INITIAL MONO- AND COMBINATON THERAPY AND HYPERTENSION CONTROL THE FIRST YEAR

Abstract

Initial antihypertensive therapy with single-pill combinations produced more rapid blood pressure control than initial monotherapy in clinical trials. Other studies reported better cardiovascular outcomes in patients achieving lower blood pressure during the first treatment year. We assessed the effectiveness of initial antihypertensive monotherapy, free combinations, and single-pill combinations in controlling untreated, uncontrolled hypertensives during their first treatment year. Electronic record data were obtained from 180 practice sites; 106,621 hypertensive patients seen from 01/2004–06/2009 had uncontrolled blood pressure, were untreated for ≥6 months before therapy, and had ≥1 one-year follow-up blood pressure data. Control was determined by the first follow-up visit with blood pressure <140/<90 mm Hg for patients without diabetes mellitus or chronic kidney disease and <130/<80 for patients with either or both conditions. Multivariable hazards regression ratios (HR) and 95% confidence intervals (95%CI) for time-to-control were calculated adjusting for age, sex, baseline blood pressure, body mass index, diabetes, chronic kidney disease, cardiovascular disease, initial therapy, final blood pressure medication number, and therapeutic inertia. Patients on initial single-pill combinations (N=9,194) were more likely to have Stage 2 hypertension than those on free combinations (N=18,328) or monotherapy (N=79,099 [all p<0.001]). Initial therapy with single-pill combinations (HR 1.53, 95%CI 1.47–1.58) provided better hypertension control in the first year than free combinations (HR 1.34, 95%CI 1.31–1.37) or monotherapy (reference) with benefits in black and white patients. Greater use of single-pill combinations as initial therapy may improve hypertension control and cardiovascular outcomes in the first treatment year.

Introduction

The National Health and Nutrition Examination Survey 2007–2008 data on prevalent hypertension and 2010 population estimates suggest there are roughly 70 million hypertensive patients in the U.S.^1,2 Although roughly 50% of all hypertensive patients have blood pressure (BP) controlled to <140/<90 mm Hg, approximately 35 million people have uncontrolled hypertension.

Multiple patient, provider, treatment and system variables impact BP control.³ One key to progress is to identify modifiable variables that impact hypertension control.⁴ Prospective clinical efficacy studies first showed in 2003 then subsequently confirmed that initial treatment with single-pill combinations achieved more rapid BP control during the first three to six months than initial monotherapy with add-on medications.^5–7 Second, adherence is better when medications are given as single-pill combinations than free combinations.⁸ Third, randomized, clinical studies suggest that hypertensive patients achieving lower blood pressures in the first three to twelve months of treatment experience fewer cardiovascular events during that time period.^9,10

A meta-analysis confirmed better adherence with single-pill than free combinations,⁸ whereas associations with greater persistence on therapy, larger BP reductions, and fewer adverse effects were not statistically significant. Of 15 trials, nine examined BP changes and/or control as a primary outcome, and all but one was less than one year in duration. The authors encouraged use of single-pill combinations given better adherence, potential for better BP control and fewer adverse effects.

While clinical efficacy studies suggest single-pill combinations improve hypertension control, most studies were short-term and conducted in research intensive settings. The impact of initial therapy with monotherapy, free combinations and single-pill combinations on BP control during the first treatment year in usual clinical practice is unknown. Our study addressed this gap^11,12 among previously uncontrolled, untreated patients receiving care in a community-based practice network.

Methods

This retrospective study used electronic record data of patients at 180 clinical sites in the Outpatient Quality Improvement Network.^13,14 Each clinic signed a Business Associate Agreement approved by legal counsel and the Office of Research Protection at the Medical University of South Carolina. The Agreement authorized use of de-identified data for research.

