Medication compliance and clinical outcomes of fixed‐dose combinations vs free combinations of an angiotensin II receptor blocker and a calcium channel blocker in hypertension treatment

Abstract

Using the National Health Insurance Research Database of Taiwan, the authors identified 1136 patients taking fixed‐dose combination and 4544 patients taking free combinations of an angiotensin II receptor blocker and a dihydropyridine calcium channel blocker from January 2009 to December 2012. At a mean follow‐up of 2.1 years, the fixed‐dose combination was associated with improved medication adherence and persistence and better survival free from major adverse cardiac events and hospitalization for heart failure compared with the free combination regimens.

1. INTRODUCTION

Hypertension has long been recognized as a global public health challenge because of its high prevalence and association with cardiovascular morbidity and mortality.1, 2 The prevalence of hypertension is estimated to reach nearly 30% of the adult population in 2025, and by then the number of hypertensive patients worldwide will have increased approximately 60% since 2000.3 Compelling evidence has accumulated on the clinical benefits of blood pressure (BP) control in preventing stroke, coronary artery disease, heart failure, and deterioration of renal function.4, 5, 6, 7 Given the increased awareness, widespread promotion of lifestyle modification, and decades of progress in pharmacologic treatment, the BP control rate has improved but remains less than optimal. In a comparison of national survey data in 2013, around 50% of hypertensive patients in the United States achieve the target of 140/90 mm Hg, compared with 34% in England and 66% in Canada.8 In Taiwan, a nationwide survey in 2002 reported a hypertension control rate of only 25% in the general population.9

Attainment of BP target is of upmost importance in hypertension treatment. Previous studies have found that the majority of hypertensive patients, approximately two thirds of the diagnosed population, require at least two different antihypertensive medications to achieve BP targets.10, 11, 12 Furthermore, international guidelines also recommend the initiation of two antihypertensive drugs as the first step of hypertension management to control BP more promptly than treatment with single agents, especially in patients with high baseline BP or at high risk for cardiovascular events.13, 14 However, the complexity of treatment regimens is considered to have a negative impact on medication adherence and persistence, which are among the fundamental reasons leading to inadequate BP control and adverse clinical outcomes.15, 16, 17 Therefore, fixed‐dose combination (FDC) regimens, combining two or three different classes of antihypertensive agents in a single pill, are designed to simplify medication regimens. Compared with free component combinations, FDC regimens have been well studied to improve medication compliance in various disease entities, including hypertension.18, 19 Nevertheless, it remains to be investigated whether the superiority of FDCs in enhancing patient compliance could translate into a reduction in cardiovascular morbidity and mortality in patients with hypertension. Thus far, there have been no prospective studies that have rigorously compared the clinical outcomes of FDCs vs free combinations in hypertension treatment.

In Taiwan, the most frequently prescribed antihypertensive medications are dihydropyridine calcium channel blockers (CCBs). The use of angiotensin II receptor blockers (ARBs) is gaining popularity in Taiwan, while the prescription of angiotensin‐converting enzyme inhibitors (ACEIs) is simultaneously decreasing. There has also been a remarkable increase in the use of FDC regimens in hypertension treatment in Taiwan.20 Therefore, this study aimed to compare the clinical outcomes of FDCs vs free combinations of ARBs and CCBs in the real‐world management of hypertension.

2. METHODS

2.1. Data source

Data used in this study were obtained from the National Health Insurance Research Database (NHIRD) of Taiwan. Taiwan's National Health Insurance program is a state‐operated insurance system. It was implemented in 1995 and currently covers approximately 99% of Taiwan's population. Derived from the National Health Insurance program, the NHIRD provides a health claims dataset containing sociodemographic data, dates of outpatient visits, dates of admissions and discharges, prescription drugs, and utilization of all medical equipment. The patient‐level data can be extracted from the database with the use of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) diagnosis and procedure codes. The personal identifiers were well encrypted before release of the database. The institutional review board of Linkou Chang Gung Memorial Hospital approved this study.

