Estimating hospital inpatient cost-savings with sucroferric oxyhydroxide in patients on chronic hemodialysis in five European countries: a cost analysis

Abstract

Aims

Hyperphosphatemia is common among patients with advanced chronic kidney disease (CKD) undergoing dialysis. The iron-based phosphate binder (PB), sucroferric oxyhydroxide (SO), has a low daily pill burden and is indicated for the control of serum phosphorus in these patients. In a retrospective database study, hemodialysis patients switched to long-term SO therapy had fewer hospitalizations compared with patients switched to other PB therapies. This economic analysis aimed to quantify potential cost-savings of reduced hospitalizations associated with SO for healthcare systems in five European countries.

Materials and methods

All-cause hospital admissions incidence data were sourced from a real-world retrospective database study comparing adult, in-center hemodialysis patients maintained on 2 years of SO therapy (mSO) versus patients who discontinued SO (dSO) within 90 days of their first prescription and switched to other PBs. A literature search was conducted to determine the cost per hospital admission for dialysis patients in the healthcare setting of each European country. A cost-model combined the incidence rate of all-cause hospital admissions and the cost per admission to estimate the country-specific inpatient costs for the mSO and dSO groups.

Results

Annual inpatient cost-savings per patient in the mSO group versus the dSO group were €1,201, €2,097, €2,059, €1,512, and €3,068 in France, Germany, Italy, Spain, and the UK, respectively. When annual PB drug costs per patient were considered, the net annual economic cost-savings per patient were €327, €1,585, €1,022, €1,100, and €2,204, respectively.

Limitations

Hospital admissions data used in the analysis were observational in nature and derived from a US hemodialysis patient population; the effect of SO therapy on hospitalization rates for US and European hemodialysis patients may differ. The analysis did not consider indirect healthcare costs associated with hospitalizations.

Conclusion

SO therapy may offer substantial inpatient cost-savings by reducing all-cause hospital admissions attributable to uncontrolled hyperphosphatemia.

Keywords:

• Chronic kidney disease
• cost analysis
• hospitalization
• phosphate binders
• hyperphosphatemia

JEL CLASSIFICATION CODES:

• D61
• D6
• D
• A11
• A1
• A

Introduction

Hyperphosphatemia is a common electrolyte disorder in chronic kidney disease (CKD), and has a high prevalence in patients with end-stage renal disease (ESRD) undergoing hemodialysis^1,2. In hemodialysis patients, the Kidney Disease Outcomes Quality Initiative (KDOQI) clinical guidelines recommend maintaining serum phosphorus levels between 3.5 and 5.5 mg/dL (1.13–1.78 mmol/L)³. However, observational data show over 40% of European hemodialysis patients in the real-world setting have serum phosphorus levels exceeding the upper KDOQI limit (>5.5 mg/dL [1.78 mmol/L])⁴. Elevated serum phosphorus levels are associated with a higher risk of mortality and other adverse outcomes, including vascular calcification, cardiovascular events, and hospitalizations among dialysis patients^5–9.

The principal strategies to manage hyperphosphatemia in patients with ESRD include dietary phosphate restriction, removal of phosphate by dialysis, and oral phosphate binder therapy^10,11. Adequate dietary phosphate restriction is difficult to achieve in clinical practice as it requires extensive patient education and it may also lead to malnutrition¹², while conventional hemodialysis treatment does not remove sufficient quantities of ingested phosphate¹³. As a result, the majority of dialysis patients also require treatment with oral phosphate binders in order to control their serum phosphorus levels^2,14. These agents lower serum phosphorus by binding dietary phosphate in the gastrointestinal tract and preventing its systemic absorption^15,16. Several types of phosphate binder are available for clinical use, and each of these agents has its own advantages and limitations (reviewed previously^10,17). Calcium-based phosphate binders are considered effective and inexpensive; however, regular administration of these agents may induce calcium overload, hypercalcemia, and vascular calcifications and has led to increased use of non-calcium based phosphate binders in clinical practice^1,15,18.