Inclusion and exclusion criteria

Patients with uncontrolled hypertension who initiated antihypertensive therapy from January 2004 through June 2008 were eligible. Patients were included if they were ≥18 year old, untreated and uncontrolled for ≥6 months before initial antihypertensive therapy, and had at least one year of follow up data. BP control was determined using the Seventh Joint National Committee Report definition.¹⁵

To facilitate generalization of study findings, exclusion criteria were limited to chronic kidney disease with estimated glomerular filtration rate <30 ml/1.7 m²/min or ICD9 403, 585, 586, active drug or alcohol abuse (ICD9 303, 303.9X, 304.XX), major psychiatric illness (ICD9 295.XX, 296.3, 297.X, 298.X), and known malignancy (ICD9 140–209).

Non-modifiable and modifiable covariables

The impact of initial antihypertensive therapy on time to control was adjusted for non-modifiable and modifiable covariables. Non-modifiable variables included initial BP, age, race, sex, diabetes mellitus (ICD9 250.XX, chronic kidney disease as defined, cardiovascular disease (ICD9 402–414, 425, 428, 431–437, 440–442, 444), depression (300.4, 311) and body mass index. Modifiable variables included therapeutic inertia, mean dose equivalents, number of antihypertensive medication pills and antihypertensive medication classes, tobacco status, number of visits in the 13 months following treatment initiation, and Veterans Affairs versus civilian care site. While the principal outcome was BP control at one year, a 31-day extension to 396 days was allowed given variations of follow-up intervals with an outpatient observational study.

Operational definitions

Time-to-control was defined by the number of days between initiation of antihypertensive therapy and the point when 50% of patients were controlled (S(T)₅₀, where S indicates survival; T denotes time in days). BP control was defined as <140/<90 for patients without diabetes mellitus or chronic kidney disease and <130/<80 for patients with either or both diagnoses.¹⁵

Dose equivalents for each antihypertensive medication were determined by dividing the dose prescribed by the maximum recommended dose according to the hypertension guidelines or by the U.S. Food and Drug Administration for medications approved after guideline publication.¹⁵ The mean equivalents dose for all medications in the final regimen taken by each patient was calculated.

Therapeutic inertia was defined as the number of visits with elevated BP in which pharmacotherapy was not changed divided by the number of visits with BP above goal.

Data Reporting and Analysis

Baseline descriptive data presented are presented as mean and standard deviation. Comparative data by group are provided as mean and 95% confidence intervals. One-way analysis of variance was used to analyze continuous variables (age, body mass index, BP) among the final therapy groups (monotherapy, free combination, single-pill combination, none). Bonferroni adjustment was applied to multiple pair-wise comparisons of continuous variables between final therapy groups. The Chi-square test was performed for categorical variables, e.g., male/female, white/black, and diabetes (yes/no) with the final therapy groups. Fisher’s combination test was used to adjust for multiple Chi-square test comparisons. The Kruskal-Wallis test (p<0.05, true/false) was used to determine if the number of BP medications and medication categories were different among final therapy groups. The median time to 50% BP control was generated from raw-data using Kaplan-Meier survival curves. A log-rank test was used to test the homogeneity of time-to-control across the three initial treatment strata for all patients and separately in Blacks and Whites. Univariable and multivariable Cox Proportional Hazard Regression analyses were performed to identify the impact of non-modifiable and modifiable factors on time to control. Independent variables that were significant in the univariable model were included in multivariable modeling. SAS Version 9.2 was used for all analyses. Two-sided p-values <0.05 were accepted as statistically significant.

Results

Patient inclusion-exclusion diagram (Figure 1)

Figure 1.

The derivation of the study sample is shown from the clinical database based on pre-defined inclusion and exclusion criteria.

From 1,414,020 adults seen at practices in the Outpatient Quality Improvement Network between January 2004 and June 2009, 106,621 were eligible for analysis based on inclusion and exclusion criteria.

Descriptive characteristics (Table 1)

Table 1.

Baseline characteristics of hypertensive patients grouped by blood pressure goal and initial therapy category.

