Cost-Effectiveness of IV-to-Oral Switch Therapy: Azithromycin vs Cefuroxime

With or Without Erythromycin for the Treatment of Community-Acquired

Pneumonia
Joseph A. Paladino, Larry D. Gudgel, Alan Forrest and Michael S. Niederman
Chest 2002;122;1271-1279
DOI: 10.1378/chest.122.4.1271

This information is current as of April 5, 2006

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.chestjournal.org/cgi/content/full/122/4/1271

CHEST is the official journal of the American College of Chest Physicians. It has been
published monthly since 1935. Copyright 2005 by the American College of Chest Physicians,
3300 Dundee Road, Northbrook IL 60062. All rights reserved. No part of this article or PDF
may be reproduced or distributed without the prior written permission of the copyright holder.
ISSN: 0012-3692.

Downloaded from www.chestjournal.org by on April 5, 2006

Cost-Effectiveness of IV-to-Oral Switch
Therapy*
Azithromycin vs Cefuroxime With or Without
Erythromycin for the Treatment of Community-
Acquired Pneumonia
Joseph A. Paladino, PharmD; Larry D. Gudgel, MBA, PharmD†;
Alan Forrest, PharmD; and Michael S. Niederman, MD, FCCP

Study objective: To conduct a cost-effectiveness analysis of IV-to-oral regimens of azithromycin vs

cefuroxime with or without erythromycin in the treatment of patients hospitalized with
community-acquired pneumonia (CAP).
Patients: Of the 268 evaluable patients enrolled into a randomized, multicenter clinical trial of adults,
266 patients had sufficient data to be included in this cost-effectiveness analysis. One hundred
thirty-six patients received azithromycin, and 130 patients received cefuroxime with or without
erythromycin.
Methods: A pharmacoeconomic analysis from the hospital provider perspective was conducted.
Health-care resource utilization was extracted from the clinical database and converted to national
reference costs. Decision analysis was used to structure and characterize outcomes. Sensitivity
analyses were performed, and statistics were applied to the cost-effectiveness ratios.
Results: The clinical success and adverse event rates and antibiotic-related length of stay were 78%,
11.8%, and 5.8 days for the azithromycin group and 75%, 20.7%, and 6.4 days for the group receiving
cefuroxime with or without erythromycin, respectively. Geometric mean treatment costs were $4,104
(95% confidence interval [CI], $3,874 to $4,334) for the azithromycin group, and $4,578 (95% CI,
$4,319 to $4,837) for the group receiving cefuroxime with or without erythromycin (p ⴝ 0.06). The
cost-effectiveness ratios were $5,265 per expected cure for the azithromycin group, and $6,145 per
expected cure for group receiving cefuroxime with or without erythromycin (p ⴝ 0.05).
Conclusions: Despite a higher per-dose purchase price, overall costs with azithromycin tended to be
lower due to decreased duration of therapy, lower preparation and administration costs, and reduced
hospital length of stay. As empiric therapy, azithromycin monotherapy was cost-effective compared to
cefuroxime with or without erythromycin for patients hospitalized with CAP who have no underlying
cardiopulmonary disease, and no risk factors for either drug-resistant pneumococci or enteric
Gram-negative pathogens. (CHEST 2002; 122:1271–1279)

Key words: azithromycin; cefuroxime; community-acquired pneumonia; pharmacoeconomics

Abbreviations: ATS ⫽ American Thoracic Society; CAP ⫽ community-acquired pneumonia; CER ⫽ cost-effectiveness
ratio; CI ⫽ confidence interval; DRG ⫽ diagnosis-related group; GMc ⫽ geometric mean cost; Ln ⫽ natural logarithm;
LOSar ⫽ antibiotic-related length of stay

*From CPL Associates LLC (Drs. Paladino and Forrest), State

University of New York at Buffalo, Buffalo, NY; Clinical Phar-
macokinetics Laboratory (Dr. Gudgel), State University of New
York at Buffalo, Buffalo; and Department of Medicine (Dr.
Niederman), State University of New York at Stony Brook, Stony
Brook, NY.
†Current affiliation Wilford Hall Medical Center, San Antonio, TX.
Presented at the Fifth International Conference on the Macro-
lides, Azolides, Streptogramins, Ketolides & Oxazolidinones,
Seville, Spain, January 2000.
The clinical trial on which this economic analysis is based was
supported by a research grant from Pfizer Pharmaceuticals;
Drs. Paladino and Niederman were principal investigators.
Dr. Paladino has received funding from Astra-Zeneca, Aventis,
D espite an array of potent antimicrobials from
which to choose, community-acquired pneumo-
nia (CAP) remains a serious illness accounting for
the sixth leading cause of death in the United States
and the number one cause of death from infectious
Bayer, Bristol, Cubist, Elan, GlaxoSmithKline, Merck, Ortho-
McNeil, Pfizer, Pharmacia, and ViroPharma. Dr. Niederman has
received funding from Astra-Zeneca, Aventis, Bayer, Bristol,
GlaxoSmithKline, Merck, Pfizer, and Pharmacia.
Correspondence to: Joseph A. Paladino, PharmD, CPL Associates
LLC, 3980 Sheridan Dr, Suite 501, Amherst, NY 14226-1727;
e-mail: paladino@cplassociates.com