Most of the available phosphate binders are associated with a high daily pill burden^19,20, with some patients required to take several phosphate binder pills with each meal, in addition to other concomitant medications^19,20. A high phosphate binder pill burden has been linked with reduced treatment adherence among dialysis patients, which, in turn, has been associated with higher serum phosphorus levels²⁰. As such, the availability of effective phosphate binders with a low daily pill burden may improve adherence to treatment, allowing better control of hyperphosphatemia. Furthermore, a lower phosphate binder pill burden could also enhance adherence to other oral medications such as anti-hypertensives and antiplatelet drugs, which may reduce hospitalizations not specifically related to uncontrolled serum phosphate concentrations.

Sucroferric oxyhydroxide (SO, Velphoro) is a potent non-calcium, iron-based phosphate binder indicated for the control of serum phosphorus levels in adult patients with CKD undergoing hemodialysis or peritoneal dialysis^21–23. The efficacy and safety of SO was confirmed by a global, open-label, randomized Phase 3 registration trial involving 1,059 patients who were randomized (2:1) to SO (n = 710) or sevelamer carbonate (n = 349) for 24 weeks. Eligible patients who completed the initial Phase 3 trial could enter a 28-week extension study continuing on their randomized treatment for up to 1 year. During the Phase 3 trial and its extension study, SO demonstrated similar efficacy to sevelamer carbonate for reduction of serum phosphorus levels, but achieved this with a lower daily pill burden (3.3 pills/day versus 8.7 pills/day, respectively) and better adherence over the 1-year treatment period^24,25.

This effectiveness of SO in the real-world setting has subsequently been demonstrated by 6- and 12-month retrospective database analyses of hemodialysis patients in the US who were switched to SO therapy from another phosphate binder therapy as part of routine clinical care^26,27. These studies demonstrated that, in comparison with their previous phosphate binder(s), switching to SO monotherapy was associated with a ∼2-fold increase in the proportion of hemodialysis patients achieving target serum phosphorus (≤5.5 mg/dL [1.78 mmol/L]) and a > 50% reduction in daily phosphate binder pill burden^26,27. A potential limitation of these analyses was the lack of a concurrent active control group. To address this, Coyne et al.²⁸ recently performed an observational analysis that utilized a novel study design in which adult in-center hemodialysis patients maintained on SO therapy for 2 years were compared with an active concurrent control group of patients who discontinued SO within 90 days of first prescription and were switched to other phosphate binder(s). The results showed patients maintained on SO were more likely to achieve target serum phosphorus levels and used 50% fewer phosphate binder pills/day than patients who discontinued SO for other phosphate binders. Another key finding of the study was that patients maintained on SO therapy for 2 years had fewer all-cause hospital admissions (35.6 fewer per 100 patient-years) compared to those who received other phosphate binder therapies²⁸. The results from this US study can be used to inform hyperphosphatemia management practices in other countries, and this is the only comparative study to have evaluated hospitalization rates that could be applied to the present cost analysis.

The present study reports the results of an economic evaluation analysis using the hospitalization rates reported in the recent observational analysis by Coyne et al.²⁸ to quantify potential differences in inpatient cost implications attributed to different phosphate binders in five European countries (France, Germany, Italy, Spain, and the UK).

Methods

Hospital admission rates (US retrospective cohort study)

Data for all-cause hospital admission rates were obtained from a published retrospective comparative cohort study by Coyne et al.²⁸ The study design has been previously described. In brief, the analysis included adult (≥ 18 years) in-center hemodialysis patients who were initially prescribed SO as part of routine care at Fresenius Kidney Care facilities in the US (1 April 2014–1 April 2015). All patients included in the analysis received phosphate binder prescriptions for a 2-year period. Comparisons were performed between patients who completed 2 years of continuous SO therapy (designated the maintenance SO [mSO] group, n = 222) and an active control group consisting of patients who discontinued SO within 90 days of initial prescription and then switched to other phosphate binder regimens (the discontinued SO [dSO] group, n = 596). All patients included in the analysis were required to have been receiving phosphate binder therapy at baseline (defined as the 3-month period prior to SO prescription) and also had serum phosphorus laboratory values available at baseline and the final quarter of the 2-year follow-up period (Q8). Phosphate binder doses were administered and modified at the discretion of the treating physician. Phosphate binder regimens taken by patients in the dSO group following discontinuation of SO were: sevelamer (33.2% of 596 patients), calcium acetate (12.4%), lanthanum carbonate (2.3%), ferric citrate (0.2%), switch between phosphate binders (36.4%), and phosphate binder polytherapy (15.5%)²⁸.