DESCRIPTIVE VARIABLES	PATIENTS WITHOUT DIABETES MELLITUS OR CHRONIC KIDNEY DISEASE			PATIENTS WITH DIABETES MELLITUS AND/OR CHRONIC KIDNEY DISEASE
Treatment category	MONO- THERAPY	FREE COMBINATIONS	SINGLE-PILL COMBINATIONS	MONO- THERAPY	FREE COMBINATIONS	SINGLE-PILL COMBINATIONS
Patients, Number (%)	44,344 (74.4%)	9,590 (16.1%)	5,695 (9.6%)	34,755 (74.0%)	8,738 (18.6%)	3,499 (7.4%)
Age, years	57.9 (57.8, 58.0)	59.7 (59.5, 60.0) ‡	54.2 (53.9, 54.6) ‡‡	62.2 (62.0, 62.3)	63.8 (63.5, 64.1) ‡	59.6 (59.2, 60) ‡‡
VA Clinic Patients	52.9%	56.1% ‡	35.0% ‡‡	47.3%	49.8% ‡	31.6% ‡‡
Male	70.7%	72.5% ‡	60.9% ‡‡	67.6%	69.6% ‡	54.3% ‡‡
White	32.2%	32.3% n	39.4% ‡‡	29.9%	32.0% ‡	34.9% ‡†
Black	17.5%	18.1% n	26.7% ‡‡	19.5%	20.1% n	31.9% ‡‡
1Other/Missing race	50.3%	49.6% n	34.0% ‡‡	50.6%	47.9% ‡	33.2% ‡‡
Height, meters	1.75 (1.74, 1.75)	1.75 (1.74, 1.75) n	1.73 (1.72, 1.73) ‡‡	1.74 (1.74, 1.74)	1.74 (1.74, 1.74) n	1.72 (1.72, 1.73) ‡‡
Weight, kilograms	90.4 (90.2, 90.6)	91.2 (90.7, 91.6) †	92.3 (91.7, 92.9) ‡†	92.9 (92.6, 93.1)	93.4 (92.9, 93.9) n	93.9 (93.1, 94.7) *n
Body mass index, kg/m2	29.8 (29.7, 29.8)	30.0 (29.8, 30.1) *	31.0 (30.9, 31.2) ‡‡	30.9 (30.8, 31.0)	31.0 (30.9, 31.2) n	32.1 (31.9,32.4) ‡‡
# visits/year	3.96 (3.93, 4.00)	4.12 (4.03, 4.21) †	3.58 (3.51, 3.66) ‡‡	4.06 (4.02, 4.1)	4.35 (4.27, 4.44) ‡	4.02 (3.93,4.11) n‡
Systolic BP, mmHg	150.4 (150.2, 150.5)	151.3 (150.9, 151.6) ‡	155.1 (154.6, 155.6) ‡‡	144.3 (144.1, 144.4)	145.4 (145.0, 145.8) ‡	148.8 (148.1, 149.5) ‡‡
Diastolic BP, mmHg	88.4 (88.3, 88.5)	88.3 (88, 88.5) n	93.1 (92.8, 93.5) ‡‡	82.3 (82.1, 82.4)	82.1 (81.8, 82.3) n	86.7 (86.2,87.2) ‡‡
PRE-Hypertension	29.0%	29.5% n	43.2% ‡‡	20.1%	22.3% ‡	30.6% ‡‡
Stage 1 hypertension	71.0%	70.5% n	56.7% ‡‡	40.1%	38.2% †	41.2% n†
Stage 2 hypertension	29.0%	29.5% n	43.3% ‡‡	20.1%	22.2% ‡	30.5% ‡‡
≥20/10 above goal	29.0%	29.5% n	43.2% ‡‡	41.6%	41.9% n	55.6% ‡‡
Diabetes Mellitus only	0%	0%  n	0%  nn	40.4%	39.4% n	47.7% ‡‡
CKD only	0%	0%  n	0%  nn	41.3%	41.6% n	38.5% ‡‡
DM and CKD	0%	0%  n	0%  nn	18.2%	19.1% n	13.8% ‡‡
Depression	19.1%	18.1% *	14.0% ‡‡	17.2%	16.5% n	14.1% ‡‡
Tobacco	21.5%	22.5% *	18.3% ‡‡	16.7%	16.1% n	14.7% †n

Data are provided as mean and 95% confidence intervals or percent.