www.chestjournal.org CHEST / 122 / 4 / OCTOBER, 2002 1271

Downloaded from www.chestjournal.org by on April 5, 2006

diseases.1,2 There are approximately four million
cases of CAP annually, with 20 to 50% of these
patients being hospitalized accounting for 6% of all
Medicare discharges.1,3,4 Among those hospitalized
with CAP, mortality approaches 25%, especially if
the patient requires admission to an ICU.3–9
Because there is much uncertainty in treating
pneumonia, many hospitals have instituted clinical
pathways to improve care in a cost-effective man-
ner.5 Initial therapy is necessarily empiric because
clinical and radiographic findings are typically non-
specific and identifying an etiologic pathogen is
problematic. In one third to one half of all cases, no
pathogen is identified.6 –17 In selecting initial ther-
apy, many factors must be carefully considered, such
as age, coexisting illness, smoking history, severity of
illness, and the patient setting prior to hospitaliza-
tion.6 –17 In 1993 the American Thoracic Society
(ATS) published guidelines for the initial manage-
ment of adults with CAP. These guidelines divide
CAP into four severity categories and suggest em-
piric antimicrobial regimens for each category: ATS
categories 1 and 2 concern outpatient management,
category 3 includes hospitalized patients with CAP,
while category 4 addresses hospitalized patients with
severe CAP, for whom anti-Pseudomonas coverage is
suggested.6
Cefuroxime is a second-generation cephalosporin
commonly employed in category 3 of the original
ATS guidelines for empiric treatment of patients
hospitalized with CAP.6,8 –10 A macrolide, often
erythromycin, is added when an atypical pathogen is
suspected.6,8 –11 Azithromycin is an azalide antimi-
crobial with a spectrum of activity for respiratory
pathogens similar to the combination of a second-
generation cephalosporin plus a macrolide but with-
out enteric Gram-negative coverage. Therefore
azithromycin can be an appropriate single agent for
empiric coverage for patients who are not at risk of
infections caused by either enteric Gram-negative
bacilli or anaerobes.9 –12,18 –24
Frequently, formulary and treatment decisions are
made after comparing the cost per dose or cost per
day of an antibiotic. Azithromycin costs more per
dose and more per day than either cefuroxime or
erythromycin.25 However, medication costs are only
a small portion of health-care costs, accounting for
approximately 8% of the total health-care expendi-
tures in the United States.26 A pharmacoeconomic
analysis, in contrast to a simple price comparison,
can provide a more accurate and complete descrip-
tion of the true cost of health care.27 Since phar-
macoeconomics is an outcomes-based science, deter-
mining an economic outcome requires a clinical
outcome.28 –31 This economic study was conducted to
compare the cost-effectiveness of azithromycin vs
1272
Downloaded from www.chestjournal.org by on April 5, 2006
cefuroxime with or without erythromycin for the
treatment of adult patients hospitalized with CAP.
Materials and Methods
Clinical Trial
A multicenter, parallel group, randomized, open-label, com-
parative clinical study is the basis for this economic evaluation.32
Adult inpatients with a new infiltrate on chest radiograph and a
clinical diagnosis of CAP that required treatment with IV antibi-
otics were eligible for enrollment. Key exclusion criteria were
major allergic reactions to macrolides or ␤-lactams; significant
renal, hepatic, cardiovascular, or hematologic disease; HIV infec-
tion; AIDS; metastatic tumor; septic shock; cystic fibrosis; me-
chanical ventilation; infection due to non-Haemophilus influenza
Gram-negative organisms; use of terfenadine, loratidine, or
astemizole; and female patients who were pregnant or nursing.
Patients had to be able to sign their own informed consent; no
surrogates were allowed.
Thirty-nine investigators from 36 centers enrolled 403 patients
between November 1993 and May 1995. The sample size was
based on the anticipated evaluability rate and clinical response.32
Patients were randomized 1:1 to receive azithromycin or
cefuroxime (control). Patients in the azithromycin group received
500 mg IV qd for 2 to 5 days, followed by 500 mg po qd for a total
of 7 to 10 days of therapy. Patients in the control group received
cefuroxime, 750 mg IV q8h, for 2 to 7 days, followed by
cefuroxime axetil, 500 mg po bid, for a total of 7 to 10 days of
therapy. For patients in the control group suspected of having
atypical pathogens (Mycoplasma, Legionella, or Chlamydia),
erythromycin, 500 to 1,000 mg IV q6h/500 mg po qid, for up to
21 days, could be added at the discretion of the investigator. The
decision to switch from IV to oral therapy was made by the
individual clinical investigators on the basis of the patient’s
clinical response.
Safety, clinical, microbiological, and radiographic assessments
were performed and recorded during therapy, 10 to 14 days after
therapy, and at long-term follow-up 4 to 6 weeks after therapy.
Detailed information on the conduct of the trial, demographic
descriptions of the study population, and clinical, bacteriological,
and radiologic results have been published.32 Retrospective
calculation of the pneumonia severity index scores33 were con-
ducted.32
Methods of the Economic Analysis
Since the major expense in the treatment hospitalized patients
with CAP is the hospital per diem cost,3 a cost-effectiveness
analysis from the hospital provider’s perspective was taken. Costs
