Use of Disease-Modifying Therapies in Patients Hospitalized with Heart Failure: A Population-Based Perspective

Data Collection 

Information was collected about demographics, medical history, clinical characteristics, and laboratory test results in hospital survivors of HF through the review of information contained in hospital medical records. This included information about patients’ age, sex, race, prior comorbidities (eg, angina, diabetes, hypertension, stroke), body mass index, presenting symptoms, physical examination findings, and clinical characteristics. We reviewed the physician’s progress notes and daily medication logs for the prescription of selected medications at the time of hospital discharge. We examined the use of cardiac medications that have been shown to be of benefit in improving the long-term prognosis of patients with HF: angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and beta-blockers. Given the low use of ARBs in the study sample (5.2%) and indications for the use of this class of agents in patients with contraindications to ACE inhibitors, we combined the use of ACE inhibitors and ARBs into a single drug category. We also examined the use of medications shown to be effective in improving the symptomatic status associated with the onset of acute HF (digoxin and diuretics). Last, we examined the hospital medical record for evidence that physicians counseled patients about different nonpharmacologic strategies for the management of HF, including reduced intake of dietary fat, limitation of fluids, increased physical activity, dietary salt restriction, and monitoring of daily weight.

Data Analysis 

Differences in the characteristics of patients treated with both classes of effective medications (angiotensin pathway inhibitors and beta-blockers) compared with those receiving neither or 1 of these cardiac therapies were examined through the use of chi-square tests or t tests and analysis of variance for discrete and continuous variables, respectively. Factors associated with the receipt of both classes of therapy were examined through the use of logistic regression modeling. The factors chosen for inclusion in the regression analyses were based on findings from the published literature and their association with the receipt of both medications in univariate analyses with a P value less than .15.

Use of Cardiac Medications 

The most commonly used class of medications was diuretics, which were prescribed at the time of hospital discharge to essentially all study patients (98.0%). The next most frequently prescribed therapies were ACE inhibitors (53.3%), beta-blockers (52.7%), and digoxin (47.5%). In addition to these primary therapies, we also examined the prescribing of other effective cardiac medications. Antiplatelet therapy was prescribed to 60.2% of patients with acute HF, and lipid-lowering agents were prescribed to 21.8%.

Approximately 1 in 5 patients were not discharged with ACE inhibitors/ARBs or beta-blockers, approximately one half of patients were treated with 1 of these agents, and approximately one third of patients were treated with both (Figure 1). Approximately one half of study patients were discharged with both digoxin or diuretics. In pooling the use of these different cardiac drugs, 14.6% of patients were discharged with only 1 medication, 31.7% were discharged with any 2 medications, 37.1% were discharged with any 3 therapies, and 16.7% were prescribed all 4 cardiac medications.

View Large Image Download to PowerPoint

Figure 1. Prescribing of cardiac medications at the time of hospital discharge in patients with newly diagnosed heart failure (HF): Metropolitan Worcester, Massachusetts, 2000.

Hospital Initiation of New Cardiac Therapies 

We also examined the initiation of each of the medications under study at the time of hospital discharge in patients who were not receiving these therapies before the time of hospital admission based on the review of information contained in hospital medical records. Trends similar to those previously noted with regard to the use of individual and combination medical therapies were observed, albeit at considerably lower absolute use rates. In the study sample, ACE inhibitors or ARBs were newly initiated in 28.5% of patients, beta-blockers in 20.5%, digoxin in 22.3%, and diuretics in 94.0%.

Characteristics Associated with Treatment Practices 

In examining differences in the characteristics of patients receiving the various effective cardiac therapies examined, patients receiving ACE inhibitors/ARBs and beta-blockers were significantly younger than respective comparison groups (Table 1). Patients receiving both of these effective cardiac medications were more likely to have a history of coronary heart disease, diabetes, HF, hyperlipidemia, and hypertension in comparison with patients who were treated with none or fewer of these cardiac medications. On the other hand, these patients were significantly less likely to have had a history of anemia or chronic lung disease. Patients discharged with both effective cardiac medications were less likely to present with cough, fatigue, generalized weakness, and nausea, but were more likely to have chest pain/discomfort, dyspnea, or orthopnea than patients receiving neither or only one of these cardiac therapies. Patients treated with both classes of effective cardiac therapies were significantly more likely to be treated with digoxin, but not diuretics, than patients not treated with any of these agents. With regard to the distribution of ejection fraction results during hospitalization in the approximately one third of patients in whom this information was available, patients treated with both effective cardiac therapies were significantly more likely to be characterized by lower ejection findings than patients who did not receive any of these therapies.

