Long-term patiromer adherence and renin-angiotensin aldosterone system inhibitor utilization: A retrospective study

Abstract

BACKGROUND

Renin-angiotensin aldosterone system inhibitors (RAASi) reduce mortality risk in patients with heart failure, and slow progression of chronic kidney disease. However, hyperkalemia limits initiation, continuation, and reaching adequate dosing of RAASi therapy. Randomized trials have shown that patiromer users are more likely to continue taking RAASi.

AIM

To test the hypothesis of whether patiromer facilitates optimal RAASi use in the real-world.

METHODS

This study compared RAASi use in patients adherent with patiromer [proportion of days covered (PDC) > 80%] to patients with lower adherence (PDC ≤ 80%). Pre-pandemic (2014 to 2019) prescription claims data, member demographics, and plan information were used for individuals whose first patiromer prescription (index date) was during 2015-2018. PDC for patiromer was measured 6 months post-index. Patients were divided into two cohorts: Those with patiromer PDC > 80% (adherent) and those with patiromer PDC ≤ 80% (lower adherence). RAASi use was compared between cohorts during the 6 months post-index period.

RESULTS

Nearly 25% of all patiromer users were adherent, with a patiromer PDC > 80%. Adherent patiromer use was associated with significantly more angiotensin-converting enzyme inhibitor, angiotensin-receptor blocker, and mineralocorticoid receptor antagonist days supply than lower patiromer adherence (10%, 8%, and 9% higher, respectively). Also, angiotensin-converting enzyme inhibitor, angiotensin-receptor blocker, and mineralocorticoid receptor antagonist PDC were significantly greater in adherent patiromer patients than in lower adherence patiromer patients (5%, 6%, and 10% higher, respectively).

CONCLUSION

This study provides evidence that patients adherent with patiromer use more RAASi than lower-adherence patients. Prior studies show RAASi use reduces mortality risk and kidney disease progression.

Key Words: Renin-angiotensin aldosterone system inhibitors; Patiromer; Adherence; Hyperkalemia; Proportion of days covered

Core Tip: Increased use and optimal dosing of guideline-recommended renin-angiotensin aldosterone system inhibitors (RAASi) improve patient outcomes in patients with heart failure, chronic kidney disease, and diabetes. However, RAASi use is also associated with increased occurrence of high potassium levels (hyperkalemia). Patiromer is indicated for the treatment of hyperkalemia, and patients taking patiromer are better able to tolerate optimal doses of RAASi than patients not taking patiromer. In this United States study from 2014 to 2019, 3323 patients who were more adherent with patiromer proportion of days covered > 80% used more RAASi than 10124 patients who were less adherent with patiromer (proportion of days covered ≤ 80%).

INTRODUCTION

Renin-angiotensin aldosterone system inhibitors (RAASi) have been shown to reduce mortality risk in patients with heart failure (HF), and slow progression of chronic kidney disease (CKD), particularly in those with diabetes mellitus and albuminuria[1-5]. RAASi are also among the most common medications used to treat hypertension, to reduce proteinuria in non-diabetic kidney disorders, and to reduce the risk of cardiovascular events including stroke[6,7]. RAASi medications include the following subcategories: Angiotensin-converting enzyme inhibitors (ACEi), angiotensin-receptor blockers (ARB), mineralocorticoid receptor antagonists (MRA), and direct renin inhibitors (DRi)[8]. These types of RAASi medications, particularly ACEi and ARB, are widely accessible to patients and commonly prescribed by physicians. ACEi have also been available in generic formulations since the 1990s, and starting in 2010 most ARB became available as generic medications as well[9]. Generic forms of MRA and DRi are also available[10,11].

The extensive benefits from uninterrupted RAASi use are widely accepted, yet occurrence of hyperkalemia remains a major barrier in initiation or continuation of therapy. Hyperkalemia is an electrolyte imbalance, which when severe can be life-threatening[12]. Hyperkalemia is more commonly associated with a greater comorbidity burden[13]. Occurrence of hyperkalemia is relatively more common in patients with CKD, HF, or diabetes making it challenging to provide RAASi to those who need it the most[6,12,14-17]. An estimated 10%-38% of hospitalized patients receiving ACEi can experience hyperkalemia during the hospitalization, and almost 10% can experience severe hyperkalemia within one year of hospitalization[17]. One in three patients with either CKD or HF will also experience hyperkalemia, and the risk continues to increase with progressive severity of these comorbid conditions[12,18].