were analyzed from three diverse levels.28 Level 1 (drug budget
perspective) only considers study drug-acquisition costs. Level 2
adds antibiotic-related costs, such as preparation and administra-
tion, therapeutic drug monitoring, and additional costs of re-
sources used to manage adverse events and therapeutic failures.
Level 3 adds the cost of hospitalization and all other nonprotocol-
driven resources. A key economic outcome measure is level 3 cost
at 10 to 14 days after therapy. Clinical outcome was categorized
as success (cure or improvement) or failure, as determined by the
original clinical investigators. Adverse events included in this
analysis are those determined by the clinical investigators to be
likely due to the study drug.
The clinical evaluation at 10 to 14 days after therapy is the most
pertinent outcome relative to the hospital perspective. The
Clinical Investigations
economic evaluation period, antibiotic-related length of stay
(LOSar), is used to define the duration of hospital stay attributed
to the treatment of CAP and its sequelae. With few exceptions,
LOSar was equivalent to overall length of stay. Any resources
(procedures or medications) not directly related to the inpatient
treatment of CAP were not included. In some patients, docu-
mentation of follow-up antibiotics/treatments was incomplete. A
blinded economic investigator assigned expected values to the
data in question.
Resource Utilization
Information collected by the clinical investigators on the
electronically maintained case report form included comprehen-
sive data for each patient. Length of stay, procedures performed,
medications administered, adverse events, clinical response, and
other factors were extracted and used to construct the pharmaco-
economic database. A partial listing of the procedures that were
documented includes computer axial tomography scans; bron-
choscopies; respiratory, physical, and occupational therapy; ra-
diographic studies; laboratory tests; thoracentesis; concomitant
medications, including those used to treat adverse events and
treatment failures; ventilation-perfusion scans; ECGs; ultra-
sound; nebulizer treatments; IV site changes; heat and cold packs
for phlebitis; incentive spirometry; echocardiograms; oxygen
therapy; and telemetry.
Resource Costs
The cost for procedures was obtained by applying the cost-to-
charge ratio (70.46%) to the charges of a reference hospital, and
then adjusting to a national cost basis. A published cost of IV site
changes was used.31 Since the clinical investigators were not
required to report whether the patients were admitted to a
regular floor bed or to an ICU, the cost per bed day was derived
from the weighted average time spent in each of seven intensity
levels of care for diagnosis-related group (DRG) 89 and DRG 90
(simple pneumonia with and without effusion, n ⫽ 2,187 pa-
tients) to the direct cost of supplying that care at the reference
hospital (Table 1).34 The weighted-average intensity level ac-
counts for the cost of time spent in intensive care settings. The
resulting per diem cost was indexed to the national average,34
then applied to each day of hospitalization.
The cost for azithromycin was calculated at $18.00/500 mg IV
and $10.00/500 mg oral. The cost for cefuroxime was $3.50/750
mg IV and $4.75/500 mg oral. The cost for erythromycin was
$1.75/500 mg IV and $0.15/500 mg oral. Preparation and admin-
istration costs were $7.75 per IV dose and $1.50 per oral dose.35,36
Costs for nonstudy medications were obtained from the direct
prices as listed in the RED BOOK.25
Economic Calculations
Because costs in clinical infectious diseases studies are typically
right skewed, the geometric mean was used as the measure of
central tendency. The geometric mean cost (GMc) was deter-
mined for each outcome: success or failure for each of the three
treatment alternatives (azithromycin, cefuroxime, cefuroxime
plus erythromycin). The GMc was computed as follows:
Exp.(P(S) ⫻ Ln(GMc-s) ⫹ P(F) ⫻ Ln(GMc-f), in which Exp.
signifies the exponential, P(S) and P(F) are observed outcomes of
success and failure, and Ln(GMc-s) and Ln(GMc-f) are the
natural logarithms (Lns) of the GMc for success and failure. The
GMc for the overall control group was computed by combining
the costs and probabilities for cefuroxime alone with those for
cefuroxime plus erythromycin, using an equation similar to that
above.
A cost-effectiveness ratio (CER) was calculated for each treat-
ment arm by dividing the GMc of treatment by the probability of
success for that regimen. The CER represents the cost per likeli-
hood of producing a successful outcome, and provides a meaningful
measure of both costs and outcomes. The treatment arm with the
lowest CER is considered to be the most cost-effective regimen,29,30
and is the primary outcome measure in this pharmacoeconomic
study. As a secondary economic outcome measure, the control group
was separated into monotherapy (cefuroxime alone) and combina-
tion therapy (cefuroxime plus erythromycin) subgroups, and each
was compared to the azithromycin group.
Statistical Analysis
Patient demographics at baseline were compared by contin-
gency tables or the Fisher exact test. Between-treatment com-
parisons of level 3 costs were made by constructing 95% confi-
dence intervals (CIs) about the GMcs and by Kruskal-Wallis
one-way analysis of variance. Differences in clinical response
rates were analyzed by contingency tables. All tests were two
sided, with a probability of a type-1 error of 0.05 used to

