Full text of DOI:10.1093/gerona/57.2.M138

Journal of Gerontology: MEDICAL SCIENCES
Copyright 2002 by The Gerontological Society of America
2002, Vol. 57A, No. 2, M138–M143
Inappropriate Use of Digoxin in Older Hospitalized
Heart Failure Patients
Ali Ahmed,1,2,3,4 Richard M. Allman,1,3,5,6 and James F. DeLong6
1Division of Gerontology and Geriatric Medicine, Department of Medicine, School of Medicine,
2Department of Epidemiology and International Health, School of Public Health, and
3Center for Aging, University of Alabama at Birmingham.
4Section of Geriatrics, 5Birmingham/Atlanta Geriatric Research, Education, and Clinical Center, Birmingham VA
Medical Center, Alabama.
6Alabama Quality Assurance Foundation, Birmingham.
Background. Older adults are more likely to suffer from the adverse effects of digoxin. Studies have described the
inappropriate use of digoxin in various populations. The objective of this study was to determine the correlates of inap-
propriate digoxin use in older heart failure patients.
Methods. We studied older hospitalized heart failure patients with documented left ventricular (LV) function evalua-
tion and electrocardiography. Digoxin use was considered inappropriate if patients had preserved LV systolic function
(ejection fraction ꢀ40%) or if they had no atrial fibrillation (AF). We compared baseline patient characteristics by indi-
cation for digoxin and tested statistical significance using Pearson’s chi-square analysis and Student’s t tests. Using lo-
gistic regression, we determined the correlates of inappropriate use and initiation of digoxin.
Results. Subjects (N ꢁ 603) had a mean age of 79 (ꢂ7) years; 59% were women, and 18% were African American.
A total of 376 patients (62%) were discharged on digoxin, and 223 (37%) had no indication for its use. Half of the pa-
tients without an indication for digoxin received the drug. Of 132 patients without an indication and not already on
digoxin, 38 (29%) were initiated on it. After adjustment for various patient and care characteristics, prior digoxin use
(adjusted odds ratio [OR] 11.47, 95% confidence interval [CI] 5.72–23.02) and pulse ꢀ100/min (adjusted OR 2.33, 95%
CI 1.10–4.94) were associated with inappropriate digoxin use. Pulse ꢀ100/min was also associated with inappropriate
initiation of the drug (adjusted OR 2.95, 95% CI 1.28–6.78).
Conclusions. Inappropriate use of digoxin was common and was associated with prior use. Tachycardia was associ-
ated with inappropriate use and initiation. Electrocardiography and echocardiography should be performed in all older
heart failure patients. Digoxin therapy should not be initiated or continued in patients without any evidence of LV sys-
tolic dysfunction or chronic AF.
The adverse effects of digoxin are common (23–29), and the
risk of adverse effects increases with age (30,31). Digoxin
toxicity is also difficult to recognize in older patients (31)
and often occurs in the presence of a normal serum digoxin
level (32). Studies have demonstrated inappropriate use of
digoxin in older community-dwelling adults (33) and in
long-term care facilities (15). To our knowledge, no study
has examined the factors associated with inappropriate use
of digoxin in older hospitalized heart failure patients.
HE only two FDA-approved and manufacturer-recom-
mended indications for digoxin use are heart failure as-
T
sociated with left ventricular systolic dysfunction (LVSD)
and chronic atrial fibrillation (AF) (1,2). The usefulness of
digoxin in heart failure patients treated with angiotensin-
converting enzyme (ACE) inhibitors, beta-blockers, and
spironolactone, and especially in those who are asymptom-
atic on these drugs, has not yet been established (3).
Digoxin has been shown to have a deleterious effect in pa-
tients with heart failure and myocardial infarct/ischemia,
which is of concern because myocardial ischemia is the
number one cause of heart failure (4,5). There is no role of
digoxin in the management of acute (6,7) or paroxysmal AF
(8,9). Digoxin is also ineffective in controlling ventricular
rate during activities (10,11).
Heart failure and chronic AF are common and often co-
exist in old age (12–14). Digoxin is one of the most com-
monly prescribed drugs (15–22). In 1999, in the United
States about 22 million prescriptions of Lanoxin (Glaxo
Wellcome) were dispensed (ranked as the 13th most sold
drug), of which about one third were new prescriptions (21).
Most of these prescriptions were written for older patients.
