Journal of Cardiac Failure Vol. 15 No.

7 2009

Predictors of Over-the-Counter Drug and Herbal Therapies

Use in Elderly Patients With Heart Failure
NANCY M. ALBERT, PhD, RN,1 LISA RATHMAN, MSN, RN,2 DONNA ROSS, MSN, RN,3 DONNA WALKER, MSN, RN,4
JAMES BENA, MS,1 SHANNON MCINTYRE, MS,1 DIANE PHILIP, MSN, RN,5 SANDRA SIEDLECKI, PhD, RN,1
RITA LOVELACE, RN,6 ANNETTE M. FOGARTY, MSN, RN,7 BARBARA MAIKUT, RN,8 AND PATRICIA ZIELINSKI, RN1
Cleveland, Ohio; Lancaster, Pennsylvania; Lakewood, Ohio; Euclid, Ohio; Toledo, Ohio; East Cleveland, Ohio; Beachwood, Ohio

ABSTRACT
Background: Over-the-counter (OTC) drug and herbal therapies (HT) may worsen heart failure or interact
with prescription medications. Frequency of and predictors for routine OTC drug and HT use are not well
studied.
Methods and Results: We examined routine use of OTC drug and HT in patients at 8 medical centers.
Medical conditions independently associated with use of OTC drugs, HT, or both were assessed using
multivariable logistic regression models. Of 374 subjects, OTC drug and HT were routinely used by
349 and 43 patients, respectively. Mean age was 69.6 6 13.1 years, 63% were male, and 81% were Cau-
casian. Common OTC drugs were antiplatelets (baby-dose aspirin), vitamins, acetaminophen, antacids,
laxatives, and calcium. The most common HT used was echinacea. History of hypercholesterolemia
was associated with higher OTC drug use (OR 4.36; 95% CI 1.60e11.87; P 5 .004); renal failure pre-
dicted less use (OR 0.09; 95% CI 0.01e0.59; P 5 .013). History of hypertension was associated with
less HT use (OR 0.47, 95% CI 0.24e0.92; P 5 .028).
Conclusions: In HF patients, routine use of OTC drugs was common, but HT use was not. OTC drugs
were used more often in patients with hypercholesterolemia and were used for a variety of reasons;
thus, routine assessment and individualized education are advocated. (J Cardiac Fail 2009;15:600e606)
Key words: Antiplatelet agents, aspirin, acetaminophen, echinacea.

Heart failure (HF) is a chronic, progressive, and dis- Literature regarding OTC drug and herbal therapy use in
abling disease that commonly affects the elderly. Nonpre- patients with HF is limited. In 1999, the first study reporting
scription over-the-counter (OTC) drugs and herbal on use of OTC drugs was published. Vitamins and minerals,
therapies are often used by people with HF to increase vi- pain relievers, herbal/health food products, antacids, and lax-
tality and decrease aches and pains. Yet patients with HF atives were most often used.1 Although reports on the use of 1
are at risk for serious and unexpected adverse effects if specific OTC drug class (ie, nonsteroidal anti-inflammatory
the delicate balance of prescription and OTC drug and agents or aspirin) or herbal therapy (ie, Hawthorne or coen-
herbal therapy benefits are disturbed by altered mechanisms zyme Q) are available in the literature, there was only 1 con-
of action or therapy interactions. temporary research-based report of herbal therapies in
patients with HF. Investigators found that about one-third
of patients with HF used complementary and alternative
From the 1Cleveland Clinic, Cleveland, OH; 2The Heart Group, Lancas- medicine.2 Additionally, in a 2002 report of Americans,
ter, PA; 3Lakewood Hospital, Lakewood, OH; 4Euclid Hospital, Euclid,
OH; 5Northwest Ohio Cardiology Consultants, Toledo, OH; 6Huron Hos- nearly 1 in every 5 adults had used an herbal therapy in the
pital, East Cleveland, OH; 7Fairview Hospital, Cleveland, OH and 8Cleve- prior 12 months, with a majority also taking prescription
land Clinic Beachwood RMP Center, Beachwood, OH. medications.3 Herbal therapies were most often used for
Manuscript received November 20, 2008; revised manuscript received
February 6, 2009; revised manuscript accepted February 10, 2009. colds, musculoskeletal conditions, and stomach/intestinal ill-
Correspondence to: Nancy Albert PhD, CCNS, CCRN, NE-BC, FAHA, nesses in uninsured people, who reported poor health, and
FCCM, Director, Nursing Research and Innovation, Nursing Institute and were between the ages of 25 to 44 years. Moreover, more
Clinical Nurse Specialist, Kaufman Center for Heart Failure, Cleveland
Clinic, 9500 Euclid Avenue, Mail code J3-4, Cleveland, OH 44195. than half of American adults who used herbal therapies along
Tel: (216) 444-7028; Fax: (216) 445-1776. E-mail: albertn@ccf.org with their prescription medications failed to inform their
No conflicts of interest and no sources of funding disclosed. health care providers.3
1071-9164/$ - see front matter
Ó 2009 Elsevier Inc. All rights reserved. The American Heart Association and the American Col-
doi:10.1016/j.cardfail.2009.02.001 lege of Cardiology do not support the use of vitamin,