Table 1—Calculations for Cost per Bed Day*

Intensity of Care Level

Variables 1 2 3 4 5 6 7

Room-related direct (fixed and 26 26 26 26 123 123 123

variable supplies)
Indirect (depreciation) 33 33 33 33 126 126 126
Direct salaries 141 169 196 225 593 801 891
Indirect salaries 132 159 186 212 420 568 631
Total cost per bed day per intensity 332 387 441 496 1262 1618 1771
of care provided (sum)
Percentage of time† 3.09 36.81 36.94 13.93 7.03 1.76 0.44
Percentage of time times cost per 10 142 163 70 89 28 8
bed day
Calculated national average cost 510
per bed day
*Costs are presented in US dollars ($).
†Percentage of time spent in each intensity of care for CAP (DRG 89 and DRG 90) 1996 data.

www.chestjournal.org CHEST / 122 / 4 / OCTOBER, 2002 1273

Downloaded from www.chestjournal.org by on April 5, 2006

determine statistical significance. The statistical analysis was Table 2—Patient Demographics*
performed using software (SYSTAT version 7.0; SYSTAT Soft-
ware; Evanston, IL). Azithromycin Control
Cost and outcome data were further analyzed using modeling Variables (n ⫽ 136) (n ⫽ 130)
software (ADAPT II; Biomedical Simulation Resource, Univer-
Male gender 58 61
sity of Southern California; Los Angeles, CA).37 This was done so
Weight (mean, SD), kg 76.9, 18.2 76.4, 20.0
that statistical procedures could be performed on the CER,
Age (mean, SD), yr 60.5, 17.6 60.1, 17.7
which is a dual measure of costs and outcomes, each of which
White race 75.3 76.9
contain independent experimental uncertainty. Model input was
Black race 21.3 20.8
the treatment group; response to treatment (success or failure)
Asian 0.7 0
and the Ln of the Level 3 cost were the model outputs. The fitted
Other race 2.7 2.3
parameters were probabilities of success and the mean Ln (cost),
Signs and symptoms at baseline
for success and failure, for each of the three treatment paths
Cough with sputum production 86 87
(azithromycin, cefuroxime alone, cefuroxime plus erythromycin).
Rales 61 64
The derived functional parameters were the three probabilities of
Fever 58 58
failure, the GMcs of success and failure for each of the three
Abnormal respiratory rate 49 54
groups (azithromycin, cefuroxime alone, cefuroxime plus eryth-
Rhonchi 47 45
romycin) and for the combined control group (cefuroxime with or
Other 43 43
without erythromycin), and the CERs for all three groups
Concurrent disease syndromes at baseline
(azithromycin, cefuroxime alone, cefuroxime plus erythromycin)
Tobacco use 34 32
and for the combined control group (cefuroxime with or without
Hypertension 34 38
erythromycin).
COPD 32 37
Sensitivity analyses were performed to assess whether different
History of pneumonia 22 22
bed costs ($200 to $1,200/d), antibiotic prices (⫾ 50%), and
Diabetes mellitus 19 20
clinical success rates (varied to the threshold point to force equal
Emphysema 12 6
cost-effectiveness) would change the economic outcome. Hospi-
Congestive heart failure 11 13
tal per diem costs vary by region, hospital type, and size.
Asthma 10 10
Therefore, testing the cost per bed day over a wide range will
account for this. By varying key factors over reasonable ranges, *Data are presented as % unless otherwise indicated. The baseline
sensitivity analysis allows for a more feasible extrapolation of the demographics between groups were not significantly different
results to a variety of divergent clinical settings while also testing (p ⬎ 0.05).
the robustness of the economic conclusions.29,30
An agreement was made between the owner of the clinical
database (Pfizer; New York, NY) and the researchers, granting us
full access and complete freedom in the study design, analysis, sion tree combines the probabilities of success or
and publication.38 – 40 Any assumptions or blinded assessments failure with the cost of each treatment and out-
were conservative in nature, specified, and justified with appro-
priate sensitivity analysis.
come path. The results shown at the terminal end
of each branch are the GMc and LOSar for each
path. For all treatment options, successful treat-
Results ment resulted in a shorter length of stay and a
lower mean cost than did clinical failure. Compar-
Of the 268 clinically evaluable patients,32 266 ing level 3 costs by Kruskal-Wallis one-way analy-
patients were economically evaluable. Posttherapy sis of variance revealed a p ⫽ 0.06 while the 95%
data from two patients (one in each group) were CIs about the geometric mean overlap only slightly:
insufficient for a comprehensive economic analy- $3,874 to $4,334 for the azithromycin group and
sis. Thus, 136 patients in the azithromycin group $4,319 to $4,837 for the group receiving cefuroxime
and 130 patients in the control group, of which 66 with or without erythromycin. The CERs were
patients received cefuroxime alone and 64 patients $5,265 per expected successful outcome with
received cefuroxime plus erythromycin, comprise azithromycin and $6,145 per expected successful
the economic study. Baseline demographics and outcome with cefuroxime with or without erythro-
comorbidities of the patients are presented in
Table 2. There were no statistically significant
differences noted in the baseline characteristics
between the groups. Table 3—Treatment Outcomes
Clinical success rates at 10 to 14 days after therapy Azithromycin Control No.
were similar to the follow-up at 4 to 6 weeks Clinical Outcome No. (%) (%) p Value
(Table 3). No statistically significant differences in
10 to 14 d posttherapy
clinical outcome were observed between the two Cure/Improved 106 (78) 97 (75) 0.54
treatment groups at 10 to 14 days after therapy Failure 30 (22) 33 (25)
(p ⫽ 0.54) or 4 to 6 weeks after therapy (p ⫽ 0.46). 4 to 6 wk posttherapy
Geometric mean LOSar and level 3 economic Cure 98 (75) 87 (71) 0.46
Failure 32 (25) 35 (29)
analysis data are depicted in Figure 1. The deci-

