Clinical and economic outcomes associated with amlodipine/renin

Abstract

Objectives:

Since treatment regimen type can influence adherence and other outcomes, this study examined adherence, cardiovascular events, and economic outcomes in patients with hypertension treated with fixed-dose combination (FDC) amlodipine/olmesartan (AML/OM), FDC AML/benazepril (AML/BEN), and loose-dose combination AML plus angiotensin II receptor blockers (LDC AML/ARBs).

Methods:

Commercial health plan enrolees aged at least 18 years with index claim(s) for AML/OM, AML/BEN, or LDC AML/ARB were identified. Absence of study drug 6 months pre index, and continuous enrolment for at least 12 months post index were required. Descriptive analyses were executed to make comparisons between treatments, as well as multivariate models adjusting for baseline demographic and clinical characteristics, including propensity for assignment to study drug.

Results:

Descriptive results suggested mean follow-up adherence was higher in the AML/OM cohort [proportion of days covered (PDC) = 0.63] compared with the AML/BEN (PDC = 0.55; p < 0.001) and LDC AML/ARB cohorts (PDC = 0.34; p < 0.001). The proportion of individuals with an incident follow-up cardiovascular event composite was lower in the AML/OM cohort versus the AML/BEN and LDC AML/ARB cohorts (5.94% versus 7.85% and 16.89% respectively). Adjusted Cox models suggested that patients initiated on LDC AML/ARB (hazard ratio 1.35; p < 0.001), but not on AML/BEN, were at greater risk of a follow-up cardiovascular event (composite) compared with AML/OM. Adjusted generalized linear models suggested that mean follow-up per-member-per-month overall costs were higher in the AML/BEN (cost ratio = 1.169; p < 0.001; unadjusted mean cost US$780) and LDC AML/ARB cohorts (cost ratio = 1.286; p < 0.001; unadjusted mean cost US$1394) compared with the AML/OM cohort (unadjusted mean cost US$740).

Conclusion:

The results suggested that treatment with FDC AML/OM was associated with greater likelihood of adherence and lower overall costs than with FDC AML/BEN and LDC AML/ARB, and lower risk of cardiovascular event composite versus LDC AML/ARB.

Keywords

adherence cardiovascular outcomes fixed-dose combinations hypertension retrospective claims analysis

Introduction

Essential hypertension is a major public health concern affecting nearly 76.4 million people in the United States alone [Roger et al. 2012]; the prevalence of hypertension is expected to increase by an additional 27 million people worldwide by the year 2030 [Heidenreich et al. 2011]. Uncontrolled hypertension can lead to stroke, cardiac failure, end-stage renal disease, and death [Psaty et al. 2001]. Although awareness and control of hypertension has increased over the last three decades, hypertension remains inadequately managed in approximately 50% of all patients [Egan et al. 2010]. Patients with uncontrolled hypertension can incur incremental costs compared with patients with adequately controlled blood pressure (BP) [Paramore et al. 2001; Flack et al. 2002]. In 2010, costs attributable to hypertension were estimated to be more than US$76 billion, with US$54.9 billion and US$21.7 billion in direct and indirect costs respectively [Lloyd-Jones et al. 2010]. Numerous treatment options are available to lower BP within goal levels recommended in the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High BP (JNC 7) [Chobanian et al. 2003]. All antihypertensive agents in general, including angiotensin-converting enzyme inhibitors (ACEIs), angiotensin II receptor blockers (ARBs), β blockers, calcium channel blockers (CCBs), and thiazide-type diuretics reduce complications of hypertension [Neal et al. 2000]; however, the majority of patients require combination therapy to achieve targeted BP goals and reduction in cardiovascular (CV) risk [Black et al. 2001; Cushman et al. 2002]. Combining agents from diverse, yet complimentary, classes provides a mechanism to target the disease through multiple physiologic actions [Faulkner and Hilleman, 2001; Sica, 2002]. According to JNC 7 guidelines, BP in excess of 20/10 mmHg of goal should be initially treated with two drugs, separately or with fixed-dose combinations (FDCs). This recommendation is based on evidence that combination therapy increases the probability of achieving BP goals within a shorter time period compared with single-drug therapy [Chobanian et al. 2003].

A CCB and ARB combination regimen is a rational choice based on their complementary mechanisms of action. Addition of an ARB to a CCB lends itself to improved tolerability, for example, CCBs stimulate arterial vasodilation to a greater extent than venodilation, thereby causing fluid accumulation in interstitial spaces, whereas ARBs increase both arterial and venous dilation and thus function to counteract some of the CCB-induced arterial dilation, which is thought to result in a corresponding reduction in lower-extremity edema [See, 2001; Sica, 2003]. Furthermore, while both CCBs and ARBs act synergistically to reduce vascular smooth muscle contractility [Oparil and Weber, 2009; Gradman et al. 2010], FDCs are associated with improved treatment outcomes and greater adherence to therapy compared with loose-dose combinations (LDCs) [Gerbino and Shoheiber, 2007; Tejada et al. 2007]. The objective of the present study is to compare CV outcomes, adherence, switching, augmentation, healthcare resource utilization, and costs for patients with hypertension initiated on FDC amlodipine/olmesartan (AML/OM) with those of patients initiated on FDC AML/benazepril (AML/BEN) or LDC AML plus ARBs (LDC AML/ARB). Single-agent ARB medications that could be used in LDC AML/ARB therapy included candesartan, eprosartan, irbesartan, losartan, OM, telmisartan, and valsartan.

Methods

Data source

A retrospective observational study was conducted using data from medical and pharmacy claims for the period from 1 March 2007 to 31 December 2009 within a large US managed care health plan affiliated with OptumInsight, Inc. (www.optuminsight.com). Member coverage is geographically diverse, with coverage across all US census regions. Since individual identities or medical records were not disclosed, and data were accessed using methods consistent with the Health Insurance Portability and Accountability Act [United States, 1996], institutional review board approval was not required for this study.

Patient identification

The process involved in selecting the study sample is outlined in Figure 1. The study included commercial health plan enrolees aged 18 years and older, who initiated FDC AML/OM, FDC AML/BEN, or LDC AML plus ARBs between 1 September 2007 and 31 December 2008. The index date was the date of first pharmacy claim for the FDC cohorts or the date of the first claim for the second of the two medications constituting the LDC cohort. Patients required at least one pharmacy claim for FDC AML/OM or FDC AML/BEN during the identification period. LDC AML plus ARBs treatment episode was defined as a set of two pharmacy claims; a single-pill pharmacy claim for an ARB on the same date as, or followed by, a single-pill pharmacy claim for AML or vice versa. The date of the pharmacy claim for single-agent ARB must have been within 30 days of the date of the pharmacy claim for AML. Additionally, the fill for the second claim must have occurred prior to the run out of the first (i.e. the days’ supply for the two active ingredients must have ‘overlapped’).