Generalized equation models with exchangeable correlation structures were used to evaluate covariate-adjusted hospital admissions between mSO and dSO patient groups. A zero-inflated Poisson distribution model was applied to analyze hospitalization counts, while a zero-inflated negative binomial distribution model was utilized to evaluate length of hospital stay. The impact of demographic and clinical baseline variables as potential independent predictors or confounders of hospital admission rates were also analyzed. Statistical analysis was performed for all hospital admissions and for admissions lasting longer than 24 h²⁸. The analysis demonstrated that following adjustment for baseline demographic and clinical variables, the incidence rate of all hospital admissions (per 100 patient-years) was lower in the mSO group compared with the dSO group (108.3 vs 143.9 respectively), corresponding to 35.6 fewer hospital admissions per 100 patient-years for mSO patients (incidence rate ratio = 0.752 [standard error = 1.138]; p = 0.02)²⁸. Similarly, the adjusted incidence rate of hospital admissions lasting longer than 24 h per 100 patient-years was lower in the mSO group vs SO group (91.8 vs 131.4, respectively) corresponding to 39.6 fewer hospital admissions per 100 patient-years for mSO patients (incidence rate ratio = 0.70 [standard error: 1.145]; p = 0.006). The adjusted incidence rates of hospital admissions and length of hospitalizations reported by Coyne et al.²⁸ are presented in Supplementary Table S1.

Cost of hospital admissions data sources

To identify source data for hospital admission costs, a literature search of the PubMed database (https://www.ncbi.nlm.nih.gov/pubmed) was performed to identify observational studies that included CKD patients and reported inpatient resource use data (search period: from inception to May 2020). The key search terms from the PubMed literature search are described in Supplementary Table S2. Results from the literature search were sorted by country and reviewed to identify relevant data for costs per hospital admission among patients with CKD undergoing dialysis. For countries where no observational study data were identified, an additional search of government websites reporting diagnosis-related group (DRG) costing sources was performed. In cases where multiple costing sources were identified, the most recently published cost data were selected. All hospital admission costs identified from these studies were adjusted to 2020 prices (€)²⁹. The GBP/EUR exchange rate used (set by the Bank of England on 23 September 2021) was 1.1628³⁰. Hospital costs were adjusted for purchasing parity prices (PPPs) to allow cross-country comparisons³¹. The derived costs per hospital admission for CKD patients undergoing dialysis were used to estimate the inpatient budget impact of using SO versus other phosphate binders in France, Germany, Italy, Spain, and the UK.

Data modelling of inpatient costs

The present analysis used a cost modelling approach to estimate annual inpatient costs for dialysis patients treated with SO versus other phosphate binders, based on the incidence of hospitalization admissions for mSO and dSO patients groups reported in the retrospective study by Coyne et al.²⁸ and the estimated costs per hospital admission in each of five European scope countries, calculated using data obtained from the literature search. This study had a 1-year analytic horizon.

To quantify annual inpatient costs per patient for the mSO and dSO groups, the mean incidence rate of hospital admissions (Equation 1; Table 1) was multiplied by the cost per admission in France, Germany, Italy, Spain, and the UK (Equation 2; Table 1). Subsequently, the difference in annual inpatient costs per patient between patients in the mSO and the dSO groups was calculated to quantify potential annual inpatient cost-savings per patient associated with SO monotherapy (Equation 3; Table 1). The inpatient costs calculated from the cost model were reported per year per patient, whereas the initial retrospective cohort study by Coyne et al.²⁸ reported the mean incidence rate, incidence rate differences, and 95% confidence intervals of hospital admissions per 100 patient-years lived for the mSO group and dSO groups.