1274 Clinical Investigations

Downloaded from www.chestjournal.org by on April 5, 2006

 Figure 1. CAP decision tree. Costs are presented as level 3 GMc per patient in 1998 US dollars; No.
of days indicate the geometric mean length of stay; Success ⫽ clinical cure plus clinical improvement;
Failure ⫽ clinical failure; Cef ⫽ cefuroxime; Eryth ⫽ erythromycin.

mycin (p ⫽ 0.05). Results of all three levels of costs
are shown in Table 4. Since the success rate was
greater for the least costly treatment, an incremental
cost analysis was not necessary.29
Subgroup Analysis
There were no statistically significant differences
noted in the baseline demographics between pa-
tients who received azithromycin, cefuroxime, or
cefuroxime plus erythromycin. The costs associated
with the azithromycin group and two control sub-
groups (cefuroxime alone, cefuroxime plus erythro-
mycin) are shown in Figure 2. There were overlap-
ping CIs of the GMcs at the 95% confidence level
between the azithromycin group ($4,104; 95% CI,
$3,874 to $4,334) and the cefuroxime subgroup
($4,451; 95% CI, $4,101 to $4,800; p ⬎ 0.05); the
corresponding CERs were not different (p ⫽ 0.36).
Between the azithromycin group and the cefuroxime-
plus-erythromycin subgroup ($4,713; 95% CI, $4,329
to $5,097) the CIs overlapped by only $5 (p ⫽ 0.051);
the CERs were statistically different (p ⫽ 0.04).
Adverse Events
Adverse events with economic consequences were
identified in 16 of 136 patients (11.8%) in the
azithromycin group and 28 of 130 patients (21.5%) in
the control arm consisting of 5 of 66 patients (7.6%)

Table 4 —Level 1, 2, and 3 Costs in US Dollars for Patients Treated With Azithromycin or Control Regimens

GMcs, $

Regimens Level 1 Level 2 Level 3 Success, % CER, $

Azithromycin (n ⫽ 136) 113 231 4,104 78.0 5,265

Cefuroxime with or without 110 339 4,578* 74.5 6,145†
erythromycin (n ⫽ 130)
*p ⫽ 0.06.
†p ⫽ 0.05.

www.chestjournal.org CHEST / 122 / 4 / OCTOBER, 2002 1275

Downloaded from www.chestjournal.org by on April 5, 2006

 Figure 2. Levels of cost. Level 1 (drug budget perspective) only considers study drug-acquisition costs.
Level 2 adds antibiotic-related costs: preparation, dispensing, and administration of the antibiotics, and
additional medications and therapies used to manage adverse events and therapeutic failures. Level 3
adds the cost of hospitalization and all other nonprotocol-driven resources.

in the cefuroxime-only subgroup and 23 of 64 pa-
tients (35.9%) in the cefuroxime-plus-erythromycin
subgroup. The costs of treating the adverse events
are included in the appropriate groups. While there
were more adverse events in the patients who re-
ceived erythromycin, they were relatively mild and
transient (such as nausea, diarrhea, and insertion site
pain), and did not require expensive therapies. Clin-
ical details of the adverse events have been pub-
lished.32
range, while the cost difference narrows at the lower
end. Again, costs when using azithromycin were
consistently lower.
Results obtained by varying the clinical rate of
success for each drug independently over a range of
values from 50 to 95% are depicted in a three-
dimensional plot in Figure 3. Overall, cefuroxime
with or without erythromycin would have to be
⬎ 15% more effective than azithromycin to be cost-
effective.