METHODS
Subjects
The study subjects were fee-for-service Medicare benefi-
ciaries discharged with a principal diagnosis of heart failure
in 1994 from 11 Alabama hospitals of varying sizes (from
64 to 908 beds). The Alabama Quality Assurance Founda-
tion identified the patients through the MEDPRO (adminis-
trative claims history for Medicare Part A) files. Interna-
tional Classification of Diseases, 9th Revision, and Clinical
Modification (ICD-9-CM) codes 428 (principal discharge
diagnosis of heart failure) or 402.91 (principal discharge di-
M138

HEART FAILURE AND INAPPROPRIATE DIGOXIN USE
M139
agnosis of hypertensive heart disease with congestive heart
failure) and Diagnosis Related Codes (DRG) code 127
(heart failure) were used to identify patients. Trained study
nurses abstracted charts of patients with the above listed
codes and verified the discharge diagnoses by reviewing the
chart. We further verified the diagnosis of HF by the pres-
ence of any two of the following: history of heart failure, or
any presenting symptoms (dyspnea on exertion, orthopnea,
or paroxysmal nocturnal dyspnea), signs (jugular venous
distension, third heart sound, displaced point of maximum
cardiac impulse, or pulmonary rales), or radiographic evi-
dence (cardiomegaly, pulmonary venous congestion, or pul-
monary edema) of heart failure. Patients who did not meet
these verification criteria were excluded from the study (n ꢁ
17). We also excluded subjects without documentation of
age and those younger than 65 years (n ꢁ 29).
for those without an indication who were not receiving the
drug at the time of the admission (inappropriate initiation).
We created multiple logistic regression models to adjust for
the confounding effects of patient and care characteristics.
At first, we entered age (continuous variable), sex, and race
into the model. We also entered absence of indication for
digoxin into the model for all patients. Then, in a forward
stepwise fashion, we entered history of heart failure, being
on digoxin during admission (removed from the model for
inappropriate initiation), use of a diuretic, symptoms of dys-
pnea on exertion, orthopnea, paroxysmal nocturnal dyspnea,
lower extremity edema, angina, pulse rate 100 or more per
minute, admission systolic blood pressure of 90 mm Hg or
less (removed from the model restricted to inappropriate use
because only one patient had systolic blood pressure less
than 90 mm Hg), serum creatinine 2.5 mg/dl or more, serum
potassium 3.5 mEq/l or less, admission by a cardiologist,
and hospitalization. We categorized hospitals into small
(0–99 beds), medium (100–399 beds), and large (400ꢃ
beds) and used them as dummy variables. We used the Hos-
mer-Lemeshow goodness-of-fit test to evaluate the overall
fit of the model (35). For all analyses, a two-tailed alpha
of ꢄ.05 was used for statistical significance.
Indications for Digoxin Use
Digoxin use was considered inappropriate if there was no
evidence of LVSD or AF. Patients were considered to have
LVSD if their left ventricular (LV) ejection fraction was
less than 40% or if their LV function was described as mod-
erate to severely depressed by a contrast left ventriculogra-
phy, a multiple-gated acquisition (MUGA) radionuclide
ventriculography, or an echocardiography during the index
admission or in the past. Patients were considered to have
AF if they had AF documented by an electrocardiography
performed during the index admission. Patients who had no
documentation of current or past LV function evaluation
(n ꢁ 445) or who had no electrocardiography performed
during the current hospitalization (n ꢁ 26) were therefore
excluded from the study.
RESULTS
All patients (N ꢁ 603) met two of the four heart failure
diagnostic criteria. Three hundred nine (51%) patients had
LVSD, 152 (25%) had AF, and 81 (13%) had both. Two
hundred twenty-three (37%) patients had neither of the two
indications.
Patient Characteristics
Baseline patient characteristics are summarized in Tables
1 and 2. Patients without an indication for digoxin were
more likely to be women and less likely to have history of
heart failure and be on ACE inhibitors or digoxin before ad-
mission. They were also less likely to have cardiomyopathy
or arrhythmias and were less likely to be cared for by cardi-
ologists.
Data Abstraction
From each hospital, data from 120 randomly selected
charts (all if less than 120 were available) with a principal
discharge diagnosis of heart failure were abstracted between
January and March 1995. Data were abstracted on patient
demographics, history of heart failure, past left ventricular
(LV) function evaluation, home use of digoxin, presenting
symptoms and signs, laboratory values, chest radiograph
and electrocardiograph findings, current LV function evalu-
ation, and discharge use of digoxin. Comorbid conditions
were ascertained from the list of secondary diagnoses on the
basis of ICD-9-CM codes. The reliability and validity of the
chart abstraction were verified respectively by random re-
abstractions of 5% of the charts by another group of nurse
abstractors and expert physicians in the panel. The concor-
dance rates for both reliability and validity on key variables
were 95% or greater.
Use of Digoxin
A total of 312 (52%) patients were receiving digoxin be-
fore admission, and 376 (62%) were discharged on it. Of the
380 patients who had indications for digoxin use, 265 (70%)
were discharged on it. Of the 81 patients with both LVSD
and AF, 77% were discharged on digoxin (Figure 1). Of the
291 patients who were not receiving digoxin at the time of
hospitalization, 42% were discharged on it. Patients (62%)
receiving a daily dosage of 0.125 mg or less were older
(mean age 80 years vs 76 years for those taking higher dos-
age; p ꢄ .001), more likely to be women, and more likely to
have chronic renal insufficiency.