Table 1. Equations for: (1) mean incidence rate of hospital admissions, (2) cost per admission in the five European countries, (3) inpatient cost-savings per patient per year, (4) incremental phosphate binder drug costs, and (5) net economic effect.

Equations
1	${\mathrm{ }\overline{IR}}_{j\mathrm{ }}=\mathrm{ }\frac{1}{n}{\sum }_{i=1}^n\frac{{\mathit{Hospital}\mathrm{ }\mathit{admissions}}_i}{{\mathit{Patient}\mathrm{ }\mathit{years}\mathrm{ }\mathit{lived}}_i}\mathrm{ }$
2	${\mathit{InC}}_{jk}\mathrm{ }=\mathrm{ }{\mathrm{ }\overline{IR}}_{j\mathrm{ }}\times \mathrm{ }{HC}_k$
3	Inpatient cost – difference (cost – savings) per patient k (per scope country k) = InC mSO k – InC dSO k
4	Incremental phosphate binder drug costs (drug costs difference) per patient k (per scope country k) = (Annual drug costs per patient mSO k – Annual drug costs per patient dSO k)
5	Net economic effect (per scope country k) = Inpatient cost-savings per patient k (benefit) – Incremental phosphate binder drug costs per patient k (costs)

Abbreviations. ${ \overline{IR}}_j,$ Mean incidence rate of hospital admissions per patient per study group (j); j, Study group (mSO or dSO); n, Population in each study group; i, Study subjects; k, Scope country; InC_jk, Inpatient costs per patient, per study group (j) and per scope country (k); HC_k, Cost per hospital admission per scope country (k).

Annual phosphate binder drug costs were quantified per patient for the mSO group and dSO groups, based on the composition of prescribed phosphate binder regimens observed in the study by Coyne et al.²⁸ and on the daily doses and acquisition costs presented in Supplementary Table S3. Subsequently, incremental annual phosphate binder drug costs and differential hospital admission costs per patient were quantified for the mSO versus the dSO group (Equation 4; Table 1). Annual cost-savings from reduced hospital admissions (benefits) for the mSO group compared to the dSO group were deducted from the corresponding incremental annual phosphate binder drug costs (costs) to estimate the “net annual economic effect” (i.e. benefits − costs) per patient resulting from the comparison of the two groups (Equation 5; Table 1).

Costs were not discounted because of the 1-year analytic horizon of this present study. A univariate sensitivity analysis for the net annual economic effect per patient attributed to SO per country was also conducted. This sensitivity analysis utilized the 95% CI limits for the incidence rate of hospital admissions difference reported in the analysis by Coyne et al.²⁸ Hence, a range of potential net annual economic effects per patient were quantified by taking into account the lower and upper 95% CI values for incidence rate differences in the adjusted rate of hospital admissions between the mSO and dSO study groups. Further sensitivity analysis was conducted evaluating the impact of ±10% variation of annual drug costs on the results.

Results

Country-specific costs per hospital admission

The costs per hospital admission (adjusted to 2020 prices) for dialysis patients in each country, and the data sources they were derived from, are summarized in Table 2. Hospital admission costs were lowest in France (€3,374) and highest in the UK (€8,231).

Table 2. Inpatient costs: cost per hospital admission.

Country	Description of inpatient cost input	Cost per hospital admission in 2020 prices adjusted for purchasing parity prices (PPPs)
France	The economic evaluation report of Haute Autorité de Santé for ESRD in France (2014) reported the average monthly cost of “all other hospitalizations” for patients treated with in-center dialysis and the average hospital admissions per year. Subsequently, cost of “other hospitalizations” per dialyzed patients was deduced.	€3,374^32
Germany	Obtained from a retrospective study assessing the costs of CKD patients at stages 3 and 4 and on dialysis in Germany. Patients on HD or PD incurred annual, all-cause hospitalization costs of €10,029. The annual number of hospital admissions per patient was 1.9 admissions, which results in a cost per hospital admission of €5,890.	€5,890^33
Italy	The study by Roggeri et al. reported weighted average costs per hospital admission (in 2011 prices) in the Piedmont region. The study collected cost data for all-cause hospital admissions for patients undergoing HD or PD.	€5,783^34
Spain	DRG code 460 for renal insufficiency.	€4,246^35
UK	Observed costs by CKD stage and by comorbidity were reported in the “Study of Heart and Renal Protection” (SHARP), a randomized trial that prospectively collected information on kidney disease progression, serious adverse events, and hospital care use in patients with moderate-to-severe CKD. The cost used was the incremental cost per hospital admission for a major vascular event for patients on HD.	€8,231^36