2.2. Study cohort and design

Patients who were diagnosed with hypertension (ICD‐9‐CM code: 401.x) from January 2009 to December 2012 and prescribed an ARB and a dihydropyridine CCB available in Taiwan were identified from the NHIRD. All patients should have been prescribed the study drugs for at least 6 months after the index date, unless any clinical events developed within this period. We excluded the patients with a baseline diagnosis of vascular diseases, including coronary heart disease, carotid artery disease, stroke, or peripheral vascular disease, or patients receiving revascularization procedures such as coronary artery bypass surgery, percutaneous coronary intervention, or peripheral vascular intervention. We also excluded patients younger than 18 years, patients with a concomitant prescription of an ACEI and ARB or prescription of an FDC other than the study regimens, pregnant women, and patients with a diagnosis of cancer. To avoid misclassification, we excluded patients with crossover in treatment strategies. The date of the first prescription of the study agents was defined as the index date. A period of 6 months before the index date was defined as the baseline period, which was used to identify the underlying comorbidities of the patients. All eligible patients were then grouped into the FDC and free combination groups based on the regimens they took during the study period. Propensity score matching was performed to balance the potential difference in the two study groups. The variables used in the matching process included age, sex, diabetes mellitus (ICD‐9‐CM code: 250), chronic kidney disease (ICD‐9‐CM code: 585), dyslipidemia (ICD‐9‐CM code: 272), Charlson score, and baseline medications, including antiplatelet agents, the other antihypertensive medications, oral hypoglycemic agents, statins, and overall pill burden. The overall pill burden was defined as the total number of pills prescribed during the 6‐month baseline period. The FDC group was matched at a 4:1 ratio to the free combination group.

2.3. Study end points

The primary end point was defined as major adverse cardiac events (MACEs), a composite of all‐cause mortality, myocardial infarction (ICD‐9‐CM codes: 410–410.9), stroke (ICD‐9‐CM codes: 430–437), and coronary revascularization procedures (percutaneous coronary intervention, ICD‐9‐CM codes: 36.0–36.03 and 36.05–36.09; coronary artery bypass surgery, ICD‐9‐CM codes: 36.1–36.99 and V45.81); hospitalization for unstable angina (ICD‐9‐CM code: 411.1); and resuscitation from sudden cardiac arrest (ICD‐9‐CM code: 427.5). The secondary end points included hospitalization for heart failure, new diagnosis of chronic kidney disease, and initiation of dialysis. Mortality was identified using the death certificate data of the NHIRD.

2.4. Statistical analysis

Continuous variables were expressed as mean±standard deviation and compared between the two study groups using Student t test. Categorical variables were expressed as percentage and analyzed by the chi‐square test. Propensity score matching, a widely used statistical method to establish the cause relationship or to estimate treatment effect in retrospective observational studies was used in this study to reduce the bias resulting from the confounding factors in the two study groups.21 Medication persistence was defined as the duration of time from initiation to discontinuation of the prescription of study drugs. Medication adherence, as assessed by the proportion of days covered (PDC), was measured by dividing the days the patients were covered by prescription drugs by the study period.22, 23 To analyze the clinical end points, a Cox proportional hazard model was used for time‐to‐event analysis, and survival free from the clinical end points was compared between the two groups using the Kaplan‐Meier method. The effect of medication adherence on clinical outcomes was assessed in subgroup analysis based on the PDC (high adherence: PDC ≥80%: medium adherence: PDC 50% to <80%; low adherence: PDC <50%). All analyses were conducted using SAS Statistical Software, version 9.3 (SAS Institute Inc) and R Statistical Software, version 3.0.1 (R Foundation for Statistical Computing). A P value <.05 was considered statistically significant.