Abbreviations: BP=blood pressure; CKD=chronic kidney disease; DM=Diabetes Mellitus; VA = Veterans Affairs; n=not significant;

^*p<0.05;

^†p<0.01;

^‡p<0.001.

Symbol in free combination column represents comparison with monotherapy; first symbol in single-pill combination reflects the comparison with the monotherapy column, while the second symbol represents the comparison with the free combination column.

Data are provided separately for groups without and with diabetes mellitus and/or chronic kidney disease.¹⁵ The two groups were divided into three initial therapy categories: monotherapy, free combinations, and single-pill combinations. The greatest number of patients is in the monotherapy category. The single-pill combination group is 3–5 years younger, has a higher body mass index, proportion of females, patients with Stage 2 hypertension and BP ≥20/≥10 mm Hg above goal and fewer Veterans Affairs patients than other initial therapy groups.

Number of antihypertensive pills and medication categories (Table 2)

Table 2.

Classes of anti-hypertensive medications prescribed by initial treatment category at baseline.

	MONOTHERAPY	FREE COMBINATIONS	SINGLE-PILL COMBINATIONS
Patients, Number (%)	79,099 (74.2%)	18,328 (17.2%)	9,194 (8.6%)
Blood Pressure Pills, N	1.0	2.0	1.0
Blood Pressure Categories, N	1.0	2.0	2.0
ACEI, %	28.0	0.0	0.0
1β-blocker, %	21.4	0.0	0.0
Diuretic, %	20.5	0.0	0.0
Thiazide, %	16.4	0.0	0.0
Loop,%	3.6	0.0	0.0
Aldosterone Antagonist, %	0.5	0.0	0.0
Calcium Channel Blocker, %	16.8	0.0	0.0
Dihyrdropyridine, %	12.7	0.0	0.0
Non-dihydropyridine, %	4.1	0.0	0.0
Angiotensin receptor blocker, %	7.2	0.0	0.0
α1–blocker, %	4.6	0.0	0.0
Sympatholytic, %	1.1	0.0	0.0
ACEI–Diuretic, %	0.0	19.7	47.5
ARB–Diuretic	0.0	3.3	20.3
% Diuretic–Diuretic	0.0	1.2	12.5
ACEI–Dihydropyridine CCB, %	0.0	9.4	9.7
1β-blocker–Diuretic, %	0.0	12.7	7.0
ARB–Dihydropyridine CCB, %	0.0	1.8	1.5
ACEI–Non-Dihydropyridine CCB, %	0.0	2.9	1.2
CCB–Diuretic, %	0.0	10.8	0.0
1β-blocker–CCB	0.0	6.2	0.0
1β-blocker–ARB	0.0	3.1	0.0
α1-blocker–ACEI	0.0	2.4	0.0
α1-blocker–CCB	0.0	1.6	0.0
α1-blocker–β-blocker	0.0	1.4	0.0
α1-blocker–Diuretic	0.0	1.2	0.0
ARB–Non-Dihydropyridine CCB, %	0.0	0.8	0.0
ACEI–ARB, %	0.0	0.6	0.0
CCB–Sympatholytic, %	0.0	0.6	0.0
D–Sympatholytic, %	0.0	0.6	0.0

Abbreviations: ACEI=angiotensin converting enzyme inhibitor, β=beta, CCB=calcium channel blocker, d=dihydropyridine, ARB=angiotensin receptor blocker, α=alpha; percentages in italics are a subset of the main category, e.g., diuretic or CCB; ¹Includes α,β-blocker. Percentages in italics represent a subset of the parent category.

Patients initiated on monotherapy (74.2%) received one antihypertensive pill and medication category. Patients initiated on free combinations (17.2%) were given two pills in different medication categories. Patients initiated on single-pill combinations (8.6%) received one pill with two different antihypertensive medication categories. Roughly half of free combinations coincided with a marketed single-pill combination.