Figure 1.

Study sample selection process. AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; FDC, fixed-dose combination; HTN, hypertension; LDC, loose-dose combination; OM, olmesartan.

Patients with medical and pharmacy benefits who were continuously enrolled in the plan for a minimum of 6 months prior to the index date (baseline period) and at least 12 months following the index date (follow-up period) were included. Patients were required to have a primary or secondary diagnosis code for hypertension during the baseline period [International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes: essential hypertension, 401.x; hypertensive heart disease, 402.xx; hypertensive chronic kidney disease, 403.xx; hypertensive heart and chronic kidney disease, 404.xx]. Patients were excluded if they had any pharmacy claims for FDC AML/OM, FDC AML/BEN, or LDC AML plus ARBs during the baseline period. Patients with any ICD-9-CM diagnosis, ICD-9-CM procedure, or Healthcare Common Procedure Coding System (HCPCS) codes for pregnancy, labor, or delivery during the study period were also excluded.

Study measures

Outcome measures

All outcomes were measured in the follow-up period, inclusive of the index date. Follow-up duration was calculated for each patient as number of days from the index date to the end of the follow-up period. Adherence to the prescribed regimen and persistence with therapy were measured using follow-up claims for the index medication inclusive of pharmacy claims on the index date. Adherence with index therapy was measured as the proportion of days covered (PDC). PDC was defined as the total number of days of therapy with the index medication during the follow-up period divided by the number of follow-up days. For the LDC AML plus ARBs cohort, days on index therapy were calculated using the methodology of Leslie and colleagues [Leslie et al. 2008], with days on index therapy equal to the days in which patients had access to both agents. PDC values were categorized as at least 0.80, which suggested good adherence, and less than 0.80, which suggested suboptimal adherence. Persistence with therapy was measured as days from the index date to therapy discontinuation date (first gap in therapy of 60 days or greater). Index therapy augmentation was assessed as a pharmacy claim for an antihypertensive medication in a class other than that of the index drug (or for an antihypertensive medication in the same class as the index drug, but with a different molecular makeup) during the follow-up period, more than 60 days prior to discontinuation of index therapy. There must have been at least one more continuous fill for the index drug after the first fill for the augmented drug. Index therapy switch was identified for patients who discontinued their index regimen. The pharmacy claim for a new antihypertensive medication other than index must have occurred within 60 days prior to the end of the index therapy through 90 days following the end of the index therapy.

The primary clinical outcome assessed was a composite measure of the first observed clinically significant CV event in the follow up, identified using diagnosis (primary or secondary) or procedure codes on medical claims. Clinically significant CV events included heart failure (ICD-9-CM diagnosis codes: 428.0, 428.1, 428.21, 428.23, 428.31, 428.33, 428.41, 428.43, 428.20, 428.30, and 428.40; or 428.20, 428.30, 428.40, and 428.9 if none of these four codes occurred in the prior 60 days); stroke (ICD-9-CM diagnosis codes, during inpatient (hospital) stay: 430, 431, 432.x, 433.x1, 434.x1, 436); myocardial infarction (MI; ICD-9-CM diagnosis code, during inpatient stay: 410.xx); acute ischemic heart disease (IHD) diagnosis (other than MI; ICD-9-CM diagnosis codes, during inpatient stay: 411.1, 411.8x); or surgery related to MI/IHD (including percutaneous coronary intervention; HCPCS codes: 33510–33523, 33533–33536, 33572, 92980–92981, 92982–92984, 92995–92996, 92973, G0290–G0291, S2205–S2209; ICD-9-CM procedure codes: 00.66, 36.01–36.02, 36.03, 36.05, 36.06, 36.07, 36.09, 36.1x).

Economic outcomes were assessed as per-member-per-month (PMPM) healthcare resource utilization and PMPM healthcare costs (all cause and hypertension attributable). Hypertension-attributable medical resource utilization and costs were defined based on claims associated with ICD-9-CM diagnosis, ICD-9-CM procedure, HCPCS, and revenue codes for hypertension; while hypertension-attributable pharmacy resource utilization and costs were defined based on pharmacy claims for antihypertensive medications. Healthcare resource utilization included ambulatory visits (office and outpatient), emergency department (ED) visits, inpatient stays, inpatient days, and pharmacy claims. Healthcare costs were computed as the combined health plan- and patient-paid costs in the follow-up period and included medical and pharmacy costs. Costs were adjusted to 2008 US$ using the annual medical care component of the consumer price index to reflect inflation between 2007 and 2008 [US Bureau of Labor Statistics, US Department of Labor, 1999]; 2009 costs were not adjusted.

Patient characteristics

Patient demographic characteristics, including age, sex, insurance type, and geographic region were captured. Baseline comorbidities were recorded using ICD-9-CM diagnosis, ICD-9-CM procedure, HCPCS, and revenue codes, and pharmacy claims, and included depression, diabetes, hypercholesterolemia or hypertriglyceridemia, heart failure, stroke or transient ischemic attack (TIA), MI, IHD (other than MI), peripheral vascular disease, left ventricular hypertrophy (cardiomegaly), cerebrovascular disease other than stroke or TIA, chronic kidney disease, and other kidney disease. Additionally, ICD-9-CM diagnosis codes were used to identify baseline comorbid conditions as defined by Quan and colleagues [Quan et al. 2005], which were used to score the Quan–Charlson comorbidity index [Charlson et al. 1987]. Baseline PMPM all-cause healthcare resource utilization and healthcare costs were also captured as described for follow-up economic outcomes.

Statistical analysis

Descriptive and multivariate approaches were executed for two sets of cohort comparisons: FDC AML/OM cohort versus FDC AML/BEN cohort, and FDC AML/OM cohort versus LDC AML plus ARBs cohort. All study variables, including baseline and outcome measures, were analyzed descriptively. χ² tests were used to examine cohort differences in categorical baseline characteristics (e.g. patient demographic and clinical characteristics) and in the distributions of follow-up clinically significant CV events. The independent samples (two-sided) t test was used to evaluate cohort differences in continuous variables (e.g. mean age, Quan–Charlson comorbidity score, baseline and follow-up economic characteristics, and length of follow up).