600
 OTC Drugs and Herbal Therapies in Heart Failure  Albert et al 601

nutritional, or hormonal supplements in patients with HF included OTC drug class; perceived effectiveness of the agent;
because of the lack of efficacy data available.4 Awareness and reason for use, based on literature in adults3,5; patients with
of OTC drug and herbal therapy use could aid in predicting cardiovascular conditions6,7; and patients with HF.1,2 Additionally,
prescription and OTC drug interactions that may affect HF before creating the data collection form, a list of OTC drug cate-
gories, drug and herbal therapies names, and potential cardiac and
outcomes. No study has assessed patient comorbidities or
noncardiac chronic care interactions and benefits was created
sociodemographic or behavior factors to learn if alternative
based on the Physicians’ Desk Reference for Nonprescription
medicine use is associated with specific patient characteris- Drugs and Dietary Supplements8 to ensure that all important
tics. Therefore, the purpose of this multicenter, prospective, OTC drug classes and herbal agents were included. The form in-
correlational study was to answer 3 research questions. cluded a place to count ‘‘other’’ herbal therapies that might have
What is the prevalence of and types of OTC drugs and been excluded from the list of names. Before use, face and content
herbal therapies routinely used in patients with HF? Do pa- validity was assessed by 10 RN members of the principal investi-
tient characteristics predict routine usage of OTC drugs and gators’ hospital-based Research and Evidence Based Practice
herbal therapies? Is the routine use of OTC drugs or herbal Council. Data collectors all received a 1-page data collection
therapies independently associated with common chronic form and in-person or telephone training by the principal investi-
non-HF health conditions? OTC drugs were defined as pills gator to facilitate conformity in data collection.
Based on inclusion and exclusion criteria, patients were ap-
purchased at a drug or nutrition store. Herbal therapies were
proached in an ambulatory or hospital setting. The reason for the pa-
defined as a health food or product in pill or bottle formula
tient’s visit did not need to be related to HF; however, ambulatory
purchased at a drug or nutrition store. Routine use was de- settings were all cardiology practices and in the hospital setting,
fined as those taken at least once per week for the previous an admitting diagnosis of HF was used to identify potential subjects.
4 weeks or taken a minimum of 5 days in a row at least 1 After receiving verbal informed consent to complete the survey,
time during the previous 3 months before the data collec- a nurse investigator interviewed patients to collect data.
tion encounter or hospitalization.
Data Analysis
Methods Categorical and continuous data were summarized using frequen-
cies and percentages or mean and standard deviation, respectively.
Settings and Sample
To assess the relationship between categorical demographic and
Subjects were recruited from 1 large, urban tertiary health care medical history variables with OTC drug or herbal therapies use,
center with hospital and ambulatory services in Cleveland, Ohio; 6 Pearson chi-square tests or Fisher’s exact test were used. In cases
hospitals in communities surrounding Cleveland, Ohio; and ambu- in which the response levels were ordered, the Mantel-Haenszel
latory cardiology practices in Lancaster, Pennsylvania; Toledo, test was used. Two-sample t-tests were used to summarize the rela-
Ohio; and Beachwood, Ohio, after receiving Institutional Review tionship between continuous demographic measures and OTC use.
Board approval at each site. Subjects were adults age 18 years or The change in risk based on presence of a condition or for a 1-
older, had a diagnosis of HF from either systolic left ventricular unit increase in continuous measures was described using odds ra-
dysfunction or preserved left ventricular function, and a willing- tios. Ninety-five percent confidence intervals for the odds ratio
ness to complete 1 survey administered by a registered nurse via were calculated, and a significance level of 0.05 was assumed
telephone or face-to-face interview. Exclusion criteria were inabil- for the tests. Multivariable logistic regression models were fit
ity to verbalize responses to questions on the survey; recent or cur- with each outcome variable, assessing the changes in risk of use
rent psychotic episode, or violent, hostile, or disruptive behaviors; based on medical history, while adjusting for demographic or pa-
intoxication (based on subjective evidence by data collector [ie, al- tient characteristics that were statistically significant in the uni-
cohol on breath]); lethargy or obtundation; patients waiting to variable results. When the number of patients in a given model
have a surgical procedure (ie, cardiac catheterization) on the day was very small or 0, exact logistic regression models were fit to
being approached for inclusion in the study; and history of HF estimate univariate and multivariable risk.
from restrictive or hypertrophic cardiomyopathy etiologies.
Because there were no data available in the literature on relation- Results
ships between medical history and routine use of OTC drugs and
herbal therapies in patients with HF, sample size was based on sam-
A total of 374 patients treated for HF completed surveys
ple sizes of 2 contemporary studies in the HF literature that assessed
predictors of use based on patient sociodemographic and economic between April 2007 and February 2008. Of the sample,
characteristics: OTC drugs (n 5 180)1 and herbal therapies (n 5 63.1% were male, 81% were Caucasian, obesity was preva-
252)2; therefore, our minimum sample size was 300 subjects. Addi- lent (mean body mass index, 30.5 6 8.2), 98.6% had chronic
tionally, our aim was to collect data from at least 4 sites to increase HF, mean ejection fraction was 38.7 6 15.4%, and etiology
subject diversity and generalizability of findings. of HF was more often ischemic cardiomyopathy (55.5%)
than other causes. Mean age was 69.6 6 13.1 years (range,
Measure and Procedures 31e98 years), 56.3% were married, 69% completed surveys
The survey was designed by the principal investigator; a second while awaiting care in an ambulatory clinic (not acutely de-
PhD, RN; and 1 MSN, RN. It contained details about HF etiology, compensated), and 31% while hospitalized or being treated
ejection fraction, and patient characteristics; specifically, comor- in emergency care. Of subjects, 55% rated themselves as
bidities, sociodemographic, and OTC drug and herbal therapies healthier or as healthy as other people their age whom they
behavior factors. Details about OTC drug and herbal therapies knew. Other patient characteristics are provided in Table 1.
602 Journal of Cardiac Failure Vol. 15 No. 7 September 2009