Final therapy category for each initial therapy group (Tables S1–S3 [http://hyper/ahajournals.org])

The majority of patients completed the year on their initial treatment category (monotherapy [57%], free combinations [91%], single-pill combinations [86.2%]). A substantial minority begun on monotherapy changed to free combinations (32%) and single-pill combinations (6.2%). Patients who ended on free combinations included both those on free combinations only and two or more free-dose medications with a single-pill combination.

Percent of patients with BP controlled as a function of time by initial therapy group (Figure 2a–c)

Figure 2.

Predicted percentage of patients remaining hypertensive over the first year after initiation of therapy (time 0) after fitting the Cox Proportional Hazards model controlled only for ‘Initial Therapy’. Data are shown for the three ‘Initial Therapy’ groups for all hypertensive patients (Panel a) and for Black (Panel b) and White (Panel c) patients separately.

are depicted for all patients combined and black and white patients separately. Among all patients, 59% initiated on free combinations and monotherapy and 68% begun on single-pill combinations were controlled to goal after 396 days (end of study).¹⁵

The percentage of hypertensives controlled at one year was significantly higher in whites than blacks, for monotherapy (BP control 65% vs. 53%, P<0.001), free combinations (67% vs. 51%, P<0.001), and single-pill combinations (73% vs. 63%, P<0.001). The disparity in hypertension control between blacks and whites was marginally lower on single-pill than free combinations (10% vs. 16%, p=0.12). While free combinations led to better BP control than monotherapy in whites, no difference was observed in blacks in unadjusted analysis (Figure 2).

Clinical factors that influence BP control (Figure 3).¹⁵

Figure 3.

Multivariable Hazard Ratios and 95% confidence intervals (Forest plots) are provided based on the probability of obtaining blood pressure control in previously untreated hypertensive patients during the initial treatment year. Abbreviations: Female/Male (male reference); BMI5, body mass index by 5 kg/m² increments; SPC/MONO, single-pill combination with monotherapy as reference: FREE/MONO, free combinations with monotherapy as reference; SBP, 10 mm Hg = baseline systolic blood pressure in 10 mm Hg increments; DBP, 5 mm Hg, baseline diastolic BP in 5 mm Hg increments; CKD, chronic kidney disease, reference group is <Stage 3 CKD; Diabetes, diabetes mellitus; CVD, cardiovascular disease; Depression, clinical diagnosis of depression; Visits/Year, number of clinical visits yearly; Therapeutic Inertia, percent of visits with uncontrolled hypertension on which antihypertensive regimen was not changed; Mean Dose Equiv, average dose equivalents for all BP medication(s) in final treatment regimen; Final # BP Pills = number of antihypertensive pills in final regimen; Final # BP Med Cats = number of different BP medication classes in final regimen.

For multivariable hazards regression modeling, the two groups with goal BP <140/<90 and <130/<80 mm Hg were combined. Diabetes mellitus and chronic kidney disease were covariables in the model. Covariables associated with better BP control in the initial treatment year included female sex, increasing number of visits, and patients with cardiovascular disease (all p<0.0001).¹⁶ Patients with higher body mass indices and baseline systolic BP, chronic kidney disease, diabetes mellitus, a larger number of prescribed unique BP medication categories, BP pills taken daily, and higher dose equivalents and therapeutic inertia were less often controlled to goal.

For all patients, the median time to 50% BP control for the single-pill combination group was 195 days [95% CI: 185, 203], and 269 [254, 285] for free-dose combinations and 280 days [273, 287] for monotherapy with overall p<0.001. For Whites and Blacks, respectively, this trend is preserved with the single-pill combination group attaining 50% control at 168 [159, 178] and 262 days [240, 280], free-dose combination in 197 [188, 210] and 391 [377, >396] days and monotherapy in 224 [217, 231] and 377 [370, 385] days. Whites attained 50% BP control more quickly than blacks in all three initial therapy groups, p<0.001. Single-pill combinations provided more rapid BP control than free-dose combinations or monotherapy in both racial groups, p<0.01.