The associations between treatment cohort and follow-up therapy PDC at least 0.80, follow-up CV event, PMPM healthcare resource utilization, and PMPM healthcare costs were analyzed with multivariate regression models, adjusting for covariates. Follow-up therapy PDC at least 0.80 was analyzed using logistic regression; time to first follow-up CV event was analyzed using Cox proportional hazards models; follow-up healthcare resource utilization outcomes were analyzed using negative binomial regression; and follow-up healthcare costs were analyzed using generalized linear models with a γ distribution and log-link function. The γ distribution and log-link account for the skewed distribution of costs.

All analyses were adjusted for baseline demographic and clinical characteristics that included age, sex, geographic region, and number of baseline nonindex hypertension medication classes (i.e. ARBs, diuretics, ACEIs, CCBs, β blockers, ACEI/CCB, ARB/CCB, and other antihypertensives). Models were further adjusted in the following ways: event models were adjusted for baseline all-cause healthcare resource utilization and baseline comorbidities (defined above); resource utilization and PDC at least 0.80 models were adjusted for baseline all-cause healthcare resource utilization, baseline Quan–Charlson comorbidity score, and select baseline comorbidities not overlapping with the Quan comorbidity algorithms (i.e. depression, hypercholesterolemia or hypertriglyceridemia, IHD other than MI, and nonchronic kidney disease); and cost models were adjusted for log-transformed baseline all-cause medical costs and pharmacy costs, baseline Quan–Charlson comorbidity score, and select baseline comorbidities not overlapping with the Quan comorbidity algorithms.

Multivariate analyses were further adjusted using propensity score subclassification. Propensity estimates for treatment cohort (for each of the two sets of cohort comparisons) were generated using logistic regression models, adjusting for age, gender, region, baseline comorbidities, baseline all-cause healthcare resource utilization, and baseline nonindex hypertension medication classes. Cox proportional hazards model estimates were also generated for each propensity score quintile (PSQ) to provide additional granularity to the comparisons between cohorts.

Estimated hazard ratios (HRs), odds ratios (ORs), incidence rate ratios and cost ratios were computed [along with standard errors, associated two-tailed p values and 95% confidence intervals (CIs)] to test the association between independent variables and follow-up outcomes (CV events, PDC ≥0.80, healthcare resource utilization and costs respectively).

Results

Patient characteristics

A total of 24,663 patients were identified for inclusion in the study; baseline demographic and clinical characteristics by cohorts are presented in Table 1. The FDC AML/BEN cohort was most populous (48.9%; n = 12,051), followed by the LDC AML plus ARBs (31.4%; n = 7748) and FDC AML/OM (19.7%; n = 4864). Patients in the FDC AML/OM cohort were younger. Larger proportions of patients in the FDC cohorts were men and located in the South. Patients in the FDC AML/OM cohort largely had a similar comorbidity profile at baseline compared with the FDC AML/BEN cohort; however, the prevalence of comorbidities in the LDC cohort at baseline was greater (Table 1). Further, the mean baseline Quan–Charlson comorbidity score in the FDC AML/OM cohort was 0.72, which was lower than the LDC AML plus ARBs cohort (1.48; p < 0.001). Use of antihypertensive medications during the baseline period was consistently higher in the LDC AML plus ARBs cohort compared with the FDC AML/OM cohort. Select baseline PMPM all-cause healthcare resource utilization and healthcare costs differed by cohorts; the mean PMPM all-cause medical costs at baseline in the FDC AML/OM cohort was lower compared with the FDC AML/BEN and LDC AML plus ARBs cohorts (US$516, US$590, and US$1302 respectively).

Table 1.

Baseline demographic, clinical, and economic characteristics of cohort groups.

	FDC AML/OM cohort 1 (N = 4864)	FDC AML/BEN cohort 2 (N = 12,051)	p value 1 versus 2	LDC AML + ARBs cohort 3 (N = 7748)	p value 1 versus 3
Age, years, mean (SD)	53.8 (11.2)	56.0 (11.9)	<0.001	60.7 (12.5)	<0.001
Men, %	58.06	56.24	0.030	50.79	<0.001
Region, %
Northeast	9.29	4.85	<0.001	9.37	<0.001
Midwest	13.59	14.08		19.72
South	67.35	73.79		57.15
West	9.77	7.27		13.76
Comorbidities of interest, %
Depression	5.98	6.07	0.821	7.81	<0.001
Diabetes	19.51	22.06	<0.001	32.32	<0.001
Hypercholesterolemia or hypertriglyceridemia	59.70	57.13	0.002	67.59	<0.001
Heart failure	2.41	2.44	0.896	8.63	<0.001
Stroke or TIA	2.30	2.12	0.451	5.64	<0.001
MI	0.31	0.36	0.626	1.48	<0.001
IHD (other than MI)	9.87	10.21	0.499	21.68	<0.001
Peripheral vascular disease	5.24	4.39	0.017	10.80	<0.001
Left ventricular hypertrophy (cardiomegaly)	2.38	2.17	0.401	5.25	<0.001
Cerebrovascular disease other than stroke or TIA	3.70	3.14	0.067	8.20	<0.001
Chronic kidney disease	4.56	4.58	0.963	16.25	<0.001
Other kidney disease	0.95	1.10	0.388	5.39	<0.001
Quan–Charlson comorbidity score, mean (SD)	0.72 (1.14)	0.76 (1.18)	0.028	1.48 (1.73)	<0.001
Baseline hypertension medication use, %
ARBs (single agent ARB or FDC ARB + HCTZ)	21.13	4.72	<0.001	54.30	<0.001
Diuretics (single-agent diuretic or FDC potassium-sparing diuretic + HCTZ)	20.09	12.75	<0.001	31.53	<0.001
ACEIs (single-agent ACEI or FDC ACEI + HCTZ)	15.65	22.40	<0.001	20.26	<0.001
CCBs (single-agent CCB)	27.69	21.82	<0.001	40.85	<0.001
PMPM all-cause healthcare resource utilization, mean (SD)
Ambulatory visits	1.39 (1.31)	1.26 (1.28)	<0.001	1.96 (1.79)	<0.001
ED visits	0.07 (0.26)	0.07 (0.26)	0.227	0.10 (0.30)	<0.001
Inpatient stays	0.01 (0.05)	0.02 (0.06)	0.002	0.04 (0.10)	<0.001
PMPM all-cause healthcare costs, mean (SD)
Medical costs	US$516 (1349)	US$590 (2317)	0.010	US$1302 (4004)	<0.001
Pharmacy costs	US$168 (297)	US$105 (214)	<0.001	US$263 (393)	<0.001

ACEI, angiotensin-converting enzyme inhibitor; AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; CCB, calcium channel blocker; ED, emergency department; HCTZ, hydrochlorothiazide; FDC, fixed-dose combination; IHD, ischemic heart disease LDC, loose-dose combination; MI, myocardial infarction; OM, olmesartan; PMPM, per member per month; SD, standard deviation; TIA, transient ischemic attack.