Sensitivity Analysis
Varying study drug-acquisition costs ⫾ 50% did
not change the overall economic decision; costs
when using azithromycin were consistently lower.
The point estimate cost per bed day was $510. When
the daily cost per hospital bed was varied between
$200/d and $1,200/d, the results become increasingly
favored toward azithromycin at the higher end of the
Discussion
The treatment regimens in this study did not differ
in efficacy32; the results are corroborated by a re-
cently published study.41 The clinical trial was com-
pleted in 1995 in accordance with the original ATS
guidelines for CAP.6 Updated and revised guidelines
from recognized authoritative bodies suggest a fluo-

1276 Clinical Investigations

Downloaded from www.chestjournal.org by on April 5, 2006


Figure 3. Sensitivity analysis on the probability of success.
Clinical success rates were varied to the threshold limit, which is
the point of equal cost-effectiveness. The differences in cost
between the two treatment groups are plotted on the diagonals.
Eighty percent of the total surface area lies beneath the zero
diagonal line where azithromycin is more cost-effective.
roquinolone as monotherapy or a combination of a
␤-lactam plus a macrolide for patients in a hospital
ward bed.42– 44 It is important to repeat that an
appropriate match of patient and empiric regimen is
of utmost importance. Knowledge of local epidemi-
ology is increasingly important with the increase of
Streptococcus pneumoniae resistance to ␤-lactams,
macrolides, fluoroquinolones, tetracyclines, and sul-
fonamides. The updated ATS guidelines delineate a
role for azithromycin, though it is limited to a
well-defined group of inpatients who have no under-
lying cardiopulmonary disease, and no risk factors for
either drug-resistant pneumococci or enteric Gram-
negative pathogens.42
Patients in each of the three study arms experi-
enced similar adverse events although at slightly
different rates. Not surprisingly, the combination of
cefuroxime plus erythromycin produced the highest
adverse event rate. The majority of adverse events
were IV site problems after IV administration and GI
problems after oral administration.
We found that patients treated with azithromy-
cin used fewer health-care resources, and that
azithromycin was cost-effective compared to the
group receiving cefuroxime with or without mac-
rolide. The GMc per patient was $474 less in the
azithromycin group than the control group
(p ⫽ 0.06 with only a $15 overlap in 95% CI), an
economically important difference. If this cost
www.chestjournal.org
Downloaded from www.chestjournal.org by on April 5, 2006
savings were applied to every appropriate patient
hospitalized with CAP in the United States, an
estimated annual savings of $400 to $700 million
would be realized.
GMc was used rather than arithmetic mean or
median values for two reasons. In concordance
with other cost-effectiveness analyses of anti-
microbial therapies,45– 48 the results are not nor-
mally distributed. Failures cost much more than
successes (Fig 1) so the data exhibit a prominent
right skew: arithmetic means are much higher than
the median values while geometric means are
more consistent and representative of the data.
Additionally, geometric means allow for the use of
more descriptive and informative statistical tests
than do median values.
The sensitivity analyses demonstrated the robust-
ness of the decision model used in this economic
analysis. Varying the prices of the study drugs by
⫾ 50% or the cost of the hospital bed between
$200/d and $1,200/d did not change the economic
decision. When varying the percentage of success-
fully treated patients, it was determined that the
cefuroxime with or without erythromycin control
group would require a clinical success rate ⬎ 15%
above that of azithromycin to be cost-effective.
Although it has a higher purchase price per dose,
azithromycin demonstrated a cost-effectiveness ad-
vantage over cefuroxime with or without erythromy-
cin. The overall decrease in costs was due to three
factors. The group receiving cefuroxime with or
without erythromycin requires three to seven paren-
teral doses or two to six oral daily doses, whereas
azithromycin is administered only once daily with
consequently lower daily preparation and adminis-
tration costs.49,50 Secondly, azithromycin is adminis-
tered for fewer days; thus, less resources are used.
Third, there was a decrease in the length of hospital
stay. There is an overall cost savings when using
azithromycin as empiric therapy for hospitalized
patients with CAP who are expected to be infected
with penicillin-susceptible S pneumoniae, Haemophi-
lus pneumoniae, Moraxella catarrhalis, Mycoplasma
pneumoniae, Legionella spp., or Chlamydia pneu-
moniae, without having underlying cardiopulmo-
nary disease or risk factors for enteric Gram-
negative pathogens.
References
1 Garibaldi RA. Epidemiology of community-acquired respira-