In multivariable modeling, patients beginning on single-pill combinations (HR: 1.53, 95%CI [1.47–1.58]) were more likely to have BP controlled in the first treatment year than patients initiating treatment with free combinations (HR 1.34, 95%CI 1.31–1.37) or monotherapy (reference). Since only ~50% of patients were identified in the electronic record as black or white (mostly missing, few other race/ethnicity), multivariable models were repeated with black (white reference) as a covariable. The multivariable hazards ratios including race for single-pill combinations (HR 1.49, 95%CI 1.43–1.56) and free combinations (HR 1.32, 95%CI 1.28–1.37) were similar to results excluding race.

Percent of hypertensive patients controlled in Cox multivariable modeling (Figure S1 [http://hyper/ahajournals.org])

The figure provides data separately for four race-sex groups. White men and women were more likely to be controlled than Black men and women on initial monotherapy, free combinations or single-pill combinations (p<0.001 for all three). In both black and white men and women, single-pill combinations provided the highest percentages of controlled hypertension (p<0.001 for both). In white men and women, initial therapy with free combinations provided better control than monotherapy (p<0.01), whereas the difference between free combinations and monotherapy was not significant in black men and women.

Discussion

The principal study finding is that untreated, uncontrolled hypertensive patients attained better BP control when treatment was initiated with a single-pill antihypertensive combination than with free combinations or monotherapy (Figure 2). The monotherapy group does not appear to ‘catch up’ during the first treatment year. Our findings in hypertensive patients receiving usual care in a diverse primary care settings, confirm shorter-term observations in randomized clinical efficacy studies.^5–7 However, in “Aliskiren and the calcium channel blocker amlodipine combination as an initial treatment strategy for hypertension control trial” (ACCELERATE),⁷ BP differences between the initial monotherapy and single-pill combination groups declined when all groups received combination treatment.

Hypertensive patients begun on single-pill combinations were ~53% and on free combination ~34% more likely than those started on monotherapy to attain BP control in the first year (Figure 3). Other covariables independently associated with better BP control included females sex, higher baseline diastolic BP, cardiovascular disease, depression, and number of visits yearly. None of these other covariables was as strongly related to BP control as initial therapy. Covariables independently associated with worse BP control included higher baseline body mass index and systolic BP, chronic kidney disease and diabetes mellitus, greater therapeutic inertia, and higher mean dose equivalents, unique BP medication categories, and number of BP pills.

Many covariable associations are not surprising.^3,5,15,16 The positive link between depression and BP control is inconsistent with the impression that depression impedes control of concomitant health conditions, but this may not apply to cardiovascular risk factors.¹⁷ A higher number of BP medications, pills and dose equivalents was associated with poorer BP control, although up-titration of antihypertensive medications is required to control BP in many patients.^15,16 The explanation may be that patients requiring treatment intensification have BP values above goal and a substantial proportion remain uncontrolled even after intensified treatment.

Hypertensive patients initiated on single-pill combinations had higher untreated BP values and were more likely to have Stage 2 hypertension than those started on monotherapy, which is consistent with clinical guidelines for initial therapy in patients ≥20/≥10 mmHg from goal.¹⁵ Despite starting with higher untreated BP, patients that began treatment with single-pill combinations were more likely to achieve BP control than those started on free combinations or monotherapy BP (Figure 2).

A second study observation is that patients are most likely to end their first treatment year on the initial treatment strategy of monotherapy, free combinations, or single-pill combinations. While this may partially reflect therapeutic inertia,¹⁶ the highest control rates at one year within each initial therapy group occurred in those that remained in the initial treatment category.