Analysis
The SPSS for Windows, Release 10 (34) was used to ana-
lyze the data. Baseline patient characteristics and digoxin
use were compared by indication for the drug, and statistical
significance was tested using Pearson’s chi-square analysis
and Student’s t tests. We used logistic regression analysis to
calculate crude odds ratios (OR) with 95% confidence inter-
vals (95% CI) of digoxin use for all patients (use), for those
without an indication for the drug (inappropriate use), and
Prior use of digoxin was strongly associated with dis-
charge use of digoxin (crude OR 5.86, 95% CI 4.06–8.45)
(Table 3). Dyspnea and pulse 100 beats per minute or
greater were associated with higher use of digoxin. Those
with renal insufficiency were less likely to receive digoxin.
However, hypokalemia was not associated with discharge
use of digoxin. On the multivariate analysis, prior digoxin
use, indication for digoxin, pulse ꢀ100 per minute, and se-

M140
AHMED ET AL.
Table 1. Patient Characteristics by Indication for Digoxin Use
Table 2. Admission Characteristics by Indication for Digoxin Use
No Indication Indication
No Indication Indication
for
for
for
for
All Patients
(N ꢁ 603)
Digoxin
(n ꢁ 223)
Digoxin
(n ꢁ 380) p Value
All Patients
(N ꢁ 603)
Digoxin
(n ꢁ 223)
Digoxin
(n ꢁ 380) p Value
Parameter
Parameter
Mean age, y
Female gender
78.5 (ꢂ7.4) 77.8 (7.7) 78.9 (ꢂ7.1)
357 (59%) 159 (71%) 198 (52%) ꢄ.001
.09
Symptoms
Dyspnea
543 (90%)
206 (34%)
251 (42%)
168 (28%)
41 (7%)
205 (92%) 338 (89%)
78 (35%) 128 (34%)
87 (39%) 164 (43%)
62 (28%) 106 (28%)
.238
.746
.319
.981
.289
.272
.728
African American
Nursing home admission
Historical features
Heart failure
Atrial fibrillation
Admission medications
Angiotensin-converting
enzyme inhibitors
Diuretics
108 (18%)
36 (6%)
40 (18%)
9 (4%)
68 (18%)
27 (7%)
.99
.125
Dyspnea on exertion
Orthopnea
PND
Fatigue
Angina
432 (72%) 145 (65%)
154 (26%)
288 (76%)
127 (33%) ꢄ.001
.003
12 (5%)
11 (5%)
29 (8%)
12 (3%)
27 (12%)
23 (4%)
365 (61%)
Leg swelling
Physical signs
Pulse (per min)
Systolic BP, mm Hg
Diastolic BP, mm Hg
Third heart sound
JVD
Pulmonary rales
PMI displaced
Laboratory values
BUN, mg/dl
137 (61%) 228 (60%)
223 (37%)
184 (31%)
312 (52%)
62 (28%)
75 (34%)
91 (41%)
161 (42%) ꢄ.001
109 (29%)
221 (58%) ꢄ.001
92 (ꢂ23)
148 (ꢂ30)
82 (ꢂ19)
129 (21%)
265 (44%)
422 (70%)
61 (10%)
86 (ꢂ20)
96 (ꢂ23)
ꢄ.001
.011
.104
.726
.017
.364
.319
.203
152 (ꢂ31) 146 (ꢂ29)
Digoxin
Comorbidities
80 (ꢂ19)
83 (ꢂ19)
46 (21%)
83 (22%)
Coronary artery disease 149 (25%)
Cardiomyopathies
Hypertension
Diabetes
Chronic Obstructive
47 (21%)
11 (5%)
48 (22%)
62 (28%)
102 (27%)
96 (25%) ꢄ.001
58 (15%)
88 (23%)
.113
84 (38%) 181 (48%)
161 (72%) 261 (69%)
107 (18%)
106 (18%)
150 (25%)
.051
.203
19 (8%)
42 (11%)
27 (ꢂ19)
30 (ꢂ21)
23 (ꢂ13)
ꢄ.001
Pulmonary Disease 189 (32%)
73 (33%)
24 (11%)
5 (2%)
116 (31%)
51 (13%)
11 (3%)
.572
.340
.630
Serum creatinine, mg/dl 1.5 (ꢂ0.96) 1.4 (ꢂ0.80) 1.6 (ꢂ1.0)
Serum creatinine
.004
Pneumonia
75 (12%)
16 (3%)
357 (59%) 110 (49%)
Renal Failure
Care by cardiologist
Hospital size
ꢆ2.5 mg/dl
57 (10%)
18 (8%)
39 (10%)
.359
.915
247 (65%) ꢄ.001
Serum potassium, mEq/l 4.3 (ꢂ0.66) 4.3 (ꢂ0.63) 4.3 (ꢂ0.68)
Serum potassium
ꢄ3.5 mEq/l
21 (10%)
Small (0–99 beds)
157 (26%) 70 (31%)
87 (23%)
203 (53%)
90 (24%)
.022
.660
.054
64 (11%)
43 (12%)
.446
.364
Medium (100–399 beds) 318 (53%) 115 (52%)
Large (400ꢃ beds) 128 (21%) 38 (17%)
Serum sodium, mEq/l
Chest radiograph
Cardiomegaly
139 (ꢂ5.4) 139 (ꢂ4.9) 139 (ꢂ5.7)
409 (68%)
181 (30%)
134 (60%) 275 (72%)
59 (27%) 122 (32%)
.002
.144
Pulmonary edema
Pulmonary venous
congestion
rum creatinine ꢀ2.5 mg/dl were independently associated
173 (29%)
51 (23%)
122 (32%)
.016
with discharge use of digoxin (Table 4).