Abbreviations. CKD, chronic kidney disease; DRG, diagnosis-related group; ESRD, end-stage renal disease; HD, hemodialysis; PD, peritoneal dialysis.

Inpatient cost-savings

Estimated annual inpatient costs per patient for all hospital admissions for patients who switched to phosphate binders other than SO (dSO group) were higher than those who were maintained on SO monotherapy (mSO group) (Table 3). The resulting annual inpatient cost-savings per patient for all hospital admissions for the mSO group versus the dSO group were estimated as €1,201, €2,097, €2,059, €1,512, and €3,068 in France, Germany, Italy, Spain, and the UK, respectively.

Table 3. Estimated annual drug costs, inpatient costs, and total annual costs per patient between treatment groups per country (all admissions analysis [EUR 2020]).

	Annual PB drug costs per patient*			Annual inpatient costs per patient			Total annual costs per patient
Country	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Incremental drug costs (drug costs difference) per patient	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Inpatient cost difference per patient**	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Net economic effect***
France	€1,763	€889	€874	€3,654	€4,855	−€1,201	€5,416	€5,743	−€327
Germany	€2,508	€1,996	€512	€6,379	€8,476	−€2,097	€8,888	€10,472	−€1,585
Italy	€1,949	€913	€1,036	€6,263	€8,322	−€2,059	€8,212	€9,235	−€1,022
Spain	€1,414	€1,003	€411	€4,599	€6,110	−€1,512	€6,013	€7,114	−€1,100
UK	€2,622	€1,758	€864	€9,334	€12,402	−€3,068	€11,957	€14,161	−€2,204

*Drug costs and inpatient costs were adjusted for PPPs.

**Calculated based on incidence rate of hospital admissions per patient, per year: SO: 1.08; other PBs: 1.44 from Coyne et al.²⁸

***Positive values suggest additional drug costs for the mSO group.

Abbreviations. dSO, patients who discontinued sucroferric oxyhydroxide and were treated with non-sucroferric oxyhydroxide phosphate binder; mSO, patients who received 2 years of maintenance therapy with sucroferric oxyhydroxide; PB, phosphate binder.

Estimated annual inpatient costs per patient per country for the two study groups based on the incidence of hospital admission rates for patients hospitalized longer than 24 h are summarized in Table 4. This analysis also showed annual inpatient costs per patient were lower in the mSO group than the dSO group, resulting in estimated annual inpatient cost-savings per patient of €1,336, €2,333, €2,290, €1,682, and €3,413 in France, Germany, Italy, Spain, and the UK, respectively. Overall, the cost model estimated average inpatient cost-savings with SO that corresponded to 25% and 30% of the estimated total hospital costs when considering all hospital admissions and hospital admissions with length of stay longer than 24 h, respectively.

Table 4. Estimated annual drug costs, inpatient costs, and total annual costs between treatment groups per country (admissions analysis > 24 h [EUR 2020]).

	Annual PB drug costs per patient*			Annual inpatient costs per patient			Total annual costs per patient
Country	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Incremental drug costs (drug costs difference) per patient	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Inpatient cost difference per patient**	Mono-therapy SO (mSO)	Patients switching to other PBs (dSO)	Net economic effect***
France	€1,763	€889	€874	€3,097	€4,433	−€1,336	€4,860	€5,322	−€462
Germany	€2,508	€1,996	€512	€5,407	€7,740	−€2,333	€7,916	€9,736	−€1,820
Italy	€1,949	€913	€1,036	€5,309	€7,599	−€2,290	€7,258	€8,512	−€1,254
Spain	€1,414	€1,003	€411	€3,898	€5,580	−€1,682	€5,313	€6,583	−€1,270
UK	€2,622	€1,758	€864	€7,912	€11,325	−€3,413	€10,534	€13,083	−€2,549

*Drug costs and inpatient costs were adjusted for PPPs.

