures.

Lack of persistence and poor compliance are known challenges with bisphosphonates, all of which may impact the overall efficacy of bisphosphonates [14, 33‐35]. A systematic literature review of 89 prospective and retrospective observational studies examining patient adherence and persistence to oral bisphosphonates in adults with osteoporosis found that mean persistence rates ranged from 17.7 to 74.8% at 1 year and from 12.9 to 60.6% at 2 years. Similarly, compliance rates ranged from 31.7 to 72.0% at 1 year and from 34.5 to 47.9% at 2 years across studies [33]. Although reasons for the lack of persistence and compliance are likely multifactorial, complex dosing instructions as well as GI side effects have been reported as potential reasons [15, 33‐35].

Adding to the detrimental impact on bisphosphonates effectiveness of poor medication persistence and compliance is the poor compliance with dosing instructions. Absorption of oral bisphosphonate by the system is poor—usually less than 1% on average—and may be further limited by concurrent intake of food and beverage [11]. As such, to derive the full benefits of immediate-release formulations of oral bisphosphonates, patients are required to follow a strict dosing schedule—the drug is typically required to be taken on an empty stomach ≥ 30 to 60 min before the first food, drink, or other medications of the day. Furthermore, patients must not lie down for 30 min after taking the medication [36‐38].

Real-world evidence indicates that a significant fraction of patients do not follow the complex dosing instructions of the immediate-release formulation of bisphosphonates [12]. In a survey of 363 women with osteoporosis aged ≥ 55 years old, 40.5% reported not being compliant with dosing instructions ensuring adequate absorption, 34.7% were not compliant with dosing instructions preventing adverse events, and as many as 56.5% were not compliant with all dosing instructions [12]. Other external factors, such as the calcium content in tap water, may also interfere with the absorption of bisphosphonates [39].

The lower rates of fractures observed among patients in the risedronate GR cohort, despite their lower persistence on the medication, could be explained by its more convenient dosing administration. Indeed, among patients remaining on bisphosphonates, those on risedronate GR are more likely to receive an improved medication benefit because of its absorption irrespective of fasting compared to the complex and, hence often not followed, dosing administration of immediate-release bisphosphonates.

Likely reflecting their lower incidence of fractures, patients prescribed risedronate GR were associated with a lower utilization of inpatient services, which translated into lower inpatient costs. The amount of health care resources and the associated costs diverted by osteoporotic fractures are substantial [4, 16, 17, 40]. For instance, using claims data from the USA, Tran et al. found that the all-cause incremental yearly health care costs of commercially insured women with versus without incident osteoporotic fracture over a period of 5 years following the fracture were $77,370, $28,737, $23,155, $21,478, and $18,835, respectively [40]. Even a small decrease in the risk of fracture has thus the potential for significant health care resources and health care costs saving on the long term.

Results from this study should be interpreted in light of some limitations. First, given the retrospective nature of the data used, this study did not attempt to establish causal inference. Instead, findings should be interpreted as association-level results. Second, because the databases used include patients with a commercial health plan or a Medicare Supplemental health plan, the study sample may not be representative of the general osteoporosis population. This is underscored by the relatively young age of studied patients. Third, the higher out-of-pocket cost associated with risedronate GR compared to that of other bisphosphonates may be the source of a selection bias. Another potential source of confounding comes from the fact that the database used does not include the necessary clinical information to identify patients at a higher risk of fracture. Fourth, osteoporosis and other conditions considered in this study were identified based on diagnostic codes recorded on administrative claims used for billing purposes, which in some instance could reflect suspected rather than confirmed clinical diagnoses. It is also possible that not all diagnoses are systematically recorded on claims; thus, underreporting of certain conditions is also a possibility. Similarly, the incidence of spine fracture may be underestimated as most cases do not come to medical attention. Fifth, claims data only provide information on filled prescriptions, not on how medications are actually consumed. Finally, the HCC reported are likely to be an underestimation of the total costs resulting from fractures as they do not capture administrative costs of third-party payers for claims processing, out-of-pocket costs, indirect costs (e.g., productivity loss), or other intangible costs.

Despite their lower persistence, patients prescribed risedronate GR were associated with a lower incidence of fractures compared to those prescribed other oral bisphosphonates, leading to lower utilization of inpatient services which translated into lower inpatient costs. These results are consistent with the hypothesis that the more convenient dosing administration of risedronate GR could improve the absorption of bisphosphonate and help patients with osteoporosis get an improved benefit of treatment.