Note: PND ꢁ paroxysmal nocturnal dyspnea; BP ꢁ blood pressure; JVD ꢁ
jugular venous distension; PMI ꢁ point of maximum impulse; BUN ꢁ blood
urea nitrogen.
Inappropriate Use of Digoxin
Half of the patients without an indication for digoxin use
were discharged on the drug (Figure 1). Prior digoxin use
was strongly associated with inappropriate use of the drug
(OR 10.03, 95% CI 5.30–19.00) (Table 3). After adjustment
for various patient and care characteristics, prior digoxin
use and pulse ꢀ100 per minute were associated with use of
digoxin (Table 4). There were no significant differences be-
tween the observed and expected number of events, sug-
gesting a reasonable fit of the multivariate logistic regres-
sion model (Hosmer-Lemeshow goodness-of-fit test ꢅ2
4.45, p ꢁ .814). The expected number of events in most
groups exceeded five, and none of the groups had an ex-
pected value less than 1. When we restricted our analyses to
patients with incident heart failure (N ꢁ 171), we found
similar results.
goodness-of-fit test: ꢅ2 7.98, p ꢁ .436). The expected num-
ber of events in all the groups exceeded five, and none of
the groups had an expected value less than 1.
DISCUSSION
The results of our study demonstrate that as many as half
of the older heart failure patients who had no indication for
digoxin use were discharged on the drug. For patients with-
out an indication for digoxin, being on the drug at the time
of admission was the strongest predictor of being dis-
charged on it. However, we have demonstrated that inap-
propriate initiation of digoxin therapy in older heart failure
patients was also common. Among these patients, a higher
admission pulse rate was associated with initiation of
digoxin therapy. Other investigators also demonstrated high
rates of inappropriate digoxin use (15,33,36,37).
Inappropriate Initiation of Digoxin Therapy
Among patients who were not receiving digoxin at the
time of the hospitalization (n ꢁ 312), 132 had no indication
for digoxin use. Of them, 38 (29%) received digoxin at dis-
charge. Patients with pulse rates ꢀ100 beats per minute had
almost three times higher odds of receiving a new prescrip-
tion for digoxin on discharge in the absence of any approved
indications (Table 5). After adjustment for other covariates,
this association remained unchanged. The multivariate lo-
gistic regression model was fit to data (Hosmer-Lemeshow
These findings are important for several reasons. First, al-
though LVSD and chronic AF are the only two FDA-
approved indications for use of digoxin, it is not the drug of
choice for either of these two conditions. The role of
digoxin in asymptomatic LVSD patients on optimum dos-
age of ACE inhibitors, beta-blockers, and spironolactone
and those with preserved LV systolic function has not yet
been established (3,38,39). There is also concern about its

HEART FAILURE AND INAPPROPRIATE DIGOXIN USE
M141
Table 4. Adjusted Odds Ratio (95% Confidence Interval) for Use
of Digoxin
Patients Without
Indication for
All Patients
Digoxin
Parameter
(N ꢁ 602)
(n ꢁ 223)
Age
Female gender
African American
Absence of indication for digoxin
Preadmission digoxin use
Pulse ꢆ100 beats/min
Serum creatinine ꢆ2.5 mg/dl
1.01 (0.99–1.04)
0.92 (0.61–1.38)
1.35 (0.82–2.23)
0.53 (0.36–0.80)
6.19 (4.15–9.23)
1.64 (1.08–2.49)
0.18 (0.09–0.34)
1.03 (0.98–1.07)
0.49 (0.24–1.01)
1.63 (0.72–3.69)
—
11.48 (5.72–23.02)
2.33 (1.10–4.94)
0.23 (0.06–0.88)
Note: Also adjusted for history of heart failure, diuretic use, admission
symptom of dyspnea on exertion, orthopnea, paroxysmal nocturnal dyspnea,
lower extremity edema and angina, serum potassium of 3.5 mEq/l or less, care
by cardiology, and hospitals (dummy variables for medium and large hospitals
by bed size, using small hospital as the reference category).