Table 1. Patient Characteristics

Over-the-Counter Drugs Herbal Therapy
n (%) unless otherwise stated Total Use No Use P Value* Use No Use P Value*
Age; mean (SD) 69.6 (13.1) 70.0 (13.0) 63.5 (13.1) 0.03 67.8 (11.6) 69.8 (13.3) .29
Gender; male 232 (62.0) 215 (62.5) 17 (70.8) 0.41 30 (69.8) 202 (62.2) .33
Ethnicity; Caucasian 298 (79.7) 283 (81.1) 15 (60.0) 0.01 36 (83.7) 262 (79.2) .49
Education 0.30 .58
#High school/GED 206 (56.3) 190 (55.4) 16 (66.8) 23 (43.5) 183 (55.7)
Some college/college graduate 162 (43.7) 155 (44.6) 7 (29.2) 20 (46.5) 142 (43.2)
Marital status; married 209 (55.8) 196 (56.2) 13 (52.0) 0.69 30 (69.8) 179 (54.1) .051
Living with 272 (72.9) 255 (73.1) 17 (68.0) 0.58 31 (72.1) 241 (72.8) .92
someone/assisted living
Family income !$40,000 208 (63.8) 193 (55.6) 15 (62.5) 0.13 22 (52.4) 186 (56.6) .25
Physician type: cardiology 258 (69.9) 247 (71.8) 11 (44.0) 0.003 38 (88.4) 220 (67.5) .005
Ejection fraction; %; mean (SD) 38.7 (15.4) 38.9 (15.3) 35.0 (16.3) 0.30 36.9 (12.6) 38.9 (15.7) .35
HF etiology, ischemic 203 (55.5) 192 (56.3) 11 (44.0) 0.42 23 (56.1) 180 (55.4) .51

EF, ejection fraction; GED, general education development transcript; HF, heart failure; SD, standard deviation.
*Pearson or Mantel-Haenszel chi-square tests, Fisher’s exact test, or t-test.

Comorbid chronic medical conditions were common, many OTC drugs, and those with a history of renal failure on di-
of which were cardiovascular or related diseases (Fig. 1). alysis (P 5 .008) were more likely to be non-OTC users
Of subjects, 283 (76.1%) bought their own OTC drugs or (Table 3). In addition, there was a trend that non-insulin di-
herbal therapies and 227 (61%) were always able to afford abetics were more likely to be non-OTC drug users (OR
prescription medications. 0.46; 95% CI 0.20e1.09; P 5 .073). After multivariable re-
Of subjects, 349 of 374 routinely used OTC medications gression, patients with hypercholesterolemia were more
(93.3%). The number of OTC drugs used per patient varied likely to take OTC drugs than those without a history of hy-
widely from 0 to 10 therapies; mean use per patient was percholesterolemia (P ! .01) and patients with a history of
2.94 6 1.89. Routinely used OTC drugs by at least one- renal failure and on dialysis remained more likely to be
fourth of patients were antiplatelet agents (baby-dose aspi- non-OTC drug users (P ! .05; Table 3).
rin), vitamins, acetaminophen, and nonesodium-based ant- Herbal therapy use was low; at only 43 subjects (11.5%).
acids (Table 2). OTC users were more likely to be Range of use per patient was 0 to 5 therapies; mean use per
Caucasian (OR 2.9; 95% CI 1.2e6.6; P 5 .011), older patient was 0.19 6 0.63. Echinacea was the most frequency
(OR 1.04; 95% CI 1.00e1.07; P 5 .03), and be under the used herbal therapy (by 14%), followed by cranberry, gar-
care of cardiology specialty providers (OR 3.2; 95% CI lic, green tea, and gingko (each by 9.3% of patients; Table
1.4e7.4; P 5 .003) as compared with non-OTC users. 2), Based on patient characteristics, herbal users were more
Use of OTC drugs varied with history of medical condi- likely to be under the care of a cardiology specialty pro-
tions. Patients with a history of coronary artery disease (P vider (OR 3.7; 95% CI 1.4e9.6; P 5 .005) and married
5 .045), coronary artery bypass graft surgery (P 5 .04), (OR 2.0; 95% CI 0.99e3.9; P 5 .051). In addition, there
and high cholesterol (P ! .001) were more likely to use was a trend that herbal therapy users were more likely