The risk or presence of hyperkalemia often leads to physician: (1) De-escalation/suboptimal dosing; (2) Conscious discontinuation; and/or (3) Involuntary patient non-adherence, with approximately 55%-75% of patients not reaching target dosing, and 10% stopping RAASi therapy[17,19,20]. In a recent study of over 50000 patients with hyperkalemia who were taking RAASi medications, only 17% were found to be both adherent and at an optimal dose[16]. This common practice of discontinuing or down-titrating RAASi is a major reason for lack of achieving desired outcomes as demonstrated by multiple large observational cohorts[18,21-24]. For example, a recent study of over 40000 patients with CKD and/or HF from five countries found that the risk of hospitalization after hyperkalemia was significantly higher in patients with reduced vs maintained RAASi[25]. Kidney Disease: Improving Global Outcomes guidelines now recommend that physicians should consider treatment of hyperkalemia with medications such as diuretics, bicarbonate, or potassium binders instead of discontinuing or down-titrating RAASi due to hyperkalemia[3].

Patiromer is a non-absorbable, daily-use, sodium-free exchange polymer indicated for the treatment of hyperkalemia independent of healthcare setting[26]. Patiromer was designed to avoid sodium exchange, since high sodium intake has been associated with edema, HF hospitalizations and may reduce the antiproteinuric efficacy of RAASi. Multiples clinical studies of patients with CKD and/or HF have demonstrated that patiromer can effectively lower serum potassium in hyperkalemic patients taking RAASi and maintain potassium balance with chronic daily treatment. Patiromer is a well-tolerated chronic therapy that can reduce recurrent hyperkalemia and reduce physician-initiated de-escalation/discontinuation of RAASi[27-30]. Real-word evidence has shown that patiromer users were more likely to continue taking RAASi and had fewer hospitalizations or emergency department visits after starting patiromer than patients with hyperkalemia who did not take patiromer[29]. These data also showed that patiromer is prescribed for longer duration than traditional sodium polystyrene sulfonate indicating an intention for chronic use[29].

Since the availability of patiromer, there has been a lack of real-world data on consistent use of patiromer and its ability to allow goal-directed use of RAASi in eligible patient populations. To address this knowledge gap, we examined RAASi use in patients who are adherent with patiromer [proportion of days covered (PDC) over 80%] and patients with lower patiromer adherence. Our hypothesis is that patients who are adherent with patiromer will have better RAASi use.

MATERIALS AND METHODS

Study population

The current study used pre-pandemic (2014 to 2019) prescription claims data, member demographics, and plan information from the Symphony Health Dataverse, a unified repository of data from a wide range of commercial, Medicare, and Medicaid payors in the United States. ZIP Code Tabulation Area information from the United States Census Bureau[31] was aggregated at the 3-digit ZIP code level and was merged with the patient data to provide racial, educational, income, and population density information for each patient’s 3-digit ZIP code region. Access to the data was provided by CSL Vifor.

The study population included individuals: (1) Whose first patiromer prescription was written from 2015 to 2018; and (2) Had prescription activity, via claims data, for both 6 months before and 12 months after the first prescription. The patient’s index date was the date of the first patiromer prescription. Patients under the age of 18.5 years on their index date or with a sodium polystyrene sulfonate prescription from 2014 to 2019 were excluded.

PDC is the industry-accepted measure of medication adherence by both insurance payors and value-based care organizations[32-34]. PDC for a given medication is measured over a specific time period, usually from the date the first prescription is filled through a specific end date (e.g. 6 months after the first fill date). PDC is defined as the total days supply from prescriptions filled during the time period divided by the total number of days in the time period. In the current study, PDC for patiromer was measured in the 6 months following each patient’s index date. Since 80% is the most common cutoff used in the medication adherence literature[35-38], patients in this analysis were divided into two cohorts: Those adherent with patiromer (PDC > 80%) and those with lower patiromer adherence (PDC ≤ 80%).

Outcome measures

The primary outcome measure was use of RAASi medication classes and was compared in patients with above or below a patiromer PDC threshold of 80%. We analyzed this outcome using the following metrics: (1) The percent of patients taking an ACEi, ARB, MRA, or DRi during the 6 months after the patient’s index date; (2) The average days supply of ACEi, ARB, MRA, or DRi medications over the 6 months after the index date; (3) Average ACEi, ARB, MRA, or DRi PDC measured from the first post-index RAASi prescription through 6 months after the index patiromer prescription; and (4) The percent of patients with ACEi, ARB, MRA, or DRi PDC > 80%.