tory tract infections in adults: incidence, etiology, and impact.
Am J Med 1985; 78:32S–37S
2 US Department of Commerce, Bureau of the Census. Statis-
tical abstract of the United States. 104th ed. Washington, DC:
USGPO, 1984
CHEST / 122 / 4 / OCTOBER, 2002 1277
 3 Dixon RE. Economic costs of respiratory tract infections in
the United States. Am J Med 1985; 78(suppl 6B):45–51
4 Health Care Financing Administration. Health care financing
review: statistical supplement, 1995, 211
5 Clare M, Sargent D, Moxley R, et al. Reducing health care
delivery costs using clinical paths: a case study on improving
hospital profitability. J Health Care Finance 1995; 21:48 –58
6 Niederman MS, Bass JB Jr, Campbell GD, et al. Guidelines
for the initial management of adults with community-
acquired pneumonia: diagnosis, assessment of severity, and
initial antimicrobial therapy; American Thoracic Society. Am
Rev Respir Dis 1993; 148:1418 –1426
7 File TM, Tan JS, Plouffe JF. Community-acquired pneumo-
nia: what’s needed for accurate diagnosis. Postgrad Med
1996; 99:95–107
8 King DE, Pippin HJ. Community-acquired pneumonia in
adults: initial antibiotic therapy. Am Fam Physician 1997;
56:544 –549
9 Bartlett JG, Breiman RF, Mandell LA, et al. Community-
acquired pneumonia in adults: guidelines for management.
Clin Infect Dis 1998; 26:811– 838
10 Fein AM, Niederman MS. Guidelines for the initial manage-
ment of community-acquired pneumonia: savory recipe or
cookbook for disaster? Am J Respir Crit Care Med 1995;
152:1149 –1153
11 Bates JH, Campbell GD, Barron AL, et al. Microbial etiology
of acute pneumonia in hospitalized patients. Chest 1992;
101:1005–1012
12 Fang GD, Fine M, Orloff J, et al. New and emerging
etiologies for community-acquired pneumonia with implica-
tion for therapy: a prospective multicenter study of 359 cases.
Medicine (Baltimore) 1990; 69:307–316
13 Marrie TJ, Durant H, Yates L. Community-acquired pneu-
monia requiring hospitalization: five-year prospective study.
Rev Infect Dis 1989; 11:586 –599
14 Woodhead MA, Macfarlane JT, McCracken JS, et al. Pro-
spective study of the aetiology and outcome of pneumonia in
the community. Lancet 1987; 1:671– 674
15 Ortqvist A, Sterner G, Nilsson JA. Severe community-
acquired pneumonia: factors influencing need of intensive care
treatment and prognosis. Scand J Infect Dis 1985; 17:377–386
16 Torres A, Serra-Battles J, Ferrer A, et al. Severe community-
acquired pneumonia: epidemiology and prognostic factors.
Am Rev Respir Dis 1991; 144:312–318
17 Pachon J, Prados MD, Capote F, et al. Severe community-
acquired pneumonia: etiology, prognosis, and treatment. Am
Rev Respir Dis 1990; 142:369 –373
18 Ishida K, Kaku M, Irifune K, et al. In vitro and in vivo
activities of macrolides against Mycoplasma pneumoniae.
Antimicrob Agents Chemother 1994; 38:790 –798
19 Rizzato G, Montemurro L, Fraioli P, et al. Efficacy of a three
day course of azithromycin in moderately severe community-
acquired pneumonia. Eur Respir J 1995; 8:398 – 402
20 Uzun O, Hayran M, Akova M, et al. Efficacy of a three-day
course of azithromycin in the treatment of community-
acquired pneumonia: preliminary report. J Chemother 1994;
6:53–57
21 Scholwald S, Barsic B, Klinaar I, et al. Three-day azithromy-
cin compared with ten-day roxithromcyin treatment of atyp-
ical pneumonia. Scand J Infect Dis 1994; 26:706 –710
22 Roblin PM, Hammerschlag MR. Microbiologic efficacy of
azithromycin and susceptibilities to azithromycin of iso-
lates of Chlamydia pneumoniae from adults and children
with community-acquired pneumonia. Antimicrob Agents
Chemother 1998; 42:194 –196
23 Cunha BA, Ortega AM. Atypical pneumonia: extrapulmonary
1278
Downloaded from www.chestjournal.org by on April 5, 2006
clues guide the way to diagnosis. Postgrad Med 1996;
99:123–132
24 Cunha BA. Community-acquired pneumonia: cost-effective
antimicrobial therapy. Postgrad Med 1996; 99:109 –119
25 Drug Topics RED BOOK. Montvale, NJ: Medical Economics
Company, 1999
26 Bootman JL, Townsend RJ, McGhan WF. Introduction to
pharmacoeconomics. In: Bootman JL, Townsend RJ,
McGhan WF, eds. Principles of pharmacoeconomics. 2nd ed.
Cincinnati, OH: Harvey Whitney Books, 1996; 5–19
27 Dickson M, Bootman JL. Pharmacoeconomics: an interna-
tional perspective. In: Bootman JL, Townsend RJ, McGhan
WF, eds. Principles of pharmacoeconomics. 2nd ed. Cincin-
nati, OH: Harvey Whitney Books, 1996; 20 – 42
28 Paladino JA, Fell RE. Pharmacoeconomic analysis of cefm-
enoxime dual individualization in the treatment of nosocomial
pneumonia. Ann Pharmacother 1994; 28:384 –389
29 Chrischilles EA. Cost-effectiveness analysis. In: Bootman JL,
Townsend RJ, McGhan WF, eds. Principles of pharmacoeco-
nomics. 2nd ed. Cincinnati, OH: Harvey Whitney Books,
1996; 76 –101
30 Detsky AS, Naglie IJ. A clinician’s guide to cost-effectiveness
analysis. Ann Intern Med 1990; 113:147–154
31 Garrelts JC. A pharmacoeconomic model to evaluate antibi-