Figure 1. Proportion of patients with left ventricular systolic dys-
function (LVSD) and atrial fibrillation (AF), alone and together, and
number without who received digoxin on discharge (*p ꢄ .5).
demonstrated a similar association (37). Studies have dem-
onstrated a general reluctance by physicians to stop a drug
that a patient is already using (43,48), especially if the drug
is not causing any adverse side effects, if discontinuation of
the drug is believed to cause worsening of symptoms, or if it
was prescribed by another physician. This is particularly
true in the light of the publication in the preceding year
(1993) of findings of the studies that withdrawal of digoxin
resulted in worsening of heart failure symptoms (49,50).
The subjects in these two studies were heart failure patients
with LVSD. However, we observed a high rate of digoxin
use and initiation in patients without LVSD or AF. Among
patients who were inappropriately initiated on digoxin, we
found that presence of tachycardia was the only predictor of
digoxin use on discharge. Increase in pulse rate by one beat
per minute was independently associated with 3% higher
odds of inappropriate initiation of digoxin therapy (adjusted
OR 1.03, 95% CI 1.01–1.05). Digoxin is not recommended
for rate control in patients with supraventricular tachycardia
other than chronic AF (51). To our knowledge, no other
study has reported an association between tachycardia and
inappropriate initiation of therapy with digoxin.
use in heart failure patients with myocardial infarction and
coronary artery disease (4,5). Similarly, clearly, digoxin has
no role in the management of acute AF, and most studies in-
dicate that it has no role in the management of paroxysmal
AF (6–9). Second, digoxin use is associated with adverse
effects with the potential for fatal complications (29–31,40).
Older adults are also more likely to experience adverse ef-
fects of digoxin at normal serum digoxin concentrations
(32). This is particularly a concern because older adults are
the users of most of the estimated more than 20 million
digoxin prescriptions dispensed every year (21).
Our observation of a high prevalence of inappropriate use
of digoxin has several possible explanations. Inappropriate
use of drugs is not uncommon in older adults (37,41–47).
Most patients receiving digoxin are older adults who also
receive multiple other medications. The risk of inappropri-
ate use of medications increases with the increase in the to-
tal number of medications prescribed (45). Second, we have
demonstrated that prior digoxin use was the strongest pre-
dictor of its discharge use. Recently, Incalzia and colleagues
Our study has several strengths. We included patients
from hospitals of various sizes and from various geographic
locations. Data were collected by chart review with direct
Table 3. Crude Odds Ratio (95% Confidence Interval) for Use
of Digoxin
Table 5. Crude and Adjusted Odds Ratio (95% Confidence
Interval) for Initiation of Digoxin Therapy Among Patients
Without Indications for its Use
Patients Without
Indication for
All Patients
Digoxin
Crude Odds Ratio
Adjusted Odds Ratio
Parameter
(N ꢁ 603)
(n ꢁ 223)
Parameter
(95% Confidence Interval) (95% Confidence Interval)
Age
Female gender
African American
Absence of indication for digoxin
Preadmission digoxin use
Dyspnea
Pulse ꢆ100 beats/min
Serum creatinine ꢆ2.5 mg/dl
Serum potassium ꢄ3.5 mEq/l
Care by cardiologist
1.02 (1.00–1.04)
0.90 (0.64–1.26)
0.89 (0.58–1.37)
0.43 (0.31–0.61)
5.86 (4.06–8.45)
1.76 (1.03–3.00)
1.56 (1.09–2.23)
0.25 (0.14–0.44)
0.77 (0.45–1.29)
1.31 (0.94–1.84)
1.03 (1.00–1.07)
0.64 (0.35–1.14)
1.14 (0.58–2.26)
—-
10.03 (5.30–19.00)
0.99 (0.38–2.60)
1.73 (0.93–3.21)
0.26 (0.08–0.83)
0.74 (0.30–1.84)
1.17 (0.70–1.99)
Age
1.02 (0.97–1.08)
0.66 (0.29–1.49)
1.21 (0.47–3.09)
2.87 (1.27–6.50)
1.03 (0.98–1.08)
0.56 (0.24–1.34)
1.26 (0.47–3.41)
2.95 (1.28–6.78)
Female gender
African American
Pulse ꢆ100 beats/min
Note: Also adjusted for history of heart failure, admission symptom of dys-
pnea on exertion, orthopnea, paroxysmal nocturnal dyspnea, lower extremity
edema and angina, serum creatinine of 2.5 mg/dl or more, serum potassium of
3.5 mEq/l or less, care by cardiology, and medium and large size hospitals (us-
ing small hospital as the reference category).

M142
AHMED ET AL.
Center for Medicare and Medicaid Services (CMS) Disclaimer
supervision by one of the coauthors (JFD) rather than from
Medicare administrative files. Several limitations of the
study also need to be recognized. Due to the retrospective
chart review design of the study, documentation bias re-
mains a concern. This is especially true for procedures such
as echocardiography or MUGA scan performed in the past.