80
70
60
of Subjects

50
40
30
20
10
0
itis

BG

PD

es

ia

ncy

A
lur
hm
atio

sio

ctio
CA

/TI
lem
bet
thr

CO
CA

icie
ten

Fai
Ast

oke
far
Dia
rill

ero
Ar

uff
per

nal
n
Fib

Str
lest

al I

Ins
Hy

Re
ial

cho

i
ard

nal
Atr

per

Re
oc
My
Hy

Fig. 1. Comorbid conditions. CABG, coronary artery bypass graft surgery, CAD, coronary artery disease; COPD, chronic obstructive
pulmonary disease; TIA, transient ischemic attack.
 OTC Drugs and Herbal Therapies in Heart Failure  Albert et al 603

Table 2. Use of OTC Drugs and Herbal Therapies (n 5 375) Table 3. Relationship between Medical History and OTC
Drug Use
Herbal
OTC Drugs* n (%) Therapyy n (%) Adjusted Results*
Antiplatelet agent 178 (47.6) Echinacea 6 (14.0) Factor Odds Ratio (95% CI) P Value
(low-dose aspirin)
Acetaminophen 173 (46.5) Cranberry 4 (9.3) OTC drug use
Vitamins 172 (46.0) Garlic 4 (9.3)
Antacid: not sodium based 96 (25.8) Ginkgo 4 (9.3) Coronary artery bypass graft 1.44 (0.38e5.53) .592
Calcium 60 (16.0) Green Tea 4 (9.3) surgery
Laxatives 60 (16.0) Flaxseed 3 (7.0) Coronary artery disease 1.42 (0.55e3.70) .470
Aspirin, full strength 53 (14.2) Ginger 3 (7.0) Hypercholesterolemia 4.36 (1.60e11.87) .004
Nonsteroidal anti-inflammatory 50 (13.4) Ginseng 3 (7.0) Renal failure; on dialysis 0.09 (0.01e0.59) .013
drugs Herbal therapy use
Iron 44 (11.8) Papaya 2 (4.7)
Cough/mucous expectorant 43 (11.5) Saw Palmetto 2 (4.7) Asthmay 0.23 (0.03e1.72) .151
Decongestant 31 (8.3) Chromium 1 (2.3) Depression 0.39 (0.11e1.31) .126
Picolinate Hypertension 0.47 (0.24e0.92) .028
Antidiarrheal 30 (8.0) Hawthorn 1 (2.3) OTC drug or herbal therapy use
Antihistamine 26 (7.0) Nettles 1 (2.3)
Sleep aid 19 (5.1) Senna 1 (2.3) Noneinsulin-dependent diabetes 0.25 (0.09e0.71) .009
Glucosamine 14 (3.7) St. John’s Wort 1 (2.3) Hypercholesterolemia 3.32 (1.18e9.32) .023
Salt substitute 11 (2.9) Valerian 1 (2.3) Renal failure; on dialysis 0.07 (0.01e0.48) .007
Vitamin Use Details n (%)
OTC, over-the-counter.
Multivitamins or multivitamin 135 (36.0) *Adjusted for age and physician group (cardiology vs. noncardiology) in
and minerals OTC drug and OTC drug or herbal therapy use analyses and adjusted for
Vitamin C 22 (5.9) physician group in herbal use analysis.
y
Vitamin B/B6, B12, B complex 16 (4.3) Calculated using exact logistic regression.
Vitamin E 9 (2.4)
Magnesium 9 (2.4)
Vitamin D 7 (1.9) dialysis were less likely to use non-prescription agents (P 5
Folic acid 6 (1.6)
Vitamin A 3 (!1)
.009 and .007, respectively; Table 3).
Zinc 2 (!1) Common reasons for OTC drug and herbal therapy use
varied; however, antiplatelet agents were the only OTC
*Over-the-counter drugs used less than 2% of the time: sodium-based drug recommended by doctors or nurses O50% of the
antacids, bronchodilators, cough suppressants, hair regrowth aids, alertness
aids, immune system support aids, menopause symptom suppressants, time. Table 4 provides the top 6 OTC drugs and reason