3. RESULTS

From January 2009 to December 2012 a total of 1136 patients taking FDCs and 4544 patients taking free combinations of an ARB and a dihydropyridine CCB were identified from the NHIRD of Taiwan using propensity score matching. The mean follow‐up duration was 2.1 years. Table 1 lists the demographics, comorbidities, and use of medications of the study patients. All of the study patients were diagnosed with hypertension but had no established vascular disease or history of revascularization. After propensity score matching, no significant difference was noted in the baseline period between the two study groups in terms of age, sex, comorbidities, and baseline medication use. Approximately 30% of the study patients had diabetes or dyslipidemia and <5% had chronic kidney disease. The Charlson scores were 0.87 and 0.9 in the FDC and free combination groups, respectively (P=.423). Therefore, the population enrolled in this study had a relatively low risk profile compared with many previous clinical trials.

Table 1.

Baseline characteristics

	Fixed‐dose combination of ARB+CCB (n=1136)	Free combination of ARB+CCB (n=4544)	P value
Age, y	60.28±14.56	60.70±14.89	.394
Men, No. (%)	617 (54.31)	2490 (54.80)	.769
Chronic kidney disease, No. (%)	44 (3.87)	204 (4.49)	.363
Diabetes mellitus, No. (%)	336 (29.58)	1413 (31.10)	.321
Dyslipidemia, No. (%)	364 (32.04)	1451 (31.93)	.943
Charlson score	0.87±1.06	0.90±1.06	.423
Medications, No. (%)
Antiplatelet agents	226 (19.89)	981 (21.59)	.212
lockers	353 (31.07)	1484 (32.66)	.307
Diuretics	207 (18.22)	817 (17.98)	.849
Other antihypertensive agents	87 (7.66)	332 (7.31)	.685
Oral hypoglycemic agents	307 (27.02)	1330 (29.27)	.135
Statins	246 (21.65)	998 (21.96)	.822
Insulin	57 (5.02)	240 (5.28)	.721
Baseline overall pill burden	264.85±396.37	282.45±414.16	.196

Values are expressed as mean±standard deviation or number of patients (percentage).

Abbreviations: ARB, angiotensin II receptor blocker; CCB, calcium channel blocker.

Table 2 compares the medication compliance between the two study groups. During the study period, the patients taking an FDC of an ARB and a CCB were associated with better medication adherence as assessed by PDC compared with the free combination group (PDC ≥80%: 64.97% vs 56.88%; PDC 50% to <80%: 22.55% vs 24.16%; and PDC <50%: 12.48% vs 18.95% [P<.001]). Similarly, medication persistence was better in the FDC group than in the free combination group (293.79 vs 275.13 days; P<.001). The Cox proportional hazard model showed that the use of an FDC was associated with a lower risk of MACE (hazard ratio [HR]: 0.72; 95% confidence interval [CI], 0.54–0.95 [P=.022]; Table 3). Kaplan‐Meier analysis also showed that the use of an FDC was associated with better MACE‐free survival than the free combination group (log‐rank P=.021; Figure 1A). The subgroup analysis stratified by medication adherence did not reveal a significant difference in MACE‐free survival between the two study groups (Figure 1B–D). With respect to the secondary end points, the FDC group was associated with a lower risk of hospitalization for heart failure (HR, 0.71; 95% CI, 0.51–0.99 [P=.041]). No difference was noted between the two study groups in terms of the risks of new diagnosis of chronic kidney disease (HR, 0.87; 95% CI, 0.64–1.17 [P=.348]) or initiation of dialysis (HR, 0.88; 95% CI, 0.49–1.56 [P=.65]). Kaplan‐Meier analysis showed that the FDC group was associated with better survival free from hospitalization for heart failure compared with the free combination group (log‐rank P=.0395; Figure S1). The two study groups had comparable results in terms of new diagnosis of chronic kidney disease (Figure S2) and initiation of dialysis (Figure S3). Subgroup analyses of the secondary end points based on adherence status revealed no significant difference in the event‐free survival between the two treatment strategies.

Table 2.