However, we believe any such underdocumentation of past
procedures would have been nondifferential to the indica-
tions for digoxin use. Moreover, when we restricted our anal-
yses to patients hospitalized with incident heart failure, we
found similar results. We have used admission electrocar-
diography as the criterion for AF. Events during the hospital
course, such as cardioversion or the use of medications for
the restoration and maintenance of sinus rhythm, might
have affected physician decision not to prescribe digoxin at
the time of discharge. However, we expect any such effect
to be minimal. The use of sinus-rhythm–restoring drugs in
outpatient settings was reportedly very low in the early
1990s (16). We were also not able to obtain data on pa-
tients’ functional status. Older heart failure patients, who
are more likely to have limited functional capacity, are also
less likely to be symptomatic (30,52). However, in our
study, symptoms were not associated with digoxin use, and
adjustment for symptoms did not alter the results. Similarly,
other unknown patient, family, or physician preference
might also have affected digoxin use.
The analyses upon which this publication is based were performed under
Contract Number 500-96-P60, entitled “Utilization and Quality Control
Peer Review Organization for the State of Alabama,” sponsored by the
CMS (formerly Health Care Financing Administration), Department of
Health and Human Services. The content of this publication does not neces-
sarily reflect the views or policies of the Department of Health and Human
Service, nor does mention of trade names, commercial products, or organi-
zations imply endorsement by the U.S. Government.
The authors assume full responsibility for the accuracy and complete-
ness of the ideas presented. This article is a direct result of the Health Care
Quality Improvement Program initiated by the CMS, which has encouraged
identification of quality improvement projects derived from analysis of pat-
terns of care, and therefore required no special funding on the part of this
contractor. Ideas and contributions to the author concerning experience in
engaging with issues presented are welcomed.
References
1. Digoxin products for oral use; reaffirmation of new drug status and
conditions for marketing. Federal Register [serial online]. Available
at: http://www.fda.gov/OHRMS/DOCKETS/98fr/112400c.htm. 2000;
65:70573–70575.
2. Glaxo Wellcome. Lanoxin tablets [PDR.net web site]. 2001. Available
at: http://physician.pdr.net/physician/index.htm
3. Krumholz HM, Baker DW, Ashton CM, et al. Evaluating quality of
care for patients with heart failure. Circulation. 2000;101:E122–E140.
4. Gheorghiade M, Pitt B. Digitalis Investigation Group (DIG) trial: a
stimulus for further research. Am Heart J. 1997;134:3–12.
5. Spargias KS, Hall AS, Ball SG. Safety concerns about digoxin after
acute myocardial infarction. Lancet. 1999;354:391–392.
Although the Digitalis Investigation Group (DIG) study
did not demonstrate any increased mortality (death rates of
23% for patients in both the digoxin and the placebo group;
risk ratio 0.99, 95% CI 0.76–1.28) from the use of digoxin
in heart failure patients with preserved LV systolic function
(ejection fraction ꢆ45%) (53), routine use of digoxin in this
group of patients is not recommended (38,39). In one popu-
lation-based study, use of digoxin was associated with in-
creased morbidity and mortality (54).
In conclusion, we observed a high prevalence of digoxin
use among older heart failure patients who had no approved
indication for its use. We also found a relative underutiliza-
tion of the drug for patients who might have benefited most
from the drug. Although the results of this observational
study should be interpreted with caution, the study provides
epidemiologic data about the high prevalence of inappropri-
ate use and initiation of digoxin, a drug with the potential
for serious complication in older patients. Electrocardio-
graphy and echocardiography should be performed in all
older heart failure patients, and consideration should be
given to discontinue digoxin if the absence of LVSD or
chronic AF can be established. Similarly, digoxin therapy
should not be initiated in the absence of these indications.
6. Falk RH, Knowlton AA, Bernard SA, Gotlieb NE, Battinelli NJ.
Digoxin for converting recent-onset atrial fibrillation to sinus rhythm: a
randomized, double-blinded trial. Ann Intern Med. 1987;106:503–506.
7. Jordaens L, Trouerbach J, Calle P, et al. Conversion of atrial fibrilla-
tion to sinus rhythm and rate control by digoxin in comparison to
placebo. Eur Heart J. 1997;18:643–648.
8. Rawles JM, Metcalfe MJ, Jennings K. Time of occurrence, duration,
and ventricular rate of paroxysmal atrial fibrillation: the effect of
digoxin. Br Heart J. 1990;63:225–227.
9. Galun E, Flugelman MY, Glickson M, Eliakim M. Failure of long-
term digitalization to prevent rapid ventricular response in patients
with paroxysmal atrial fibrillation. Chest. 1991;99:1038–1040.