memory enhancers, motion sickness suppressants, smoking cessation for use. In general, relief of irritating or painful bodily sen-
aids, and weight control aids. sations was frequently sited as a reason for using a specific
y
49 herbal therapies (not listed) had 0% usage.
OTC drug; family and friend advice was less commonly se-
lected. Herbal therapy users most frequently thought the
therapy would promote good health or improve their med-
patients without acute decompensation requiring hospital or ical condition and, unlike OTC drugs, family and friend ad-
emergency care at the time the survey was completed (OR vice was the leading method of learning about an herbal
0.58; 95% CI 0.30e1.1; P 5 .10). Herbal therapy users therapy, followed by doctor or nurse advice (Table 4).
were less likely to have a medical history of asthma (P 5 Patient perceptions about OTC drug and herbal therapy
.043), depression (P 5 .042), and hypertension (P 5 use differed from actual behaviors in some patients. Of
.011); however, after multivariable regression, only hyper- the 54 patients (14.4%) who said they would never take
tension remained significantly associated with non-herbal OTC drugs, 38 (10.9% of total OTC users) had routinely
therapy use (P 5 .028; Table 3). taken an OTC drug, based on our definition of routine us-
Of subjects, 352 (94.1%) used OTC drugs or herbal thera- age. Of 265 (70.8%) patients that stated they would never
pies. Significant patient characteristics in OTC drug or herbal take an herbal therapy, 9 (20.9% of total herbal therapy
therapy users were Caucasian ethnicity (OR 2.9; 95% users) routinely used herbal therapies within the past 3
CI1.2e7.2; P 5 .025), older age (70.0 6 13.0 years vs. months, based on our definition of routine usage (Table 5).
62.3 6 12.3 years; P 5 .01) and being cared for by a cardiol-
ogy specialist provider (OR 3.7; 95% CI 1.5e8.9; P 5 .002). Discussion
Of medical history, OTC drug or herbal users were more
likely to have hypercholesterolemia (P ! .001) and less The results of this analysis suggest that OTC drugs are
likely to have renal failure on dialysis (P 5 .005) and none used routinely by a majority of patients and when compared
insulin-dependent diabetes (P 5 .03). After multivariable re- with previous reports of North American patients with car-
gression, patients with hypercholesterolemia were more diovascular conditions, it appears that use of OTC drugs
likely than those without a history of hypercholesterolemia may be on the rise. Use of OTC drugs or herbal therapies
to take OTC drugs or herbal therapies (P 5 0.02), and those in Alberta, Canada, was 82% in patients with HF (defined
with noneinsulin-dependent diabetes or renal failure and on as use at least 1 time per week) in 1995,1 compared with
604 Journal of Cardiac Failure Vol. 15 No. 7 September 2009