In this study, one-fourth (25.8%) of patients began treatment on combination therapy. Our data suggest that beginning more patients on combination therapy, especially single-pill combinations, which comprised only 8.6% of the total, could improve hypertension control. The growing number of generic single-pill combinations may facilitate greater use as initial therapy. However, high costs including co-payments for some proprietary single-pill antihypertensive combinations may limit use.¹⁸

Our findings are consistent with the Simplified Intervention to Control Hypertension Study (STITCH). Single-pill combination use rose from roughly 10% at baseline to >80% with STITCH compared to ~16% with stepped-care. Hypertension control at 6 months was better with STITCH than stepped-care (64.7% vs 52.7%, p=0.028).¹⁹ Medication uptitration occurred more often in patients assigned to STITCH than traditional stepped-care, which is consistent with evidence that single-pill combinations reduce therapeutic inertia.²⁰

A third study finding is that initial therapy with single-pill combinations improved BP control in both black and white hypertensive patients, which may be important for several reasons. First, blacks are more likely than whites to have Stage 2 hypertension, which typically requires two or more classes of BP medications to obtain control.^15,21 Second, adherence to antihypertensive medications appears to be lower in blacks than whites.^21,22 Single-pill combinations improve adherence.^5,8,23 Third, blacks have more economic barriers to care than whites.²¹ For some patients, co-payments or cost of purchasing medication may be less for single-pill than free combinations. Greater use of single-pill combinations in black hypertensive patients with BP ≥15/10 mm Hg from goal may reduce racial differences in hypertension control and time to control.²¹

In whites, the time to 50% control was roughly one month longer for free-dose combinations and two months longer for monotherapy than for single-pill combinations. For blacks the time to 50% control was nearly four months longer for free-dose combinations and monotherapy than for single-pill combinations. Given the greater risk for hypertensive cardiovascular disease in blacks than whites,²¹ the differences in time to control may magnify racial disparities in outcomes in addition to absolute differences in hypertension control achieved.⁹

Our study confirms and extends previous reports that initial antihypertensive treatment with single-pill combinations improves BP control compared to free combinations or monotherapy.^5–7 The meta-analysis, reporting a favorable trend but non-significant difference between single-pill and free combinations, rigorously compared the same medications and doses. In our study, only ~50% of free combinations corresponded to the medication classes in single-pill combination (Table 2). Available single-pill combinations were rigorously tested to document that the combination provides better BP control than the individual components. Free-dose combinations not corresponding to marketed single-pill combinations may not satisfy this important efficacy criterion.

Limitations of this study include use of observational, community-based practice data. Patients were not randomized. BP measurements and visit frequency were not standardized. Nearly 9,200 patients were initiated on single-pill combinations but comprised only 8.6% of patients. In multivariable hazards regression, we controlled for differences that were measured, but potentially important differences, e.g., adherence, were not assessed. Prescription information was obtained from various electronic health record systems. These data do not always reflect prescriptions the patient received.^24–26 For example, the comparatively high rates of BP control for patients who were untreated at one year (Tables 3a–c) may include patients on medication(s) not captured in the electronic record.²⁵ The available electronic record data precluded an assessment of adverse effects, although previous studies documented that initial therapy with single-pill combinations is well tolerated.^5–7,27

Perspective

Uncontrolled hypertensive patients initiated on combination antihypertensive therapy, especially single-pill combinations, were more likely to obtain blood pressure control to goal,¹⁵ than patients initiated on monotherapy. However, <10% of previously untreated patients began treatment with single-pill combinations, which may reflect the fact that most physicians have not been trained in this approach. Single-pill combinations are beneficial for improving blood pressure control in black patients who are more likely to have untreated blood pressure ≥15–20/≥10 mm Hg from goal than white hypertensive patients.^15,21 The findings suggest that initiating antihypertensive therapy more often with combination therapy, and particularly single-pill combinations, could improve blood pressure control and health equity in community-based practices.

Novelty and Significance.

What Is New?

• Starting patients with high blood pressure (BP) on two medications leads to better BP control in the first treatment year than starting treatment with one medication.

• When the two medications are in one pill, BP control is even better than when the two medications are given separately.

What Is Relevant?

• Better BP control in the first treatment year can reduce complications, especially in high-risk patients.

Summary: Greater use of two medications in a single pill as initial high BP treatment improves control and may reduce complications.