**Calculated based on incidence rate of hospital admissions per patient, per year: SO: 0.92; other PBs: 1.31 from Coyne et al.²⁸

***Positive values suggest additional drug costs for the mSO group.

Abbreviations. dSO, patients who discontinued sucroferric oxyhydroxide and were treated with non-sucroferric oxyhydroxide phosphate binder; mSO, patients who received 2 years of maintenance therapy with sucroferric oxyhydroxide; PB, phosphate binder.

When annual phosphate binder drug costs per patient were considered, cost-savings persisted. Cost-savings from reduced annual inpatient costs per patient in the mSO group were greater than the incremental annual phosphate binder drug costs per patient, resulting in net annual economic effects per patient that favored the mSO group versus the dSO group both for all hospital admissions (Table 3) and hospital admissions longer than 24 h (Table 4). The net annual economic cost-savings per patient were €327, €1,585, €1,022, €1,100, and €2,204 for all hospital admissions, and €462, €1,820, €1,254, €1,270, and €2,549 for hospital admissions longer than 24 h in France, Germany, Italy, Spain, and the UK, respectively. The results show that for every additional €1 spent on SO, high returns on investment, ranging from €1.4 to €4.6, may be expected for the healthcare settings under study.

The results of the sensitivity analysis to evaluate how changes in the incidence rate ratios of all hospital admissions and hospital admissions longer than 24 h affected inpatient costs in each country are summarized in Tables 5 and 6, respectively. In the observational study the estimated 95% CI for the incidence rate ratios were persistently lower than 1.0; hence, the results show persistent cost-savings for mSO patients, compared with dSO patients. In the sensitivity analysis the net economic effects were shown to favor the mSO group for the majority of extreme scenarios performed. The results were not sensitive to the variation of annual drug costs and cost-savings for mSO patients persisted for the majority of scenarios.

Table 5. One-way sensitivity analysis for the net economic effect* (all admissions).

	Sensitivity analysis for the difference in the incidence of admissions**		Sensitivity analysis for the difference in annual PB drug costs
All admissions	Minimum	Maximum	−10%	10%
France	€273	−€762	−€415	−€240
Germany	−€536	−€2,345	−€1,636	−€1,533
Italy	€7	−€1,768	−€1,126	−€919
Spain	−€345	−€1,648	−€1,142	−€1,059
UK	−€670	−€3,316	−€2,290	−€2,118

*Positive values suggest additional drug costs for the mSO group.

**All admissions incidence rate difference adjusted for covariates: Mean (95% CI): −35.6 (−17.8, −48.5) per 100 patient-years.

Abbreviations. CI, confidence interval; mSO, patients who received 2 years of maintenance therapy with sucroferric oxyhydroxide; PB, phosphate binder.

Table 6. One-way sensitivity analysis for the net economic effect* (admissions > 24 h).

	Sensitivity analysis for the difference in the incidence of admissions**		Sensitivity analysis for the difference in annual PB drug costs
All admissions	Minimum	Maximum	−10%	10%
France	€115	−€907	−€550	−€375
Germany	−€813	−€2,598	−€1,872	−€1,769
Italy	−€265	−€2,017	−€1,357	−€1,150
Spain	−€544	−€1,831	−€1,311	−€1,229
UK	−€1,075	−€3,686	−€2,635	−€2,462

*Positive values suggest additional drug costs for the mSO group.

**Admissions > 24 h incidence rate adjusted for covariates: Mean (95% CI): −39.6 (−22.5, 52.8) per 100 patient-years.

Abbreviations. CI, confidence interval; mSO, patients who received 2 years of maintenance therapy with sucroferric oxyhydroxide; PB, phosphate binder.