Table 4. Most Common Reasons for Frequently Used Over- Table 5. Patients’ Perception of their own OTC Drug/
the-Counter Drugs, Reasons for Using Herbal Therapy, and Herbal Therapy Use
How Patients Learned about Herbal Therapy
OTC use by Herbal Use
Drug/Therapy Reason n (%) Patients by Patients
Replied Yes n Replying Replying
Acetaminophen (n 5 173) Noncardiac aches and pains 159 (91.4) Patient Statement (%) Yes; n (%) Yes; n (%)
Cold, flu, cough, fever 11 (6.9)
Other reason 24 (13.9) I will replace it for 18 (4.8) 18 (5.2) 1 (2.3)
Family/friend advice 6 (3.5) 1 or more of my
Doctor/nurse recommendation 43 (24.9) prescriptions
Antiplatelet (n 5 178) Chest pain 1 (0.6) I will take it in addition 208 (59.9) 205 (58.9) 34 (79.1)
Heart/stroke prevention 108 (60.7) to my prescription
Family/friend advice 1 (0.6) drug
Doctor/nurse recommendation 117 (65.2) I will never take OTC 54 (14.5) 38 (10.9)
Antacid (n 5 96) Heartburn 38 (39.2) drugs,
Indigestion 42 (43.3) not even for pain
Stomach pain, gas, cramps 22 (22.9) relief
Family/friend advice 23 (23.7) I will never take herbal 265 (71.2) 9 (20.9)
Doctor/nurse recommendation 8 (8.3) Tx
Calcium (n 5 60) Osteoporosis 12 (20.0)
Preventive 33 (55.0) OTC, over-the-counter; Tx, treatment.
Other reason 7 (11.7)
Family/friend advice 4 (6.7)
Doctor/nurse recommendation 21 (35.0)
Laxative (n 5 60) Constipation 49 (83.1) 2002,3 a 32.5% use of complementary therapies during
Irregularity or hard stool 12 (20.4)
Painful bowel movement; 5 (8.5) a 6-month period in patients with HF between 2000 and
stomach cramps 20042 and a 21.3% use at a single-center HF clinic in Italy
Family/friend advice 3 (5.1) between 2004 and 2005,10 we reported a low use of only
Doctor/nurse recommendation 13 (22.0)
Vitamins (n 5 172) Overall well-being 71 (42.3) 11.5%. In the 32.5% prevalence statistic presented previ-
Promote or maintain good health 77 (44.8) ously, vitamin C was used by 17% of respondents and
Energy/vitality; decrease fatigue 22 (12.8) was the most frequently used herbal therapy, whereas, in
Prevent colds 5 (3.0)
Aging concerns 7 (4.2) our study, it was reported as an OTC drug and not included
Family/friend advice 15 (8.9) in herbal therapy use prevalence. As with OTC drugs, var-
Doctor/nurse recommendation 70 (41.9) iability in definitions of routine use and therapies studied
Herbal therapy Provide energy; decrease fatigue 10 (25.0)
(n 5 43) Overall well-being 12 (30.0) impedes comparisons of prevalence and trends in routine
Promote or maintain health 20 (46.5) use. Another explanation might be that use is based on
Aging concerns 4 (10.0) the environment of care. In American adults, herbal therapy
Improve medical condition 18 (45.0)
Other reasons for use 6 (15.0) use was lower in people living in the Northeast compared
Family/friend advice 6 (15.0) with those living in the West.3 Our sample was obtained
Doctor/nurse recommendation 2 (5.0) solely from the Midwest and Northeast areas of the United
Learned by:
States. Herbal therapy use may have been mediated by
Nutrition store worker 5 (12.5) availability, marketing, and general acceptance of use.
Family/friend 11 (27.5) Based on patient demographic characteristics, we found
Doctor or nurse 9 (22.5)
Internet 2 (5.0) routine OTC drug users to be Caucasian, older, and under
Other* 5 (12.5) the care of a cardiologist, compared with nonusers. Re-
searchers of a single-center clinic in Italy assessed patient
*All users of herbal therapies responded ‘‘no’’ to learning about them
from reading books, magazines, or bottle labels, and discussing with
characteristics and found no differences in OTC drug use
a pharmacist in a drug store or a homeopathic doctor. by gender, age, education level, or HF duration.10 When pa-
tients with HF were compared with age- and gender-
matched controls without a heart condition, there were no
our use of 96%. Use of OTC drugs in patients with cardio- differences by group except that OTC drug users were older;
vascular diseases across Canada was 67% (defined as use at however, in patients older than 74 years, OTC drug use de-
least 1 time per week over the last 6 months) between 1998 clined.1 Because study populations and characteristics var-
and 19999 and in a single-center HF clinic in Italy, OTC ied, future research is needed to learn if specific patient
drug use (not including vitamins or nutritional supple- demographic and socioeconomic status factors differ based
ments) was 75.8% (defined as having ever used OTC drugs) on use or nonuse of OTC drugs.
between 2004 and 2005,10 compared with our use of 93.3%. We found that routine herbal users were more likely
However, heterogeneity in drugs studied and definitions of married and under the care of a cardiologist. In 3 reports,
routine use prevents true comparisons. researchers assessed patient characteristics between herbal
Paradoxically, use of herbal therapies may be on the de- therapy users and nonusers and found no differences be-
cline. Compared with a 18.6% use in adults based on re- tween groups; however, marital status was assessed in
sponses to a census bureau in-home survey conducted in only 1 study 11 and physician type was not a variable in
 OTC Drugs and Herbal Therapies in Heart Failure  Albert et al 605