Comparison of medication persistence and PDC covered between the fixed‐dose and free combination groups

	Fixed‐dose combination of ARB+CCB	Free combination of ARB+CCB	P value
Persistence, d	293.79±78.49	275.13±90.22	<.001
PDC, % of patients			<.001
<50%	12.48	18.95
50% to <80%	22.55	24.16
≥80%	64.97	56.88

Abbreviations: ARB, angiotensin II receptor blocker; CCB, calcium channel blocker; PDC, proportion of days covered.

Table 3.

Cox proportional hazard model for the comparison of event‐free survival between the fixed‐dose vs the free combination groups

Events	Subgroups	HR	95% CI	P value
MACE	All patients	0.72	0.54–0.95	.022a
	PDC <50%	0.52	0.25–1.07	.074
	PDC 50% to <80%	0.93	0.57–1.52	.768
	PDC ≥80%	0.79	0.53–1.17	.24
Hospitalization for heart failure	All patients	0.71	0.51–0.99	.041a
	PDC <50%	0.75	0.36–1.56	.449
	PDC 50% to <80%	0.79	0.43–1.47	.459
	PDC ≥80%	0.73	0.47–1.15	.18
New diagnosis of CKD	All patients	0.87	0.64–1.17	.348
	PDC <50%	0.7	0.32–1.53	.375
	PDC 50% to <80%	1.03	0.57–1.88	.918
	PDC ≥80%	0.92	0.62–1.35	.661
Initiation of dialysis	All patients	0.88	0.49–1.56	.65
	PDC <50%	1.08	0.42–2.76	.879
	PDC 50% to <80%	0.7	0.21–2.38	.57
	PDC ≥80%	1.26	0.5–3.14	.625

Abbreviations: CI, confidence interval; CKD, chronic kidney disease; HR, hazard ratio; MACE, major adverse cardiac events; PDC, proportion of days covered.

Figure 1.

Comparison of major adverse cardiac event (MACE)–free survival between the fixed‐dose and the free combination groups of angiotensin II receptor blockers (ARBs) and calcium channel blockers (CCBs). (A) All patients; (B) proportion of days covered (PDC) <50%; (C) PDC 50% to <80%; and (D) PDC ≥80%. The fixed‐dose combination (FDC) group of an ARB and a CCB was associated with better MACE‐free survival compared with the free combination group. There was also a trend toward better MACE‐free survival in the subgroup of PDC <50% with the use of an FDC.

4. DISCUSSION

This claims database study used the NHIRD of Taiwan to retrospectively analyze the clinical outcomes of an FDC vs free combinations of an ARB and a dihydropyridine CCB in the real‐world treatment of hypertension in patients without established vascular disease. During the study period, compared with the free combination regimens, the use of an FDC of an ARB and a CCB was associated with better medication adherence and persistence and survival free from MACE and hospitalization for heart failure.

Evidence suggests that most patients with hypertension require two or more antihypertensive drugs to achieve target BP. However, increased pill number has been shown to negatively affect medication compliance. In a meta‐analysis conducted by Bangalore and colleagues,18 patients with chronic illnesses who took FDC regimens had a 26% reduction in noncompliance rate compared with those who received free‐drug combination regimens. The subanalysis on hypertension also reported a 24% reduction in the risk of noncompliance with the use of an FDC. However, data on the clinical outcomes of an FDC in hypertension treatment remain sparse. In the Use of a Multidrug Pill in Reducing Cardiovascular Events (UMPIRE) study, an FDC of aspirin, a statin, and two antihypertensive drugs resulted in better medication adherence and a small but significant reduction in systolic BP and low‐density lipoprotein cholesterol levels compared with usual care.24 Nevertheless, the cardiovascular event rates were comparable between the two treatment strategies. On the contrary, our previous study using the NHIRD of Taiwan found that treatment with an FDC of valsartan and amlodipine was associated with higher medication compliance, lower health care costs, and better MACE‐free survival compared with the free combination regimens.25 Consistently, the use of an FDC of an ARB and a CCB in the current study was associated with an increase in medication adherence and persistence. Given the enrollment criteria that the patients should be prescribed with the study drugs for at least 6 months, the overall medication compliance in our study cohorts would be better than that in the real‐world treatment of hypertension. Nevertheless, the use of an FDC was still associated with an in medication compliance and MACE‐free survival compared with the free combination group.