Discussion

This study used a cost model approach to estimate inpatient cost-savings associated with use of SO versus other phosphate binders among patients undergoing hemodialysis in five European countries (France, Germany, Italy, Spain, and the UK). It found the lower rate of all-cause hospitalizations previously reported for hemodialysis patients treated with SO²⁸ translated into substantial in-patient cost-savings in each of these countries. The present analysis estimated annual inpatient cost-savings per patient for five countries in Europe, which varied from €1,201 in France to €3,068 in the UK for patients treated with SO monotherapy (mSO group) versus patients who switched to phosphate binders other than SO (dSO group). When annual phosphate binder drug costs per patient were considered, cost-savings persisted: net annual economic cost-savings per patient varied from €327 in France to €2,204 in the UK in the mSO group versus the dSO group.

Given that hyperphosphatemia increases the risk of vascular calcification, cardiovascular events, cardiovascular mortality and hospitalizations^5–9, and hyperphosphatemia is associated with higher all-cause and cardiovascular mortality^37,38, reducing mortality and hospitalizations is an important public health goal that also promotes efficient use of healthcare resources. Traditional phosphate binders pose a high pill burden and low tolerability leading to poor adherence and mild-to moderate benefits in achieving and maintaining serum phosphate levels targets^20,39,40. Moreover, calcium-based phosphate binders contribute to an increase in calcium phosphate product, associated with high risk of cardiovascular disease^41–43. Newer iron-containing phosphate binders therefore offer potential benefits, such as a lower pill burden with SO and improved iron parameters with ferric citrate, and the biggest challenge to efficacy for these newer phosphate binders is non-adherence⁴⁴. Data reporting the distinct benefits of phosphate binders on hard outcomes such as mortality or hospitalizations are, however, scarce^44,45.

The lower rate of hospital admissions observed for patients treated with SO monotherapy versus other phosphate binder therapy reported by Coyne et al.²⁸ may have been due to better control of hyperphosphatemia in the former treatment group. The greater improvements in serum phosphorus control with SO may, in part, be explained by improved treatment adherence to a phosphate binder treatment that had a lower daily pill burden. Adherence studies in hemodialysis patients indicate that higher pill burden may be inversely related to adherence, and that poor adherence is associated with higher serum phosphorus levels⁴⁶. In the phase 3 trial and its extension study, SO demonstrated similar efficacy to sevelamer carbonate, but with a lower pill burden and better adherence^24,25. In the retrospective study by Coyne et al.²⁸, improvements in serum phosphorus control among the mSO group was achieved with ∼50% fewer phosphate binder pills versus the dSO group (5.1 SO pills/day vs 10.9 non-SO phosphate binder pills/day). The potential impact of pill burden on treatment adherence and consequently control of hyperphosphatemia suggests that reducing daily pill burden in CKD patients on hemodialysis while maintaining efficacy for serum phosphorus control may ultimately result in fewer hospital admissions and help to optimize health service resource use.

Further evidence that treatment with SO may reduce the risk of hospitalizations compared with other phosphate binder therapies is provided by a recent observational analysis of 24 end-stage renal disease seamless care organizations in the US⁴⁷. This study analyzed hospital admission rates of dialysis patients prescribed various phosphate binder therapies over a 3-year period from 2016 to 2018. The results showed a lower hospitalization rate (per 100-member months [MM]) among patients treated with SO (7.97 per 100-MM), compared with those prescribed calcium acetate (11.28 per 100-MM), sevelamer (10.52 per 100-MM), ferric citrate (9.54 per 100-MM), or lanthanum carbonate (8.86 per 100-MM).