Statistical analysis

Two-tailed t-tests were used to assess differences between cohorts in the RAASi subcategory average days supply and average PDC. The percent of patients taking each RAASi subcategory of medication and the percent of patients with RAASi subcategory PDC > 80% were compared between cohorts using z-tests.

RESULTS

Study population descriptive characteristics

Of the 13447 patients who met the study’s inclusion and exclusion criteria, 41% were female, and the average age was 62 ± 13 years. Five age categories are shown in Table 1, with 50% of patients being between the ages of 65 and 85. The study sample had a nationwide presence. Subjects were located in all 10 regions of the United States (defined by the patient’s ZIP code first digit), and 17% of patients lived in the South-East. Over half of the patients (56%) were in Medicare, by far the largest plan type.

Table 1 Baseline population characteristics.

	Count	Mean or percent	SD
Number of patients taking patiromer	13447
Female	5527	41%
Average age1		62	13
Age < 45	1293	10%
45 ≤ age < 55	1963	15%
55 ≤ age < 65	3388	25%
65 ≤ age < 75	3825	28%
75 ≤ age < 85	2978	22%
Pharmacy benefit manager plan	1568	12%
Health insurance exchange commercial plan	74	1%
Other commercial plan	444	3%
Cash plan	334	2%
Employer group plan	1092	8%
Medicaid plan	1045	8%
Medicare plan	7567	56%
Other plan	1323	10%
Number of unique daily-use medications in pre-index period		9.5	5.3
Avg total pre-index out-of-pocket cost per month any prescription ($)		45.6	82
Community census information from patient’s 3-digit ZIP code
Unemployment		7%	0.019
White		78%	0.221
Other race		22%	0.221
Poverty		14%	0.046
Disabled		10%	0.030
Not high school graduate		14%	0.058
High school graduate		27%	0.061
Some college		29%	0.042
Bachelors		19%	0.052
At least masters		12%	0.049
Per capita income		$30327	$7770
Population per square mile		2498	6772

¹Age based only on year of birth.

Years before 1940 were set to 1940 by Symphony Health. ZIP: Zone improvement plan.

During the 6-month pre-index period, patients averaged 9.5 ± 5.3 unique daily-use medications, and they spent $45.60 ± $82.00 out-of-pocket per month for all prescription medications. Table 1 also provides summary racial, educational, income, and population density values from the patients’ 3-digit ZIP code areas.

Outcomes

Nearly 25% of all patiromer users were adherent over the 6-month post-index period, with a patiromer PDC > 80%. Table 2 shows the RAASi outcomes comparisons between those with patiromer PDC over 80% (adherent) and those with PDC less than or equal to 80% (lower adherence). A significantly greater proportion of patients with high patiromer adherence had at least one ACEi prescription than patients with lower patiromer adherence (26.2% vs 23.7%, respectively). There were no significant differences between patiromer cohorts in the percent of patients with at least one prescription from each of the other RAASi subcategories, but there were significant differences in the amount of use within most subcategories. Greater patiromer adherence was associated with greater RAASi use. Specifically, in ACEi, ARB, and MRA, patients who were adherent with patiromer had significantly more RAASi days supply than patients with lower patiromer adherence (Figure 1). Also, ACEi, ARB, and MRA utilization (as measured by PDC) was significantly greater in patients with patiromer adherence than in patients with lower patiromer adherence (Figure 2). Finally, a significantly greater proportion of patients who were adherent with patiromer reached the 80% PDC threshold for ACEi, ARB, and MRA use (Figure 3) than patients with lower patiromer adherence.

Figure 1 Renin-angiotensin-aldosterone system inhibitors subcategory days supply in patients by patiromer adherence. RAASi: Renin-angiotensin-aldosterone system inhibitors; ACEi: Angiotensin-converting enzyme inhibitor; ARB: Angiotensin II receptor blocker; MRA: Mineralocorticoid receptor antagonist; DRi: Direct renin inhibitor; PDC: Proportion of days covered.

Figure 2 Average renin-angiotensin-aldosterone system inhibitors subcategory proportion of days covered in patients by patiromer adherence. RAASi: Renin-angiotensin-aldosterone system inhibitors; PDC: Proportion of days covered; ACEi: Angiotensin-converting enzyme inhibitor; ARB: Angiotensin II receptor blocker; MRA: Mineralocorticoid receptor antagonist; DRi: Direct renin inhibitor.