any study.2,10,11 In a large sample of American adults, and more than 40% of our sample had a history of atrial fi-
herbal therapy users were more likely to be Caucasian, brillation. Both garlic and ginseng have important drug in-
middle aged, have a higher education level, have higher teractions with warfarin and antiplatelet agents,
income, live on the west coast, and have seen a health pro- respectively.6,7 It might be that our subjects were well in-
fessional within 6 months of taking the survey.3 Further, in formed of specific herbal therapies they chose to use and un-
community-dwelling elderly women from Turkey, herbal derstood indications and adverse effects. In a European
therapy users had more physician visits in a 1-year period study of patients with HF, only 42% treated with the antico-
than nonusers.11 Based on findings related to health care agulant acenocoumarol and 32% treated with aspirin recog-
provider visits, herbal therapy users may be more likely nized the actions of these drugs.12 We did not assess
to have medical conditions that cause more intense sensa- knowledge of drug or herbal therapy effects and side effects
tions or symptoms, requiring more medical care or moni- or ask questions about confusion or complexity associated
toring; however, research is needed to determine if these with polypharmacy to learn if our low herbal therapy usage,
factors are truly important to discerning herbal therapy especially in patients with hypertension, was related to these
use. factors.
This study is the first to report associations between rou- Key findings of our study were that nearly 59% of patients
tine OTC drug and herbal therapy use and medical history in stated they would take an OTC drug or herbal therapy in ad-
patients with HF. Even though more than 50% of subjects dition to their HF therapies and that only 5.2% of patients
had coronary artery disease, hypertension, or hypercholes- stated they would replace an OTC drug or herbal therapy
terolemia, only hypercholesterolemia was independently as- for a prescription agent. Drugs known to be extremely
sociated with OTC drug use and no medical condition was effective in treating HF are underused, especially use of an
associated with greater likelihood of herbal therapy use. angiotensin-converting enzyme inhibitor or angiotensin re-
These results may reflect availability of OTC drugs known ceptor blocker in patients without contraindications.13,14 In
to decrease cardiovascular events in people with hypercho- patients with acute coronary syndromes, those who used
lesterolemia (ie, antiplatelet agents), and a lack of OTC complementary and alternative therapies were more likely
drugs known to be targeted toward preventing or treating to adhere to b-blocker therapy (64.4% vs. 45.9%); however,
other common chronic medical conditions. there were no differences in aspirin and statin adherence,
Some medical conditions were associated with a de- even after adjusting for demographic and clinical factors.15
creased likelihood of using therapies: history of renal fail- Education about the need to use and adhere to evidence-
ure requiring dialysis was associated with decreased OTC based HF drug therapies is imperative and should be height-
drug use; hypertension was associated with decreased ened in patients known to use OTC drug and herbal
herbal therapy use, and noneinsulin-dependent diabetes therapies. Additionally, because the relative risk of hospital-
and renal failure requiring dialysis were associated with ization is increased with utilization of nonsteroidal anti-
the combined end point: decreased OTC drug or herbal inflammatory drugs,16,17 efforts to minimize use in patients
therapies use. Because both noneinsulin-dependent diabe- with HF continue to be needed.
tes and hypertension usually require more than 1 prescrip-
tion drug therapy to achieve ideal blood glucose and blood Implications
pressure, respectively, and because renal failure is fre-
A uniform definition of routine use of OTC drugs and
quently associated with hypertension, our results may re-
herbal therapies is needed. Routine or, at minimum, sea-
ally reflect less herbal therapy use when HF is
sonal assessment of OTC drug and herbal therapies by
complicated by other chronic medical conditions that
healthcare providers is needed because use is commonly re-
use multiple prescription drugs. Simply put, it might be
lated to cold and flu symptoms or common irritating or
that herbal therapy would increase the number of overall
painful sensations or conditions. Therefore, OTC drug
drugs taken per day. In addition to the uncertainty of add-
and herbal therapy use should be a component of medica-
ing herbal therapies to regimens containing multiple pre-
tion reconciliation. Further, individualized education about
scription therapies, patients might have felt that their
OTC drug and herbal therapies is needed to minimize the
diabetes, hypertension, or renal failure were adequately
risk of new or worsening of adverse events due to interac-
controlled by prescription medications and dialysis and
tions between OTC and prescription drugs or worsening or
that herbal therapy would not further improve their gen-
new HF symptoms. Future research is needed to learn the
eral health or medical conditions.
effects OTC drug and herbal therapies may have on mask-
Our finding of decreased use of herbal therapies in pa-
ing signs or symptoms of worsening HF condition.
tients with HF and hypertension was noteworthy. There is
evidence in hypertension of some benefit with garlic, gin- Limitations
seng, and hellebore, and also evidence of potential harm.
For example, garlic can inhibit platelet aggregation, ginseng This study used a convenience sample and a survey
can cause hypertension, and hellebore can cause hypoten- format to learn about OTC drug and herbal therapy use;
sion and bradycardia.6,7 In addition, nearly one-half of our thus, interpretations of these data are limited; definitive
subjects were taking aspirin at low dose as an antiplatelet cause-and-effect relationships cannot be established.
606 Journal of Cardiac Failure Vol. 15 No. 7 September 2009