Follow-up index therapy utilization and adherence patterns

The mean duration for follow up was 543 days, 585 days, and 625 days for the FDC AML/OM, LDC AML plus ARBs, and FDC AML/BEN cohorts respectively (p < 0.001; Table 2). Mean time to switch and mean time to augmentation were lower (p < 0.001 for both) in the FDC AML/OM cohort compared with the LDC AML plus ARBs cohort respectively.

Table 2.

Follow-up index therapy utilization and adherence patterns.

	FDC AML/OM cohort 1 (N = 4864)	FDC AML/BEN cohort 2 (N = 12,051)	p value 1 versus 2	LDC AML + ARBs cohort 3 (N = 7748)	p value 1 versus 3
Index therapy utilization
Length of follow up, days, mean (SD)	543.03 (113.72)	625.43 (133.14)	<0.001	585.32 (132.39)	<0.001
Time to switch, days, mean (SD)	221.14 (201.99)	214.70 (225.30)	0.259	165.20 (206.34)	<0.001
Time to augmentation, days, mean (SD)	47.63 (86.13)	47.25 (99.37)	0.873	27.24 (59.71)	<0.001
Adherence
PDC, mean (SD)	0.63 (0.34)	0.55 (0.37)	<0.001	0.34 (0.34)	<0.001
PDC ≥0.80, n (%)	2146 (44.12)	4394 (36.46)	<0.001	1513 (19.53)	<0.001
PDC <0.80, n (%)	2718 (55.88)	7657 (63.54)		6235 (80.47)	<0.001
Persistence, days, mean (SD)	339.76 (224.08)	334.71 (267.28)	0.210	233.72 (238.56)	<0.001
		OR (95% CI)	OR (95% CI)
Multivariate logistic regression results estimating likelihood of PDC (≥0.80) (Reference = FDC AML/OM cohort)		0.758* (0.635−0.906)	0.242** (0.205−0.287)

AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; CI, confidence interval; FDC, fixed-dose combination; LDC, loose-dose combination; OM, olmesartan; OR, odds ratio; PDC, proportions of days covered; SD, standard deviation.

p < 0.01; **p < 0.001.

A greater proportion of patients (44.12%) in the FDC AML/OM cohort achieved good adherence (PDC ≥0.80) to index therapy compared with those in the FDC AML/BEN (36.46%; p < 0.001) and LDC AML plus ARBs (19.53%; p < 0.001) cohorts. Multivariate logistic regression analysis results modeling PDC at least 0.80 suggested that, compared with patients in the FDC AML/OM cohort, those in the FDC AML/BEN (OR 0.76; p = 0.002) and LDC AML plus ARBs (OR −0.24; p < 0.001) cohorts were less likely to be adherent to the index regimen. Furthermore, mean persistence was lower in the LDC AML plus ARBs cohort compared with the FDC AML/OM cohort (p < 0.001); however, there was no statistically significant difference when the FDC AML/BEN cohort was compared with the FDC AML/OM cohort (p = 0.210).

Follow-up clinical outcomes

The proportion of patients who experienced an incident composite CV event in the follow up was lower in the FDC AML/OM cohort versus FDC AML/BEN and LDC AML plus ARBs cohorts (5.94% versus 7.85% and 16.89% respectively) (Table 3). Similar differences existed among the cohorts in terms of the individual components of the CV event composite (i.e. heart failure, stroke, MI, acute IHD diagnosis, and MI/IHD-related surgery). Multivariate adjusted Cox models suggested that, compared with FDC AML/OM, patients initiated on LDC AML plus ARBs were at greater risk (p < 0.001) of an incident CV event composite. There was no statistical difference in the risk of an incident CV event composite between FDC AML/OM and FDC AML/BEN (p = 0.085). Compared with subjects in the FDC AML/OM cohort, the Cox models calculated a higher adjusted risk for subjects in the LDC AML plus ARBs cohort for heart failure (p = 0.005), stroke (p = 0.007), and MI/IHD-related surgery (p = 0.042), but not for MI (p = 0.313) or acute IHD diagnosis (p = 0.065). The adjusted estimated risk of stroke was higher for patients in the FDC AML/BEN cohort (p = 0.038) compared with the AML/OM cohort.

Table 3.

Follow-up clinical outcomes.

	FDC AML/OM (N = 4864)	FDC AML/BEN (N = 12,051)	LDC AML + ARBs (N = 7748)	Multivariate Cox proportional hazards regression model results (Reference = FDC AML/OM)
				FDC AML/BEN	LDC AML/ARBs

	%	%	%	HR (95% CI)	HR (95% CI)
Clinical outcomes
Cardiovascular event composite	5.94	7.85	16.89	1.144 (0.982–1.333)	1.353*** (1.151–1.590)
Heart failure	3.99	4.76	12.33	1.143 (0.941–1.390)	1.318** (1.086–1.600)
Stroke	1.01	1.83	3.30	1.441* (1.021–2.036)	1.645** (1.148–2.357)
MI	0.72	1.05	1.88	1.185 (0.778–1.807)	1.255 (0.808–1.950)
Acute IHD diagnosis	0.72	1.10	2.17	1.180 (0.783–1.780)	1.530 (0.973–2.404)
MI/IHD-related surgery	1.25	1.75	3.07	1.067 (0.781–1.457)	1.434* (1.013–2.030)

AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; CI, confidence interval; FDC, fixed-dose combination; HR, hazard ratio; IHD, ischemic heart disease; LDC, loose-dose combination; MI, myocardial infarction; OM, olmesartan.

p < 0.05; ^†p < 0.01; ^‡p < 0.001.