There are several limitations in the methodological approach of the present analysis that should be considered when interpreting the evidence and generalizability of the findings. First, in the absence of data for European hemodialysis patients, the hospital admission rates applied in the present analysis were obtained from an observational study conducted on US hemodialysis patients prescribed SO therapy. Hospitalization rates may differ between European and US hemodialysis patients due to differences in patient characteristics and clinical management practices for hyperphosphatemia. Second, outcomes were derived from a retrospective observational analysis that utilized existing electronic clinical records, and therefore data relating to treatment indication, adherence or tolerance to phosphate binder therapies were not available. Specific reason(s) for SO discontinuation among dSO patients were not captured, but may have included lack of effectiveness, phosphate binder treatment intolerance, non-adherence to therapy, a lack of insurance coverage, and out-of-pocket costs. Although prescription data can provide an accurate insight into prescribed pill burden, these data cannot be used as a surrogate for patient adherence to prescribed regimens. Furthermore, the observational, real-world evidence described here is based on a limited number of clearly defined hemodialysis patients in the US that had recently switched phosphate binder therapy as part of routine care. The switch in phosphate binder therapy could be attributable to a multitude of factors, including tolerability, suggesting these results might not be applicable to all CKD patients. The clinical data on which this analysis was performed were obtained for all-cause hospital admissions in a cohort of patients with elevated serum phosphorus at baseline. The present analysis applies the cost per hospital admission based on patients diagnosed with CKD and undergoing dialysis. As specific data relating to the actual cause of hospital admission (e.g. cardiovascular or infection-related) were not available, there is the possibility that our estimates have over- or underestimated the actual cost of hospital admissions. Moreover, the analysis described here was based on direct costs only and did not consider indirect costs from lost productivity of the individual who was hospitalized or their caregivers; hence, the cost estimates may be underestimated from the societal perspective.

With the estimated annual costs of phosphate binders ranging from $750 million¹⁶ to $1 billion worldwide in 2010⁴⁸, it is essential to determine the most cost-effective way to treat hyperphosphatemia in patients with ESRD. Recent UK economic analyses of SO, that utilized data from its clinical trial program, demonstrated the cost-effectiveness of SO compared with sevelamer carbonate^49,50. Specifically, the results from these economic analyses indicated that patients accrued 2.826 quality-adjusted life-years (QALYs) with SO and 2.835 QALYs with sevelamer, and annual cost-savings with SO were £744 per patient based on total annual direct costs (£2,140 and £2,884 per patients for SO and sevelamer, respectively)⁵⁰. The net medicines budget impact of SO in Scotland is expected to result in savings of £3,000 in year 1 and £10,000 in year 5. These economic analyses show SO results in savings for the healthcare system without compromising efficacy. The US observational study by Coyne et al.²⁸ has shown the potential of SO to increase the efficiency of the healthcare system by reducing hospitalizations and the associated costs. Further country-specific, economic studies are needed to evaluate the clinical and economic benefits of SO.

Conclusions

SO has been shown to be an effective phosphate binder therapy for the management of hyperphosphatemia. This analysis, based on retrospective observational data, shows treatment with SO may result in substantial inpatient cost-savings by reducing hospital admission rates and hospitalization costs attributable to uncontrolled hyperphosphatemia among hemodialysis patients with CKD. These potential inpatient cost-savings observed with SO monotherapy may help to optimize healthcare resource utilization. Further real-world evidence studies across different healthcare settings would confirm whether SO therapy reduces hospitalization rates among hemodialysis patients, compared with other phosphate binders.

Transparency

Declaration of financial/other interests

JAH has nothing to disclose; MS has nothing to disclose; ARdA, TS, and SW are all employees of Vifor Pharma.

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

A reviewer on this manuscript has disclosed that they are a consultant to Vifor/Fresenius which markets SO, sucroferric oxyhydroxide. They are also an author on the study that they draw all their hospital reduction data.

Author contributions

All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; took part in drafting the article or revising it critically for important intellectual content; gave final approval of the version to be published; and agree to be accountable for all aspects of the work.

Previous presentations

Some of the data reported in this manuscript were previously presented at a scientific congress (Ramirez de Arellano et al. ISPOR Europe 2019, Copenhagen, Denmark; Abstract # PUK15).

Acknowledgements

This study and medical writing assistance for the manuscript was funded by Vifor Pharma, Glattbrugg, Switzerland.

Additional information

Funding

This study was supported by Vifor Pharma. Medical writing support was provided by AXON Communications (United Kingdom) and Global Market Access Solutions (Switzerland) and funded by Vifor Pharma.