Figure 3 Percent of patients with renin-angiotensin-aldosterone system inhibitors subcategory proportion of days covered > 80% by patiromer adherence. RAASi: Renin-angiotensin-aldosterone system inhibitors; PDC: Proportion of days covered; ACEi: Angiotensin-converting enzyme inhibitor; ARB: Angiotensin II receptor blocker; MRA: Mineralocorticoid receptor antagonist; DRi: Direct renin inhibitor.

Table 2 Renin-angiotensin aldosterone system inhibitor outcomes by patiromer adherence¹.

	Adherent patiromer patients (patiromer PDC > 80%)			Lower adherence patiromer patients (patiromer PDC ≤ 80%)			Difference	P value
	Count	Mean (95%CI) or percent	SD	Count	Mean (95%CI) or percent	SD
Number of patients	3323	24.7%		10124	75.3%
Patients with ACEi	870	26.2%		2395	23.7%		2.5%	0.0032
Patients with ARB	790	23.8%		2324	23.0%		0.8%	0.3319
Patients with MRA	209	6.3%		586	5.8%		0.5%	0.2877
Patients with DRi	4	0.1%		4	0.0%		0.1%	0.0972
Average ACEi 6-month days supply		122.8 (120.0-125.6)	42.2		111.2 (109.4-113.1)	45.8	11.6	< 0.0001
Average ARB 6-month days supply		121.2 (118.2-124.3)	43.8		111.8 (110.0-113.7)	45.7	9.4	< 0.0001
Average MRA 6-month days supply		114.5 (108.1-121.0)	47.5		105.0 (101.2-108.7)	46.5	9.6	0.0122
Average DRi 6-month days supply		104.0 (48.7-159.3)	56.4		101.3 (56.3-146.2)	45.8	2.7	0.9426
Average ACEi 6-month PDC2		88.8% (87.5%-90.0%)	0.19		84.2% (83.3%-85.1%)	0.23	4.6%	< 0.0001
Average ARB 6-month PDC2		90.9% (89.8%-92.1%)	0.17		85.5% (84.6%-86.3%)	0.22	5.5%	< 0.0001
Average MRA 6-month PDC2		87.2% (84.3%-90.1%)	0.22		79.1% (77.0%-81.2%)	0.26	8.1%	< 0.0001
Average DRi 6-month PDC2		65.0% (30.2%-99.9%)	0.36		63.1% (36.3%-89.9%)	0.27	1.9%	0.9344
Percent of patients with ACEi PDC > 80%	689	79.2%		1694	70.7%		8.5%	< 0.0001
Percent of patients with ARB PDC > 80%	660	83.5%		1693	72.8%		10.7%	< 0.0001
Percent of patients with MRA PDC > 80%	165	78.9%		371	63.3%		15.6%	< 0.0001
Percent of patients with DRi PDC > 80%	2	50.0%		1	25.0%		25.0%	0.4652

¹Patiromer adherence determined by proportion of days covered threshold of 80%.

²Renin-angiotensin-aldosterone system inhibitor proportion of days measured from 1^st post-index renin-angiotensin-aldosterone system inhibitor prescription through 6 months after index patiromer or sodium polystyrene sulfonate prescription.

PDC: Proportion of days covered; CI: Confidence interval; ACEi: Angiotensin-converting enzyme inhibitor; ARB: Angiotensin II receptor blocker; MRA: Mineralocorticoid receptor antagonist; Dri: Direct renin inhibitor.

The weighted average of days supply across RAASi subcategories was 121.2 days in patients adherent with patiromer and 110.8 days in patients with lower patiromer adherence, a difference of 10.4 days (P < 0.0001). Similarly, the weighted average of RAASi PDC across RAASi subcategories was 89.5% in patients adherent with patiromer and 84.2% in patients with lower patiromer adherence (P < 0.0001).

DISCUSSION

This study is among the first to examine the impact of patiromer on RAASi use in a large national, real-world pharmaco-epidemiological dataset. It is also the first study to compare the impact of high vs low patiromer adherence on RAASi utilization. This study of 13447 patiromer users found that RAASi use was significantly greater in patients whose patiromer adherence was high (PDC > 80%) than in patients whose patiromer adherence was lower (PDC ≤ 80%). Hyperkalemia management with patiromer was associated with significantly higher RAASi days supply and significantly higher RAASi PDC. Additionally, a greater proportion of patients adherent with patiromer reached the 80% ACEi, ARB, or MRA PDC threshold than patients with lower patiromer adherence.