Additional analyses examined the association between treatment cohort and the composite CV event within each PSQ. Table 4 presents the baseline Quan–Charlson comorbidity score by PSQ, as well as the results from the Cox proportional hazards regression model comparing risk of composite CV events between treatments in each PSQ. For the comparison between FDC AML/OM and AML/BEN, there were no differences between comparators within any of the PSQ. For the comparison between FDC AML/OM and LDC AML plus ARBs, there was a greater risk of a composite CV event in the LDC cohort in quintiles 1, 2, and 3. For this comparison, PSQ 1 represented patients with the fewest baseline comorbidities while PSQ 5 was reflective of patients who had the most comorbidities at baseline.

Table 4.

Comparison of cardiovascular event risk by propensity score quintiles.

Propensity score quintiles	Baseline Quan–Charlson comorbidity index score (n)		Risk of CV event relative to AML/OM, HR (p value)
Quintiles	FDC AML/OM (N = 4864)	LDC AML + ARBs (N = 7748)	LDC AML + ARBs
Overall	0.72	1.48	1.353 (<0.001)
Quintile 1	0.45 (2199)	0.58 (322)	1.841 (0.021)
Quintile 2	0.72 (1240)	0.66 (1283)	2.026 (<0.001)
Quintile 3	0.83 (630)	0.87 (1892)	1.582 (0.019)
Quintile 4	1.12 (430)	1.33 (2092)	1.211 (0.296)
Quintile 5	1.61 (365)	2.78 (2159)	1.155 (0.306)
	FDC AML/OM (N = 4864)	FDC AML/BEN (N = 12,051)	FDC AML/BEN
Overall	0.72	1.48	1.144 (0.085)
Quintile 1	0.77 (1997)	0.78 (1383)	1.152 (0.306)
Quintile 2	0.55 (1082)	0.56 (2267)	0.999 (0.996)
Quintile 3	0.65 (832)	0.66 (2585)	1.108 (0.598)
Quintile 4	0.79 (592)	0.77 (2788)	1.111 (0.606)
Quintile 5	0.93 (361)	0.97 (3028)	1.387 (0.133)

AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; HR, hazard ratio; LDC, loose-dose combination; OM, olmesartan.

Follow-up healthcare resource utilization and costs

Follow-up healthcare resource utilization and costs are reported in Table 5. Both descriptive and multivariate analysis results are shown for each treatment cohort. Results from the negative binominal regression models, controlling for baseline demographics and clinical characteristics, indicated that the rate of follow-up all-cause ambulatory visits was higher in the FDC AML/BEN (p < 0.001) and LDC AML plus ARBs (p < 0.001) cohorts compared with the FDC AML/OM cohort (Table 5). Similar results were observed in the case of hypertension-attributable ambulatory visits. Utilization of all-cause ED visits was higher for the FDC AML/BEN cohort versus the FDC AML/OM cohort (p < 0.006), but not different versus the LDC AML plus ARBs cohort. However, utilization of hypertension-attributable ED visits was higher in the LDC AML plus ARBs cohort (p < 0.001) and the FDC AML/BEN cohort (p < 0.002) compared with the FDC AML/OM cohort. The rates of both all-cause and hypertension-attributable inpatient stays were higher in the FDC AML/BEN (all-cause: p < 0.001; hypertension attributable: p < 0.001) and LDC AML plus ARBs (all cause: p < 0.001; hypertension attributable: p < 0.001) cohorts compared with the FDC AML/OM cohort. In comparison with the AML/OM cohort, the utilization of all-cause and hypertension-attributable inpatient days was also greater in the AML/BEN and LDC AML plus ARBs cohorts. The rates of both all-cause and hypertension-attributable pharmacy utilization were higher in the FDC AML/BEN (all cause: p < 0.001; hypertension attributable: p < 0.001) and LDC AML plus ARBs (all cause: p < 0.001; hypertension attributable p < 0.001) cohorts compared with the FDC AML/OM cohort.

Table 5.

Follow-up resource utilization and cost outcomes.

	FDC AML/OM (N = 4864)	FDC AML/BEN (N = 12,051)	LDC AML + ARBs (N = 7748)	FDC AML/BEN	LDC AML + ARBs

				Multivariate negative binomial regression results (Reference = AML/OM)
PMPM all-cause healthcare resource utilization, mean (SD)				IRR (95% CI)	IRR (95% CI)
Ambulatory visits	1.17 (1.10)	1.23 (1.16)	1.85 (1.62)	1.249** (1.211–1.289)	1.282** (1.242–1.324)
ED visits	0.06 (0.22)	0.06 (0.22)	0.08 (0.22)	1.189* (1.050–1.345)	1.099 (0.971–1.244)
Inpatient stays	0.01 (0.03)	0.01 (0.04)	0.02 (0.06)	1.354** (1.217–1.505)	1.560** (1.392–1.749)
Inpatient days	0.05 (0.26)	0.08 (0.41)	0.20 (0.86)	1.566** (1.345–1.824)	1.910** (1.612–2.264)
Pharmacy claims	2.79 (2.16)	2.91 (2.20)	4.43 (2.92)	1.329** (1.299–1.361)	1.432** (1.396–1.469)
PMPM HTN-attributable healthcare resource utilization, mean (SD)				IRR (95% CI)	IRR (95% CI)
Ambulatory visits	0.38 (0.31)	0.40 (0.30)	0.55 (0.41)	1.209** (1.177–1.242)	1.256** (1.219–1.295)
ED visits	0.00 (0.02)	0.01 (0.03)	0.01 (0.03)	1.282* (1.095–1.502)	1.439** (1.204–1.720)
Inpatient stays	0.01 (0.03)	0.01 (0.03)	0.02 (0.05)	1.329** (1.193–1.480)	1.580** (1.408–1.773)
Inpatient days	0.05 (0.24)	0.08 (0.38)	0.18 (0.78)	1.570** (1.345–1.833)	1.919** (1.612–2.285)
Pharmacy claims	1.09 (0.74)	1.12 (0.77)	1.74 (1.04)	1.293** (1.263–1.323)	1.582** (1.542–1.623)
				Multivariate generalized linear model results (Reference = AML/OM)
PMPM unadjusted all-cause healthcare costs, mean (SD)				CR (95% CI)	CR (95% CI)
Overall cost	740 (1393)	780 (1807)	1394 (3534)	1.169** (1.092–1.251)	1.286** (1.196–1.383)
Medical costs	502 (1289)	575 (1738)	1049 (3420)	1.235** (1.128–1.352)	1.412** (1.285–1.551)
Inpatient costs	142 (710)	199 (1038)	366 (1799)	1.473** (1.223–1.773)	1.599** (1.307–1.957)
Pharmacy costs	238 (328)	206 (271)	345 (451)	1.017 (0.979–1.056)	1.125** (1.081–1.171)
PMPM unadjusted HTN-attributable healthcare costs, mean (SD)				CR (95% CI)	CR (95% CI)
Overall costs	303 (763)	345 (1089)	555 (1875)	1.235** (1.128–1.352)	1.295** (1.165–1.438)
Medical costs	223 (762)	283 (1088)	476 (1873)	1.361** (1.200–1.542)	1.476** (1.284–1.697)
Inpatient costs	134 (674)	189 (1012)	351 (1761)	1.471** (1.216–1.781)	1.607** (1.308–1.973)
Pharmacy costs	80 (43)	62 (45)	79 (48)	0.837** (0.819−0.857)	0.889** (0.869−0.908)