Combination therapy with an ACEI or an ARB plus a CCB has been recommended by several guidelines on hypertension treatment.13, 14, 26 Clinical outcomes of the combination of a RAAS inhibitor and a CCB have been evaluated in the in the Anglo‐Scandinavian Cardiac Outcomes Trial‐Blood Pressure Lowering Arm (ASCOT‐BPLA), which revealed that an amlodipine‐based antihypertensive regimen adding perindopril as required prevented more adverse cardiovascular events compared with an atenolol‐based regimen with adding bendroflumethiazide as required.27 In addition, the Avoiding Cardiovascular Events Through Combination Therapy in Patients Living With Systolic Hypertension (ACCOMPLISH) study also reported the superiority of a benazepril‐amlodipine combination over a benazepril‐hydrochlorothiazide combination in reducing the composite of cardiovascular events and mortality.28 Although ACEIs were studied in the ASCOT‐BPLA and ACCOMPLISH studies, currently available evidence does not support significant differences in the cardiovascular outcomes with the use of an ACEI vs an ARB in hypertension treatment.29, 30, 31 In a previous analysis using the NHIRD of Taiwan, we found that the FDC of an ARB and a CCB was as effective as that of an ACEI and a CCB in reducing cardiovascular events and adverse renal outcomes.32

Thus far, no prospective randomized controlled trials have been performed to examine whether the increased use of an FDC compared with free combination regimens in hypertension would translate into a reduction in cardiovascular morbidity and mortality. Nevertheless, prior studies in various disease entities including hypertension have drawn convincing conclusions that an FDC could improve medication compliance. Numerous trials have also confirmed the association between medication compliance and BP control rates as well as reduction in adverse cardiovascular events. The current study provided real‐world outcomes of an FDC in hypertension treatment. Nevertheless, more rigorous evidence is still warranted to determine whether increased medication compliance with the use of an FDC could translate into improvement in clinical outcomes.

4.1. Study limitations

There are several limitations in this retrospective cohort study. The current study utilized the NHIRD of Taiwan, which is a claims database based on prescription and patient registration information and therefore not designed for research purpose. The NHIRD does not provide BP records. The efficacy of BP lowering, the critical link between medication compliance and patient outcomes, could not be estimated in this study. In addition, the rationale behind the choice between the FDC and free combination regimens was not recorded in the database. We could not exclude the possibility of selection bias in this study. We compared the medication adherence and persistence between the two strategies and categorized the study patients for subgroup analysis based on the PDC. Although widely used in the studies of pharmacy claims datasets, these surrogate markers of medication compliance did not ensure that the patients consumed the medications accordingly. We used the propensity score matching to balance the potential differences between the two study groups. However, there should be parameters that were not taken into consideration and may confound the study results, which is an inherent limitation of retrospective studies. Finally, we enrolled hypertensive patients without established vascular diseases in the current study. It remains to be determined whether the results could be extrapolated to patients with documented atherosclerotic cardiovascular diseases.

5. CONCLUSIONS

In this retrospective claims database study, the use of the FDC of an ARB and a dihydropyridine CCB was associated with better medication adherence and persistence compared with the free combination regimens. The FDC also seemed to be associated with better survival free from MACE and hospitalization for heart failure.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

Supporting information

 

 

 

Tung Y‐C, Huang Y‐C, Wu L‐S, Chang C‐J, Chu P‐H. Medication compliance and clinical outcomes of fixed‐dose combinations vs free combinations of an angiotensin II receptor blocker and a calcium channel blocker in hypertension treatment. J Clin Hypertens. 2017;19:983‐989. 10.1111/jch.13035