Table 1. Use of Combination Medical Therapies⁎ in Patients with Acute Heart Failure According to Selected Characteristics: Metropolitan Worcester, Massachusetts 2000

		None	ACE Inhibitors/ARBs	Beta-blockers	Both Medications	P Values
	Age (y) (%)	(n=544)	(n=621)	(n=507)	(n=791)
	<55	6.8	9.8	3.6	7.1
	55-64	5.9	10.5	7.9	10.8
	65-74	18.9	19.7	19.4	22.6	<.001
	75-84	39.9	32.7	38.3	40.3
	≥85	28.5	27.4	30.8	19.2
	Age (mean, y)	77.3	75.2	77.8	74.8	<.001
	Male (%)	41.2	46.4	43.0	43.0	.35
	White (%)	95.0	89.5	94.7	91.3	.06
	Body mass index (%)
	25-29.9	27.0	25.1	31.0	27.5	.96
	≥30	29.8	31.6	26.7	30.9
	Serum creatinine (mean, mg/dL)	1.6	1.5	1.8	1.6	.36
	Medical history (%)
	Anemia	26.1	23.8	28.4	20.7	<.05
	Angina	10.3	13.9	20.9	23.6	<.001
	Chronic lung disease	45.4	40.4	30.4	24.5	<.001
	Coronary heart disease	44.7	53.8	60.0	66.6	<.001
	Diabetes	20.8	27.7	23.3	32.5	<.001
	Heart failure	70.2	79.2	73.0	77.9	<.005
	Hyperlipidemia	16.9	19.5	30.8	34.3	<.001
	Hypertension	52.9	67.2	71.4	76.0	<.001
	Myocardial infarction	33.6	39.7	48.9	56.0	<.001
	Renal failure/disease	22.6	20.3	35.5	22.0	<.05
	Stroke	13.1	15.0	17.4	14.2	.23
	Symptoms (%)
	Chest pain/discomfort	21.7	28.3	34.7	37.2	<.001
	Cough	56.8	54.6	48.3	46.0	<.005
	Dyspnea	90.4	94.2	92.9	94.4	<.01
	Edema	72.8	72.3	66.5	69.2	.31
	Fatigue	10.1	7.3	8.3	6.3	<.05
	Generalized weakness	35.3	29.0	26.0	23.1	<.001
	Nausea	11.4	15.0	15.8	17.8	<.005
	Orthopnea	30.0	40.7	35.1	41.1	<.001
	Weight gain	8.8	9.5	8.7	9.7	.84
	Other medications
	Digoxin (%)	38.1	56.8	38.3	52.5	<.001
	Diuretics (%)	97.6	99.0	97.0	98.0	.78
	Ejection fraction findings (%)
	<40	22.9	45.8	20.5	47.7
	40-49	12.7	13.7	15.1	15.9	<.001
	≥50	64.4	40.5	64.4	36.4

ACE=angiotensin-converting enzyme; ARB=angiotensin receptor blocker.

⁎ Receipt of ACE inhibitors/ARBs and beta-blockers.

Important associations were observed between the prescribing of few or multiple cardiac therapies that have been shown to improve the long-term prognosis of patients with HF and recommendations to alter various lifestyle or dietary practices (Table 2). Patients who were treated with both ACE inhibitors/ARBs and beta-blockers were significantly more likely to be recommended to monitor or modify 3 or more lifestyle factors (46.5%) compared with patients who were prescribed none (32.9%) of these effective cardiac medications (P < .001). Relatively similar trends were observed in patients with a first episode of HF.

Table 2. Prescribing of Different Cardiac Therapies in Relation to Health Care Provider Counseling on Selected Nonpharmacologic Management Approaches

		Receipt of Selected Medications
	Nonpharmacologic intervention†	None	ACE inhibitors/ARBs	Beta-blockers	Both medications⁎
	0, 1	35.5%	20.9%	23.5%	13.0%
	2	31.8%	41.9%	39.8%	40.7%
	3	20.4%	22.2%	23.1%	24.8%
	>4	12.3%	15.0%	13.6%	21.5%

ACE=angiotensin-converting enzyme; ARB=angiotensin receptor blocker.

⁎ Receipt of ACE inhibitors/ARBs and beta-blockers. † Recommendations to monitor or modify diet, sodium intake, physical activity, limitation of fluids, monitoring of daily weight.