AML, amlodipine; ARB, angiotensin II receptor blocker; BEN, benazepril; CI, confidence interval; CR, cost ratio; ED, emergency department; FDC, fixed-dose combination; HTN, hypertension; IRR, incidence rate ratio; LDC, loose-dose combination; OM, olmesartan; PMPM, per member per month.

p < 0.01; **p < 0.001.

Mean follow-up PMPM all-cause overall, medical, and inpatient costs were lower in FDC AML/OM cohort (US$740, US$502, and US$142, respectively) compared with the FDC AML/BEN (US$780, US$575, and US$199, respectively) and LDC AML plus ARBs cohorts (US$1,394, US$1,049, and US$366, respectively). A similar trend in costs was observed for hypertension-attributable costs. Adjusted results from generalized linear regression analyses suggest higher adjusted follow-up PMPM overall healthcare costs in the FDC AML/BEN (all cause: p < 0.001; hypertension attributable: p < 0.001) and LDC AML plus ARBs cohorts (all cause: p < 0.001; hypertension attributable: p < 0.001). Compared with the FDC AML/OM cohort, medical costs were higher in the FDC AML/BEN cohort (all cause: p < 0.001; hypertension attributable: p < 0.001) and LDC AML/ARB cohorts (all cause: p < 0.001; hypertension attributable: p < 0.001). Only in the case of hypertension-attributable pharmacy costs were costs lower in both the FDC AML/BEN (p < 0.001) and LDC AML plus ARBs (p < 0.001) cohorts compared with the FDC AML/OM cohort. Compared with the FDC AML/OM cohort, inpatient costs were higher in the FDC AML/BEN (all cause: p < 0.001; hypertension attributable: p < 0.001) and LDC AML plus ARBs cohorts (all cause: p < 0.001; hypertension attributable: p < 0.001).

Discussion

To the best of our knowledge, this is the first study that addressed questions with respect to the real-world outcomes associated with combination therapy in the management of hypertension by examining the association between the use of combination (FDC and LDC) AML with renin–angiotensin system inhibitor therapies and clinical outcomes, a critical gap in the medical literature. Further, the magnitude of the clinical outcome difference between these therapies was estimated in terms of healthcare resource utilization and cost. Lastly, the differential impact on clinical and economic outcomes was also evaluated for AML-based FDCs that include OM or an ACE inhibitor, specifically, BEN. The study also showed an apparent superiority of the FDC of AML/OLM versus the FDC of AML/BEN in terms of CV events. The additional benefit of the combination of AML/OLM may be related to the higher specificity of angiotensin II blockade through direct blockade of angiotensin II receptors, although additional factors such as higher tolerability cannot be excluded.

Findings from this study showed that persistence and adherence with therapy was greater with FDC than with LDC therapy. Furthermore, persistence with FDC AML/OM was greater than with LDC AML plus ARBs. It was also observed that the risk of an incident CV event and utilization of healthcare resources was lowest in the FDC AML/OM cohort; at the same time, the highest mean adherence among the comparators was observed in the FDC AML/OM cohort. These outcomes translated into lower all-cause and hypertension-attributable healthcare costs in the FDC AML/OM cohort relative to the FDC AML/BEN and LDC AML plus ARBs cohort.

The results of this study echo findings from previous studies comparing FDC and LDC regimens on adherence and persistence outcomes [Bangalore et al. 2007; Brixner et al. 2008; Hess et al. 2008; Gupta et al. 2010; Yang et al. 2010; Zeng et al. 2010]. Patients receiving FDC versus LDC regimens had 6.6% greater adherence (p < 0.001) in a nationally representative retrospective claims analysis [Brixner et al. 2008] and were more likely to be adherent (OR 2.915; p < 0.001) and less likely to discontinue therapy (HR 0.537; p < 0.001) in another analysis [Zeng et al. 2010]. Similarly, FDC therapy compared with LDC therapy was associated with a 26% lower risk of nonadherence (p < 0.0001) in one meta-analysis [Bangalore et al. 2007] and associated with 29% (95% CI 11%–50%) greater adherence and persistence in another meta-analysis [Gupta et al. 2010].

An association between improved therapy adherence and reduced risk of CV events has previously been established and past results are consistent with our current findings. In a recent retrospective cohort evaluation, increasing adherence to a once-daily antihypertensive regimen by one pill/week reduced multivariate-adjusted mortality risk by 7% (HR 0.93; 95% CI 0.90−0.096) [Bailey et al. 2010]. Similarly, in separate evaluations, high adherence or persistence with antihypertensive therapy was shown to significantly reduce the risk of CV events by 38% (HR 0.62; 95% CI 0.40−0.96) [Mazzaglia et al. 2009] while nonadherent patients had 68% more ED visits (p < 0.0001) [Butler et al. 2011].

Greater persistence with hypertensive therapy has been associated with lower healthcare resource use [Mccombs et al. 1994; Rizzo and Simons, 1997; Sokol et al. 2005] and hospitalization rates [Hess et al. 2008]. Pittman and colleagues found that regression-adjusted healthcare costs increased as adherence decreased [Pittman et al. 2010]; mean costs for patients with a medication possession ratio (MPR) of 80% or higher (US$7182) were significantly lower than patients with an MPR ranging from 60% to 79% (US$7560) or for patients with an MPR less than 60% (US$7995) (p < 0.001 for both). The odds of CV-related hospitalizations (OR 1.33; 95% CI 1.25–1.41) and ED visits (OR 1.45; 95% CI 1.33–1.58) (p < 0.001) were significantly greater among patients with low or moderate adherence [Pittman et al. 2010]. Similar findings from the literature were compiled in a meta-analysis of 12 retrospective claims database studies which showed that the mean difference in combined total annual hypertension-related and all-cause healthcare costs was US$1357 lower in the single-pill FDC group compared with the LDC group [Sherrill et al. 2011].