In addition to our analysis of real-world post-approval data in a large, nationwide population, post hoc analysis of clinical trial data from small populations has indicated improved RAASi use in patients taking patiromer. In the OPAL-HK clinical trial, Weir et al[27] found that only 6% of patiromer users required discontinuation of RAASi, compared with 56% of patients taking a placebo. In a study of patients who were age 65 or older Weir et al[27] also found that 100% of patiromer users and 48% of those not taking patiromer were able to remain on RAASi therapy. Similar results were found for patients under age 65[39]. Another study found that in patients with and without HF, those taking patiromer were significantly more likely than patients taking a placebo to remain on guideline-recommended RAASi therapy[40]. RAASi use scores in the DIAMOND trial[30] were significantly higher in users of patiromer vs placebo.

These studies show that patiromer use is associated with better adherence to RAASi guidelines in clinical trials, but the implementation of guidelines in real-world practice settings has had wide gaps, particularly due to hyperkalemia[41]. Hyperkalemia results in an inability to initiate RAASi, premature discontinuation of RAASi therapy, RAASi non-adherence, and non-goal-directed RAASi dosing[16-24]. The gap between implementation of RAASi-guideline-adherent therapy in clinical trials and its implementation in real-world practice highlights the need for larger studies of the impact of patiromer and other potassium binders on RAASi adherence in patients with hyperkalemia from non-clinical-trial settings.

Although it did not compare the impact of high potassium binder adherence vs low potassium binder adherence, a prior database study of RAASi use in 610 patients taking patiromer, sodium polystyrene sulfonate (another potassium binder), or no potassium binder found that 78% of patiromer users were able to continue RAASi therapy, while only 57% of sodium polystyrene sulfonate users and 57% of those not using potassium binders were able to continue RAASi therapy. Continuous patiromer use led to statistically significant reductions in hospital admissions and emergency department visits. There were no significant changes in these outcomes in the continuous sodium polystyrene sulfonate users, and there were significant increases in patients who did not use potassium binders[29].

The current study helps address the need for real-world analysis given its large population size (13477 patiromer patients), its nationwide setting, and by comparing RAASi use in patients with high vs low patiromer adherence. This real-world analysis helps fill the knowledge gap left by smaller studies and studies from clinical-trial settings.

Reducing hospital admissions, inpatient lengths of stay, and emergency department visits is important in a United States healthcare system that is improving patient care by shifting from a volume- and fee-for-service-based approach to a value-based care approach[33]. Value-based care is expected to improve patient health and satisfaction and reduce health care costs by incentivizing guideline-specific care via quality metrics, improving communication between care providers, fostering higher value care at lower cost, and promoting guideline-recommended medication adherence[42]. Particularly in chronic conditions, medication adherence is associated with better health outcomes and lower overall cost of care[32]. The medication adherence metric used in this study (PDC) is the industry-accepted measure of medication adherence by both insurance payors and value-based care organizations[32-34].

The impacts of patient characteristics and comorbidity, physician specialty, insurance type, social determinants of health, and other factors on patiromer adherence (PDC) have been studied previously. Kleinman et al[43] found that patients starting patiromer who were older, male, covered by Medicare or Medicaid, received the medication from a nephrologist, or were taking RAASi had higher patiromer adherence than their counterparts. Patients with higher out-of-pocket medication costs or those living in regions of high unemployment, poverty, or disability had lower patiromer adherence. Patiromer adherence was higher in regions with higher average education or income levels.

The current study has limitations. We were not able to measure other demographic or treatment characteristics (such as the use of diuretics or sodium-glucose co-transporter-2 inhibitors) or control for potential descriptive characteristic differences between patients adherent with patiromer and patients with lower patiromer adherence due to termination of the Symphony data source. Not all patients had pre-index medical claims data availability, so comorbidities were not measured. We acknowledge that although there are unique strengths to the analyses that we performed on data from a variety of patients, payors, and payor types, the claims-level deidentified data lacked patient-level clinical information. Medication use and PDC calculations were based on days supply and prescription fill date information provided in the claims data, which is the next best metric to direct observed therapy to assess adherence; hence the possibility also exists that patients did not take every unit of medication in the prescriptions that were filled. Finally, the results were United States-based and may not be generalizable to other countries.

CONCLUSION

Although further research is needed, in this study, patients adherent with patiromer (PDC > 80%) were more adherent with RAASi medications than patients with lower patiromer adherence (PDC ≤ 80%). Increased adherence to and optimal dosing of guideline-recommended RAASi have been shown in prior studies to improve patient outcomes, including reduced mortality risk and slowed disease progression in patients with HF, CKD, and diabetes.