There were relatively similar recommendations for the monitoring or modification of various lifestyle and nonpharmacologic regimens in relation to ejection fraction findings. Among the 329 patients with an ejection fraction of less than 40%, 84.8% received counseling on 2 or more lifestyle approaches and 21.6% received counseling on 4 or more of the lifestyle approaches examined. Among the 423 patients with an ejection fraction of 50% or higher, 82.3% were recommended to monitor or modify at least 2 lifestyle factors, whereas 18.7% were recommended to monitor or modify 4 or more of the lifestyle approaches examined.

Factors Associated with the Receipt of Selected Therapeutic Approaches 

We carried out several logistic regression analyses to more systematically examine factors associated with the receipt of both effective cardiac therapies compared with none or few disease-modifying medications in patients with new-onset HF (Table 3).

Table 3. Factors Associated with Receipt of Multiple⁎ Cardiac Medications: Metropolitan, Worcester, Massachusetts, 2000

		ACE Inhibitors/ARBs and Beta-blockers
	Characteristic	OR	95% CI
	Age (y)
	≥85	0.69	0.46-1.02
	Medical history
	Anemia	0.75	0.60-0.94
	Coronary heart disease	1.80	1.45-2.24
	Chronic lung disease	0.46	0.38-0.57
	Diabetes	1.35	1.10-1.66
	Heart failure	1.25	1.00-1.57
	Hyperlipidemia	1.37	1.12-1.68
	Hypertension	1.65	1.34-2.03
	Renal failure/disease	0.67	0.54-0.84
	Stroke	0.75	0.58-0.98
	Symptoms
	Cough	0.78	0.65-0.94
	Weakness	0.76	0.61-0.95

OR=odds ratio; CI=confidence interval; ACE=angiotensin-converting enzyme; ARB=angiotensin receptor blocker.

⁎ Receipt of ACE inhibitors/ARBs and beta-blockers.

Younger patients and individuals with a history of coronary heart disease, diabetes, hyperlipidemia, and hypertension were significantly more likely to be prescribed ACE inhibitors/ARBs and beta-blockers at discharge than respective comparison groups. On the other hand, patients with anemia, chronic lung disease, renal disease, and stroke were significantly less likely to be prescribed both effective cardiac medications than patients without these conditions. Patients who presented with cough and generalized weakness were significantly less likely to be prescribed both effective cardiac medications than patients who did not present with these symptoms.

When we added the documentation of health care provider advice about nonpharmacologic regimens to our regression analyses, patients who were treated with both of these effective cardiac medications were significantly more likely to be recommended to monitor or modify 2 or more lifestyle or dietary practices (adjusted odds ratio [OR] = 1.94; 95% confidence interval [CI], 1.51-2.49) in comparison with patients who were treated with fewer effective cardiac medications.

Because patients with a first episode of HF might be more aggressively treated with combination medical therapy than patients with prior episodes of HF, we also examined the relation among various demographic, medical history, and clinical characteristics with the receipt of multiple versus single disease-modifying cardiac medications in patients with an initial diagnosis of HF. The results of this analysis showed that relatively similar factors were associated with the prescribing of multiple effective cardiac medications in these patients.

We carried out an additional regression analysis in the sample of patients with data on ejection fraction results. After we controlled for several previously described demographic and clinical variables that differed between patients with varying ejection fraction findings, patients with an ejection fraction between 40% and 49% (adjusted OR=0.87; 95% CI, 0.56-1.36), and those with an ejection fraction of 50% or greater (adjusted OR=0.47; 95% CI, 0.33-0.65), were less likely to be treated with both effective cardiac medications in comparison with patients with an ejection fraction less than 40%.

Medication Use in Patients with Heart Failure 

The results of our study suggest that only one third of patients with acute HF were treated with both classes of cardiac medications of proven efficacy in improving the long-term prognosis associated with HF. These prescribing practices were observed whether it was the patient’s first or prior episode of HF. Moreover, the initiation of these therapies in patients not previously taking these medications was shown to be considerably less than desirable.

Previous studies have primarily examined the use of individual medications in patients with HF, with little data available about the use of combination medical therapy. In a study of 387 patients hospitalized for HF at a single academic medical center in the early 1990s, approximately half of the study patients were discharged with an ACE inhibitor19; a distinct minority (18%) of patients, however, were discharged with an optimal dose of this medication. In a previous study, these investigators examined changing prescribing practices for several medications used in the management of patients with HF between 1986 and 1987, and 1992 and 1993.20 Increases in the use of ACE inhibitors, aspirin, beta-blockers, and calcium channel antagonists, and marked declines in the use of digoxin, were observed in this single-center study. The combination use of ACE inhibitors, digitalis, and diuretics did not change over time.