The findings from this study can have significant implications from the perspective of providers, payers, and patients. Often, providers lean towards LDCs given the flexibility to titrate dose for individual patients to the optimal level. This may inadvertently lead patients to not use all the prescribed agents as intended either due to regimen complexity, lack of understanding regarding the importance of taking all medications, or access to one component of the regimen, which may have higher co-pays [Bangalore and Ley, 2012]. These barriers to optimal adherence may be compounded by inadequate follow-up visits when there is maximal opportunity to influence patients. In many cases, the opportunity to achieve goal or good control with low doses of FDC therapy may have been lost: with FDC therapy, patients ‘may achieve the goal of reaching target BP sooner as a first-line approach’, with a regiment that may be cost effective as well as simple to take for the patient [Neutel, 2011: 88].

From a payer perspective, plans with formulary restrictions such as prior authorizations, step-up therapy requirements, or higher co-pays for FDC therapies may not be factoring in the potential medical cost savings that may not only offset the incremental acquisition costs of FDCs, but also may result in cost savings through the efficient use of FDC antihypertensive therapy. From a patient perspective, ‘use of true once-daily single-pill combination therapy with effective and well-tolerated agents can reduce pill burden, simplify treatment regimens and improve treatment adherence, which can, in turn, help patients to reach and maintain their blood pressure target and achieve the short- and long-term treatment goal of CV risk reduction’ [Bangalore and Ley, 2012: 345].

Administrative claims are a strong data asset that allow evidence-based examination of healthcare utilization and costs and typically provide robust sample sizes coupled with diverse patient makeups. Other studies have supported the evaluation of such claims for the same purpose [Christensen et al. 1994; Choo et al. 1999]. However, administrative claims have some limitations that should be considered in this retrospective study. Due to the observational nature of claims data, evidence of a pharmacy fill does not guarantee that a subject adhered to the regimen; however, claims data most often serve as a proxy for measuring adherence. Additionally, clinically significant CV events may result from a multitude of factors, and no causal link to treatment may be inferred from claims data. Another limitation of this study is the selection bias associated with subjects identified for inclusion in each cohort. The LDC AML plus ARBs cohort had a higher representation of older and sicker patients. Although propensity score subclassification techniques were implemented, residual confounding by selection bias cannot be ruled out. Given the over representation of patients from the ‘South’ region in the database, it is plausible that the cost data may not be uniform to the entire USA due to differential regional variations in payment and cost structures.

In summary, this study extends findings from the literature that good adherence rates on average are quite low, ranging between 20% and 45% among studied treatment groups, which further highlights the significance of nonadherence to treatment as a major public health issue. However, our data demonstrate that patients on FDCs were more adherent and persistent with therapy than patients on LDCs. Treatment with AML/OM was associated with greater adherence and persistence than treatment with AML/BEN and LDC AML plus ARBs. Greater adherence, lower healthcare resource utilization costs, and lower incidence of CV events over the follow up were observed concurrently. These data suggest that large improvements in patient outcomes and system-wide cost savings may be realized from relatively modest improvements in adherence. The economic impact of chronic care diseases on the US healthcare system continues to grow unabated. Among the various approaches that are needed to reduce healthcare costs, improved adherence to therapy is one factor that is known to decrease the burden of CV events. Additional large-scale studies on the use of FDCs in terms of clinical outcomes may be one effective approach to address these issues.

Footnotes

Acknowledgements

Medical writing support was provided by Nadine Aawar, PhD.

Funding

This research project was supported by funds from Daiichi Sankyo USA.

Conflict of interest statement

CM Ferrario has received grants and honorarium from Daiichi Sankyo, Inc. Sumeet Panjabi is a former employee of Daiichi Sankyo, Inc.

References

Bailey

Wan

Tang

Ghani

Cushman

(2010) Antihypertensive medication adherence, ambulatory visits, and risk of stroke and death. J Gen Intern Med 25: 495–503.

Bangalore

Kamalakkannan

Parkar

Messerli

(2007) Fixed-dose combinations improve medication compliance: a meta-analysis. Am J Med 120: 713–719.

Bangalore

Ley

(2012) Improving treatment adherence to antihypertensive therapy: the role of single-pill combinations. Expert Opin Pharmacother 13: 345–355.

Black

Elliott

Neaton

Grandits

Grambsch

Grimm

Jr . (2001) Baseline characteristics and early blood pressure control in the CONVINCE trial. Hypertension 37: 12–18.

Brixner

Jackson

II Sheng

Nelson

Keskinaslan

(2008) Assessment of adherence, persistence, and costs among valsartan and hydrochlorothiazide retrospective cohorts in free-and fixed-dose combinations. Curr Med Res Opin 24: 2597–2607.

Butler

Davis

Johnson

Gardner

(2011) Effects of nonadherence with prescription drugs among older adults. Am J Manag Care 17: 153–160.

Charlson

Pompei

Ales

Mackenzie

(1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40: 373–383.

Chobanian

Bakris

Black

Cushman

Green

Izzo

Jr . (2003) The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 Report. J Am Med Assoc 289: 2560–2572.

Choo

Rand

Inui

Lee

Cain

Cordeiro-Breault

. (1999) Validation of patient reports, automated pharmacy records, and pill counts with electronic monitoring of adherence to antihypertensive therapy. Med Care 37: 846–857.

10.

Christensen

Williams

Goldberg

Martin

Engelberg

Logerfo

(1994) Comparison of prescription and medical records in reflecting patient antihypertensive drug therapy. Ann Pharmacother 28: 99–104.

11.

Cushman

Ford

Cutler

Margolis

Davis

Grimm

. (2002) Success and predictors of blood pressure control in diverse North American settings: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). J Clin Hypertens (Greenwich) 4: 393–404.

12.

Egan

Zhao

Axon

(2010) US trends in prevalence, awareness, treatment, and control of hypertension, 1988–2008. J Am Med Assoc 303: 2043–2050.

13.