otic costs. Pharmacother 1994; 14:438 – 445
32 Plouffe J, Schwartz DB, Kolokathis A, et al. Clinical efficacy
of intravenous followed by oral azithromycin monotherapy in
hospitalized patients with community-acquired pneumonia.
Antimicrob Agents Chemother 2000; 44:1796 –1802
33 Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to
identify low-risk patients with community-acquired pneumo-
nia. N Engl J Med 1999; 336:243–250
34 The American Hospital Association statistics, 1995–1996.
Chicago, IL: American Hospital Association, 1996
35 Tanner DJ. Cost containment of reconstituted parenteral
antibiotics: personnel and supply costs associated with prep-
aration, dispensing, and administration. Rev Infect Dis 1984;
6(suppl 4):924 –937
36 Hatoum HT. Microcost analysis of inpatient dispensing
and administration of oral solids. Am J Hosp Pharm 1990;
47:800 – 805
37 D’Argenio DZ, Schumitzky A. ADAPT II user’s guide (re-
lease 4). Los Angeles, CA: Biomedical Simulations Resource,
University of Southern California, 1997
38 Hillman AL, Eisenberg JM, Pauly MV, et al. Avoiding bias in
the conduct and reporting of cost-effectiveness research
sponsored by pharmaceutical companies. N Engl J Med 1991;
324:1362–1365
39 Clemens K, Townsend R, Faye L, et al. Methodological and
conduct principles for pharmacoeconomic research. Phar-
macoeconomics 1995; 8:169 –174
40 Siegel JE, Torrance GW, Russell GW, et al. Guidelines for
pharmacoeconomic studies. Pharmacoeconomics 1997; 11:
159 –168
41 Vergis EN, Indorf A, File TM, et al. Azithromycin vs
cefuroxime plus erythromycin for empirical treatment of
community-acquired pneumonia in hospitalized patients.
Arch Intern Med 2000; 160:1294 –1300
42 Niederman MS, Mandell LA, Anzueto A, et al. Guidelines for
the management of adults with community-acquired pneu-
monia: diagnosis, assessment of severity, antimicrobial ther-
apy, and prevention. Am J Respir Crit Care Med 2001;
163:1730 –1754
43 Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines
for the management of community-acquired pneumonia in
adults. Clin Infect Dis 2000; 31:347–382
Clinical Investigations
44 Mandell LA, Marrie TJ, Grossman RF, et al. Canadian
guidelines for the initial management of community-acquired
pneumonia: an evidence-based update by the Canadian In-
fectious Diseases Society and the Canadian Thoracic Society.
Clin Infect Dis 2000; 31:383– 421
45 Paladino JA, Sperry H, Backes JM, et al. Clinical and
economic evaluation of oral ciprofloxacin following an abbre-
viated course of intravenous antibiotics. Am J Med 1991;
91:462– 470
46 McKinnon PS, Paladino JA, Grayson ML, et al. Cost-effec-
tiveness comparison of ampicillin/sulbactam vs imipenem/
cilastatin in the treatment of limb-threatening foot infections
in diabetic patients. Clin Infect Dis 1997; 24:57– 63
www.chestjournal.org
Downloaded from www.chestjournal.org by on April 5, 2006
47 Walters DJ, Solomkin JS, Paladino JA. Cost-effectiveness of
ciprofloxacin plus metronidazole vs imipenem/cilastatin in the
treatment of intra-abdominal infections. Pharmacoeconomics
1999; 16:551–561
48 Dresser LD, Niederman MS, Paladino JA. Cost-effectiveness
of gatifloxacin vs ceftriaxone/macrolide for the treatment of
community-acquired pneumonia. Chest 2001; 119:1439–1448
49 Smith CF, Amen R. Comparison of seven methods of pre-
paring and administering small-volume injections. Am J Hosp
Pharm 1988; 45:1896 –1901
50 Foran RM, Brett JL, Wulf PH. Evaluating the cost impact of
intravenous antibiotic dosing frequencies. DICP, 1991; 25:
546 –552
CHEST / 122 / 4 / OCTOBER, 2002 1279
Cost-Effectiveness of IV-to-Oral Switch Therapy: Azithromycin vs Cefuroxime
With or Without Erythromycin for the Treatment of Community-Acquired
Pneumonia
Joseph A. Paladino, Larry D. Gudgel, Alan Forrest and Michael S. Niederman
Chest 2002;122;1271-1279
DOI: 10.1378/chest.122.4.1271
This information is current as of April 5, 2006

Updated Information Updated information and services, including high-resolution

& Services figures, can be found at:
http://www.chestjournal.org/cgi/content/full/122/4/1271
References This article cites 42 articles, 13 of which you can access for
free at:
http://www.chestjournal.org/cgi/content/full/122/4/1271#BIB
L
Citations This article has been cited by 1 HighWire-hosted articles:
http://www.chestjournal.org/cgi/content/full/122/4/1271#other
articles
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.chestjournal.org/misc/reprints.shtml
Reprints Information about ordering reprints can be found online:
http://www.chestjournal.org/misc/reprints.shtml
Email alerting service Receive free email alerts when new articles cite this article
sign up in the box at the top right corner of the online article.
Images in PowerPoint format Figures that appear in CHEST articles can be downloaded for
teaching purposes in PowerPoint slide format. See any online
article figure for directions.

Downloaded from www.chestjournal.org by on April 5, 2006