10. Koh KK, Kwon KS, Park HB, et al. Efficacy and safety of digoxin alone
and in combination with low-dose diltiazem or betaxolol to control ven-
tricular rate in chronic atrial fibrillation. Am J Cardiol. 1995;75:88–90.
11. Farshi R, Kistner D, Sarma JS, Longmate JA, Singh BN. Ventricular
rate control in chronic atrial fibrillation during daily activity and pro-
grammed exercise: a crossover open-label study of five drug regimens.
J Am Coll Cardiol. 1999;33:304–310.
12. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial
fibrillation in adults: national implications for rhythm management
and stroke prevention: the Anticoagulation and Risk Factors In Atrial
Fibrillation (ATRIA) study. JAMA. 2001;285:2370–2375.
13. Congestive Heart Failure in the United States: A New Epidemic. Data
Facts Sheets. Bethesda, MD: National Heart, Lung, and Blood Insti-
tute; 1996.
14. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG.
Prevalence, age distribution, and gender of patients with atrial fibril-
lation: analysis and implications. Arch Intern Med. 1995;155:469–
473.
Acknowledgments
15. Aronow WS. Prevalence of appropriate and inappropriate indications
for use of digoxin in older patients at the time of admission to a nurs-
ing home. J Am Geriatr Soc. 1996;44:588–590.
Dr. Ali Ahmed is supported by a faculty training support grant from the
Southeast Center of Excellence in Geriatric Medicine. During data analysis,
he was also supported by a Hartford/American Federation for Aging Re-
search Academic Geriatric Fellowship Award.
16. Stafford RS, Robson DC, Misra B, Ruskin J, Singer DE. Rate control
and sinus rhythm maintenance in atrial fibrillation: national trends in
medication use, 1980–1996. Arch Intern Med. 1998;158:2144–2148.
17. Li-Saw-Hee FL, Lip GY. Digoxin revisited. QJM. 1998;91:259–264.
18. Hauptman PJ, Kelly RA. Digitalis. Circulation. 1999;99:1265–1270.
19. Clinical Quality Improvement Network Investigators. Contemporary
utilization of digoxin in patients with atrial fibrillation. Ann Pharma-
cother. 1999;33:289–293.
Address correspondence to Dr. Ali Ahmed, Division of Geriatric Medi-
cine, University of Alabama at Birmingham, 1530 3rd Ave South, CH-19,
Ste-219, Birmingham, AL 35294-2041. E-mail: aahmed@uab.edu
A paper based on the preliminary findings of the study was presented at
the Annual Scientific Meeting of the American Geriatric Society, May
2001, Chicago, IL.

HEART FAILURE AND INAPPROPRIATE DIGOXIN USE
M143
20. Haas GJ, Young JB. Inappropriate use of digoxin in the elderly: how
widespread is the problem and how can it be solved? Drug Saf. 1999;
20:223–230.
21. Latner A. The top 200 drugs of 1999: the more things change, the
more they stay the same. Pharmacy Times [serial online] 2000. Avail-
able at: http://www.pharmacytimes.com/.
22. Heninger MM. Commonly encountered prescription medications in
medical-legal death investigation: a guide for death investigators and
medical examiners. Am J Forensic Med Pathol. 2000;21:287–299.
23. Beller GA, Smith TW, Abelmann WH, Haber E, Hood WB. Digitalis
intoxication: a prospective clinical study with serum level correlations.
N Engl J Med. 1971;284:989–997.
24. Storstein O, Hansteen V, Hatle L, Hillestad L, Storstein L. Studies on
digitalis. XIII. A prospective study of 649 patients on maintenance
treatment with digitoxin. Am Heart J. 1977;93:434–443.
25. Marik PE, Fromm L. A case series of hospitalized patients with ele-
vated digoxin levels. Am J Med. 1998;105:110–115.
40. Hershman DL, Simonoff PA, Frishman WH, Paston F, Aronson MK.
Drug utilization in the old old and how it relates to self-perceived
health and all-cause mortality: results from the Bronx Aging Study. J
Am Geriatr Soc. 1995;43:356–360.
41. Beers MH, Fingold SF, Ouslander JG, Reuben DB, Morgenstern H,
Beck JC. Characteristics and quality of prescribing by doctors practic-
ing in nursing homes. J Am Geriatr Soc. 1993;41:802–807.
42. Willcox SM, Himmelstein DU, Woolhandler S. Inappropriate drug
prescribing for the community-dwelling elderly. JAMA. 1994;272:
292–296.
43. Britten N, Brant S, Cairns A, et al. Continued prescribing of inappro-
priate drugs in general practice. J Clin Pharm Ther. 1995;20:199–205.
44. Anderson GM, Beers MH, Kerluke K. Auditing prescription practice
using explicit criteria and computerized drug benefit claims data. J
Eval Clin Prac. 1997;3:283–294.