Medical history was ascertained by medical record re- 2. Zick SM, Blume A, Aaronson KD. The prevalence and pattern of com-
view and patient feedback to survey questions, both of plementary and alternative supplement use in individuals with chronic
heart failure. J Card Failure 2005;11:586e9.
which may have affected the reported prevalence rate; 3. Gardiner P, Graham R, Legedza ATR, Ahn AC, Eisenberg DM,
however, there was congruence between patient report Phillips RS. Factors associated with herbal therapy use by adults in
of a chronic medical condition and medical record docu- the United States. Altern Ther 2007;13:22e9.
mentation. When collecting data on medical history of 4. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats
depression, a distinction of depression as a symptom ver- TG, et al. ACC/AHA 2005 guideline update for the diagnosis and
management of chronic heart failure in the adult: a report of the Amer-
sus as a syndrome was not made. For all chronic medical ican College of Cardiology/American Heart Association Task Force
history conditions sited, information regarding severity, on Practice Guidelines (Writing Committee to update the 2001 Guide-
current symptoms impacting quality of life, and medica- lines for the Evaluation and Management of Heart Failure). Available
tions were not captured. OTC drug and herbal therapies at: http://www.acc.org/clinical/guidelines/failure//index.pdf.
used were reported by patients dichotomously (yes/no). 5. Kuhn MA. Herbal remedies: drug-herb interactions. Crit Care Nurse
2002;22:22e32.
Adherence rates of each identified drug or herbal therapy 6. Valli G, Giardina EV. Benefits, adverse effects and drug interactions of
class were not studied. The association of patient charac- herbal therapies with cardiovascular effects. J Am Coll Cardiol 2002;
teristics and medical history on OTC drug and herbal 39:1083e95.
therapy use may have been influenced by unmeasured 7. Miller KL, Liebowitz RS, Newby LK. Complementary and alternative
confounders. Findings reported in this study may not medicine in cardiovascular disease: a review of biologically based ap-
proaches. Am Heart J 2004;147:401e11.
apply to hospitals or ambulatory clinics that differ in 8. Murray L. (Senior Editor). Physicians’ Desk Reference for Nonpre-
patient characteristics from sites used in this study. scription Drugs and Dietary Supplements. 24th ed. Montvale, NJ:
In conclusion, most patients with HF routinely used OTC Thomson PDR: 2003.
drugs, but only a small portion routinely used herbal thera- 9. Pharand C, Ackman ML, Jackevicius CA, Paradiso-Hardy FL,
pies. Acetaminophen, aspirin (full strength and as an antipla- Pearson GJ. for the Canadian Cardiovascular Pharmacists network.
Use of OTC and herbal products in patients with cardiovascular dis-
telet agent), and vitamins were the most common routinely ease. Ann Pharmacother 2003;37:899e904.
used OTC medications. Echinacea was the most common 10. Corso ED, Bondiani AL, Zanolla L, Vassanelli C. Nurse educational
routinely used herbal therapy. For OTC drugs, herbal thera- activity on non-prescription therapies in patients with chronic heart
pies or both, patient characteristics and medical history failure. Euro J Cardiovasc Nurs 2007;6:314e20.
were associated with use. Older age, Caucasian ethnicity, 11. Gözüm S, Ünsal A. Use of herbal therapies by older, community-
dwelling women. J Adv Nurs 2004;46:171e8.
and cardiologist provider were significant univariate predic- 12. Martinez-Sellés M, Garcia Robles JA, Muñoz R, Serrano JA,
tors of OTC drug and OTC drug or herbal therapy use and car- Frades E, Nuñoa MD, et al. Pharmacological treatment in patients
diologist provider was associated with herbal therapy use. with heart failure: patients knowledge and occurrence of polyphar-
After adjusting for patient age and physician type, patients macy, alternative medicine and immunizations. Eur J Heart Fail
with hypercholesterolemia were more likely to use OTC 2004;6:219e26.
13. Fonarow GC, Abraham WT, Albert NM, Stough WG, Gheorghiade M,
drugs and patients with renal failure requiring dialysis were Greenberg BH, et al. OPTIMIZE-HF Investigators and Hospitals. As-
less likely to use OTC drugs. Herbal therapy use was not as- sociation between performance measures and clinical outcomes for
sociated with specific chronic medical conditions; however, patients hospitalized with heart failure. JAMA 2007;297:61e70.
patients with asthma, depression, or hypertension were less 14. Fonarow GC, Yancy CW, Albert NM, Curtis AB, Gattis Stough W,
likely to use herbal therapy. More than 50% of patients stated Gheorghiade M, et al. Heart failure care in the outpatient cardiology
practice setting: findings from IMPROVE-HF. Circ Heart Fail 2008;
they would take OTC drugs in addition to prescription med- 1:98e106.
ications; thus, it is important for health care providers to rou- 15. Decker C, Huddleston J, Kosiborod M, Buchanan DM, Stoner C,
tinely ask questions about OTC drug use, especially in Jones A, et al. Self-reported use of complementary and alternative
patients with HF and hypercholesterolemia. medicine in patients with previous acute coronary syndrome. Am J
Cardiol 2007;99:930e3.
16. Mamdani M, Juurlink DN, Lee DS, Tochon PA, Kopp A, Naglie G,
et al. Cyclo-oxygenase-2 inhibitors versus non-selective non-steroidal
References anti-inflammatory drugs and congestive heart failure outcomes in el-
derly patients: a populationebased cohort study. Lancet 2004;363:
1. Ackman ML, Campbell JB, Buzak KA, Tsuyuki RT, Montague TJ, 1751e6.
Teo KK. Use of nonprescription medications by patients with conges- 17. Howard PA, Delafontaine P. Nonsteroidal anti-inflammatory drugs and
tive heart failure. Ann Pharmacother 1999;33:674e9. cardiovascular risk. J Am Coll Cardiol 2004;43:519e25.