Faulkner

Hilleman

(2001) Amlodipine/benazepril: fixed dose combination therapy for hypertension. Expert Opin Pharmacother 2: 165–178.

14.

Flack

Casciano

Doyle

Arikian

Tang

. (2002) Cardiovascular disease costs associated with uncontrolled hypertension. Manag Care Interface 15: 28–36.

15.

Gerbino

Shoheiber

(2007) Adherence patterns among patients treated with fixed-dose combination versus separate antihypertensive agents. Am J Health Syst Pharm 64: 1279–1283.

16.

Gradman

Basile

Carter

Bakris

Materson

Black

. (2010) Combination therapy in hypertension. J Am Soc Hypertens 4: 90–98.

17.

Gupta

Arshad

Poulter

(2010) Compliance, safety, and effectiveness of fixed-dose combinations of antihypertensive agents: a meta-analysis. Hypertension 55: 399–407.

18.

Heidenreich

Trogdon

Khavjou

Butler

Dracup

Ezekowitz

. (2011) Forecasting the future of cardiovascular disease in the United States: a policy statement from the American Heart Association. Circulation 123: 933–944.

19.

Hess

Hill

Lau

Dastani

Chaudhari

(2008) Medication utilization patterns and hypertension-related expenditures among patients who were switched from fixed-dose to free-combination antihypertensive Therapy. P T 33: 652–666.

20.

Leslie

Gwadry-Sridhar

Thiebaud

Patel

(2008) Calculating medication compliance, adherence, and persistence in administrative pharmacy claims databases. Pharm Program 1: 13–19.

21.

Lloyd-Jones

Adams

Brown

Carnethon

Dai

De Simone

. (2010) Heart disease and stroke statistics – 2010 update: a report from the American Heart Association. Circulation 121: e46–e215.

22.

Mazzaglia

Ambrosioni

Alacqua

Filippi

Sessa

Immordino

. (2009) Adherence to antihypertensive medications and cardiovascular morbidity among newly diagnosed hypertensive patients. Circulation 120: 1598–1605.

23.

Mccombs

Nichol

Newman

Sclar

(1994) The costs of interrupting antihypertensive drug therapy in a Medicaid population. Med Care 32: 214–226.

24.

Neal

Macmahon

Chapman

Blood Pressure Lowering Treatment Trialists’ Collaboration. (2000) Effects of ACE inhibitors, calcium antagonists, and other blood-pressure-lowering drugs: results of prospectively designed overviews of randomised trials. Blood Pressure Lowering Treatment Trialists’ Collaboration. Lancet 356: 1955–1964.

25.

Neutel

(2011) Long-term blood pressure control: what can we do? Postgrad Med 123: 88–93.

26.

Oparil

Weber

(2009) Angiotensin receptor blocker and dihydropyridine calcium channel blocker combinations: an emerging strategy in hypertension therapy. Postgrad Med 121: 25–39.

27.

Paramore

Halpern

Lapuerta

Hurley

Frost

Fairchild

. (2001) Impact of poorly controlled hypertension on healthcare resource utilization and cost. Am J Manag Care 7: 389–398.

28.

Pittman

Tao

Chen

Stettin

(2010) Antihypertensive medication adherence and subsequent healthcare utilization and costs. Am J Manag Care 16: 568–576.

29.

Psaty

Furberg

Kuller

Cushman

Savage

Levine

. (2001) Association between blood pressure level and the risk of myocardial infarction, stroke, and total mortality: the Cardiovascular Health Study. Arch Intern Med 161: 1183–1192.

30.

Quan

Sundararajan

Halfon

Fong

Burnand

Luthi

. (2005) Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 43: 1130–1139.

31.

Rizzo

Simons

(1997) Variations in compliance among hypertensive patients by drug class: implications for health care costs. Clin Ther 19: 1446–1457; discussion 1424–1445.

32.

Roger

Lloyd-Jones

Benjamin

Berry

Borden

. (2012) Executive summary: Heart disease and stroke statistics–2012 update: A report from the American Heart Association. Circulation 125: 188–197.

33.

See

(2001) Angiotensin II receptor blockers for the treatment of hypertension. Expert Opin Pharmacother 2: 1795–1804.

34.

Sherrill

Halpern

Khan

Zhang

Panjabi

(2011) Single-pill vs free-equivalent combination therapies for hypertension: a meta-analysis of health care costs and adherence. J Clin Hypertens (Greenwich) 13: 898–909.

35.

Sica

(2002) Rationale for fixed-dose combinations in the treatment of hypertension: the cycle repeats. Drugs 62: 443–462.

36.

Sica

(2003) Calcium channel blocker-related periperal edema: can it be resolved? J Clin Hypertens (Greenwich) 5: 291–294, 297.

37.

Sokol

Mcguigan

Verbrugge

Epstein

(2005) Impact of medication adherence on hospitalization risk and healthcare cost. Med Care 43: 521–530.

38.

Tejada

Fornoni

Lenz

Materson

(2007) Combination therapy with renin-angiotensin system blockers: will amlodipine replace hydrochlorothiazide? Curr Hypertens Rep 9: 284–290.

39.

United States (1996) Health Insurance Portability and Accountability Act of 1996. Public Law 104–191. US Statut Large 110: 1936–2103.

40.

US Bureau of Labor Statistics, US Department of Labor (1999) Chained Consumer Price Index 1999. Available at: http://www.bls.gov/data/ (accessed 27 November 2012).

41.

Yang

Chang

Kahler

Fellers

Orloff

. (2010) Evaluation of compliance and health care utilization in patients treated with single pill vs. free combination antihypertensives. Curr Med Res Opin 26: 2065–2076.

42.

Zeng

Patel

Andrews

Frech-Tamas

Rudolph

(2010) Adherence and persistence of single-pill ARB/CCB combination therapy compared to multiple-pill ARB/CCB regimens. Curr Med Res Opin 26: 2877–2887.

Clinical and economic outcomes associated with amlodipine/renin–angiotensin system blocker combinations

Abstract

Objectives:

Methods:

Results:

Conclusion:

Keywords

Introduction

Methods

Data source

Patient identification

Study measures

Outcome measures

Patient characteristics

Statistical analysis

Results

Patient characteristics

Follow-up index therapy utilization and adherence patterns

Follow-up clinical outcomes

Follow-up healthcare resource utilization and costs

Discussion

Footnotes

Acknowledgements

Funding

Conflict of interest statement

References