45. Golden AG, Preston RA, Barnett SD, Llorente M, Hamdan K, Silver-
man MA. Inappropriate medication prescribing in homebound older
adults. J Am Geriatr Soc. 1999;47:948–953.
46. Straand J, Rokstad KS. Elderly patients in general practice: diagnoses,
drugs and inappropriate prescriptions. A report from the More &
Romsdal Prescription Study. Fam Pract. 1999;16:380–388.
47. Mott DA, Meek PD. Evaluating prescriptions for the elderly: drug/age
criteria as a tool to help community pharmacists. J Am Pharm Assoc.
2000;40:417–424.
48. Physicians slow to adopt newer drugs for atrial fibrillation: study shows
tradition plays a powerful role in physician practice. [Massachusetts Gen-
eral Hospital web site]. 1998. Available at: http://www.mgh.harvard.edu/
depts/pubaffairs/releases/oct_98_drugs_for_artial_fibrilla.htm.
49. Packer M, Gheorghiade M, Young JB, et al. Withdrawal of digoxin
from patients with chronic heart failure treated with angiotensin-con-
verting-enzyme inhibitors: RADIANCE Study. N Engl J Med. 1993;
329:1–7.
26. Williamson KM, Thrasher KA, Fulton KB, et al. Digoxin toxicity: an eval-
uation in current clinical practice. Arch Intern Med. 1998;158:2444–2449.
27. Moore AR, O’Keeffe ST. Drug-induced cognitive impairment in the
elderly. Drugs Aging. 1999;15:15–28.
28. Ma G, Brady WJ, Pollack M, Chan TC. Electrocardiographic manifes-
tations: digitalis toxicity. J Emerg Med. 2001;20:145–152.
29. Giron MS, Wang HX, Bernsten C, Thorslund M, Winblad B, Fastbom
J. The appropriateness of drug use in an older nondemented and de-
mented population. J Am Geriatr Soc. 2001;49:277–283.
30. Gambassi G, Forman DE, Lapane KL, et al. Management of heart fail-
ure among very old persons living in long-term care: has the voice of
trials spread? The SAGE Study Group. Am Heart J. 2000;139:85–93.
31. Hanratty CG, McGlinchey P, Johnston GD, Passmore AP. Differential
pharmacokinetics of digoxin in elderly patients. Drugs Aging. 2000;
17:353–362.
32. Miura T, Kojima R, Sugiura Y, Mizutani M, Takatsu F, Suzuki Y. Ef-
fect of aging on the incidence of digoxin toxicity. Ann Pharmacother.
2000;34:427–432.
33. Fishkind D, Paris BE, Aronow WS. Use of digoxin, diuretics, beta
blockers, angiotensin-converting enzyme inhibitors, and calcium
channel blockers in older patients in an academic hospital-based geri-
atrics practice. J Am Geriatr Soc. 1997;45:809–812.
34. SPSS for Windows. Version 10.0. Chicago, IL: SPSS; 2000.
35. Hosmer D, Lemeshow S. Applied Logistic Regression. New York:
Wiley-Interscience; 1989.
36. Cunha UG, Barbosa MT, Paradela EM, Carvalho FG. Use of digitalis
in elderly patients admitted to a geriatric unit at a general hospital. Arq
Bras Cardiol. 1998;71:695–698.
50. Uretsky BF, Young JB, Shahidi FE, Yellen LG, Harrison MC, Jolly
MK. Randomized study assessing the effect of digoxin withdrawal in
patients with mild to moderate chronic congestive heart failure: results
of the PROVED trial. J Am Coll Cardiol. 1993;22:955–962.
51. Aronow WS. Management of the older person with atrial fibrillation. J
Am Geriatr Soc. 1999;47:740–748.
52. Kitzman D. Heart failure and cardiomyopathy. The Merck Manual
of Geriatrics [serial online]. 2000. Available at: http://www.merck.com/
pubs/mm_geriatrics/sec11/ch92.htm.
53. The Digitalis Investigation Group. The effect of digoxin on mortality
and morbidity in patients with heart failure. N Engl J Med. 1997;336:
525–533.
54. Casiglia E, Tikhonoff V, Pizziol A, et al. Should digoxin be prescribed
in elderly subjects in sinus rhythm free from heart failure? A popula-
tion-based study. Jpn Heart J. 1998;39:639–651.
37. Incalzia RA, Pedonea C, Pahorc M, et al. Reasons prompting digitalis
therapy in the acute care hospital. J Gerontol Med Sci. 2001;56A:
M361–M365.
38. Packer M, Cohn J, Abraham W, et al. The consensus recommendation
for the management of chronic heart failure. Am J Cardiol. 1999;83:
1A–38A.
39. Massie BM, Abdalla I. Heart failure in patients with preserved left
ventricular systolic function: do digitalis glycosides have a role? Prog
Cardiovasc Dis. 1998;40:357–369.
Received June 20, 2001
Accepted June 26, 2001