In comparing the rates of ACE inhibitor use in patients hospitalized at 10 community hospitals in 1995 with those hospitalized in 1992, the proportion of hospital survivors prescribed an ACE inhibitor at the time of discharge increased from 51% to 64%.21 Among more than 1500 patients treated for HF at nine hospitals throughout Connecticut in 1992, the majority of patients without contraindications to an ACE inhibitor were prescribed this therapy, although a minority of these patients (∼1 in 7) received doses consistent with available national guidelines.11 In an analysis of 297 patients with HF who were admitted to a single midwest academic medical center between 1990 and 1995, the use of ACE inhibitors and beta-blockers increased significantly over this period, including the use of more optimal doses in patients treated with an ACE inhibitor.22

The results of our study suggest that patients with HF were less than optimally treated with medications shown to be effective in the management of HF. Moreover, considerable gaps remain in the timely initiation of new cardiac therapies in patients who were not previously receiving these agents.

Factors Associated with Receipt of Minimal Treatment Modalities 

Younger patients, patients with preexisting cardiac risk factors or cardiac disease, and patients experiencing more typical cardiac symptoms were more likely to be treated with multiple disease-modifying medications of proven efficacy than respective comparison groups. Patients with a history of noncardiac conditions, such as renal failure, anemia, or chronic lung disease, were less likely to be treated with both effective cardiac therapies.

In the limited studies that have examined patient characteristics associated with the use of different HF therapies, the underuse of ACE inhibitors was previously related to older patient age, impaired renal function, normal left ventricular systolic function, and the use of alternative medical therapies.21, 22 Although avoidance of beta-blockers or ACE inhibitors in several of these patient groups may be warranted, these patients often have the most to gain from aggressive therapy given their increased rates of subsequent morbidity and mortality. Previous findings from the Worcester Heart Attack Study suggest that in many of these patients beta-blockers and ACE inhibitors are not truly contraindicated.23 Our findings reaffirm and extend the observation that internists and cardiologists involved in the management of patients with acute HF, particularly those involved in managing patients who are at high risk for adverse outcomes after HF, need to be aware of the underuse of effective disease-modifying therapies in these patients and prescribe accordingly.

Of considerable concern, there was a strong association between the use of less than optimal pharmacologic treatment regimens with the prescribing of few effective nonpharmacologic approaches. Although the reasons for these observations are unknown, these findings suggest a possible “all or nothing” approach to treatment. As has been demonstrated in other disease states, particularly in patients with acute myocardial infarction, patients more likely to receive 1 effective therapy are also those most likely to receive other effective therapeutic interventions. For example, a number of studies have suggested that patients with acute myocardial infarction who receive thrombolytic therapy are also more likely to receive other effective cardiac medications than patients who do not receive coronary reperfusion modalities.23, 24

Our findings, in conjunction with the results of previous studies, support the increasing enthusiasm by quality control agencies and hospitals for the development of comprehensive guidelines for the management of patients with chronic HF.3, 5, 17 These guidelines strongly recommend the use of both pharmacologic and nonpharmacologic management, emphasize close follow-up, and stress patient education and involvement in their own care. Numerous studies conducted in a variety of settings over the last decade have documented the efficacy of these multidisciplinary programs with improved use of effective therapies and significant reductions in recurrent episodes of HF, hospital readmissions, and overall health care costs.25, 26, 27

The results of our study, in a relatively typical community setting, suggest, however, that as late as 2000 the effective management of patients hospitalized with HF is more the exception than the norm. In addition, relatively few patients were started on new cardiac therapies during their index hospitalization. It must be acknowledged that physicians may be waiting for more opportune times during subsequent office visits to introduce new medications and nonpharmacologic regimens. However, a number of studies have documented an association between failure to prescribe effective therapy at the time of hospital discharge and suboptimal long-term use. We suggest that by failing to initiate therapy in more of these patients during an acute hospital admission, physicians are missing important windows of opportunity for the more effective long-term management and counseling of patients with this serious clinical syndrome. These findings clearly highlight the need for both improved patient and health care provider education to enhance patients’ functional status, reduce the need for subsequent hospitalizations, and improve the long-term prognosis of patients with HF.