American Journal of Infection Control 44 (2016) 1381-4

Contents lists available at ScienceDirect

American Journal of Infection Control American Journal of

Infection Control

j o u r n a l h o m e p a g e : w w w. a j i c j o u r n a l . o r g

Practice Forum

Introducing an antibiotic stewardship program in a humanitarian

surgical hospital
Naina Bhalla MD a, Nagham Hussein MD b, Maha Atari BSc b, Rasheed M. Fakhri MD b,
Chiara Lepora MD a, Nadia Walsh RN a, Sara E. Cosgrove MD c,
Richard A. Murphy MD, MPH d,*
a
Médecins Sans Frontières, Paris, France
b
Doctors Without Borders, Amman, Jordan
c
Johns Hopkins University School of Medicine, Baltimore, MD
d Division of Infectious Diseases, Los Angeles Biomedical Research Institute at Harbor–UCLA Medical Center, Torrance, CA

Key Words: Antibiotic stewardship program (ASP) implementation in humanitarian settings is a new endeavor. Doctors
Antimicrobial stewardship Without Borders/Médecins Sans Frontières introduced an ASP within a hospital in Amman, Jordan, where
antibiotic resistance patients from Iraq, Syria, and Yemen with chronic, often multidrug-resistant, infections related to
war surgery
war are managed. Antibiotics were reviewed, and real-time recommendations were made to optimize
trauma surgery
choice, dose, duration, and route by a small team. Over the first year of implementation, acceptance of
chronic osteomyelitis
reconstructive surgery the ASP’s recommendations improved. When compared with the year prior to implementation, antibi-
bacterial infections otic cost in 2014 declined considerably from approximately $252,077 (average, $21,006/month) to <$159,948
low- and middle-income countries ($13,329/month), and a reduction in use of broad-spectrum agents was observed. An ASP in a humani-
developing countries tarian surgical hospital proved acceptable and effective, reducing antibiotic expenditures and use of broad-
resource-limited settings spectrum agents.
© 2016 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier
Inc. All rights reserved.

Antibiotic resistance is growing in low- and middle-income coun- in Iraq. Prior to arrival for definitive care, the median time since initial
tries (LMICs), and antibiotic overuse is considered the major driver. injury was 19 months, and the median number of prior surgical pro-
Antibiotic stewardship programs (ASPs) aim to promote more cedures was 4 (interquartile range, 2-6). The surgical program grew
optimal antibiotic use in hospitalized patients with some provid- to include patients from Syria and Yemen.5 Existing orthopedic in-
ing real-time support for better prescribing.1 ASPs have been shown, fection was found to be common in patients at program entry, and
in high-income settings, to reduce emergence of antibiotic resis- the prevalence of multidrug-resistant pathogens was high.6,7
tance in hospitals, lower costs, and improve care quality, but few To optimize management of chronic trauma-related infections,
ASPs have been established in LMICs.2-4 a medical-surgical strategy was established, a collaboration with a
In 2006, Médecins Sans Frontières (MSF) opened a surgical local microbiology laboratory was developed, and broad-spectrum
program in Amman, Jordan, for Iraqi victims of war, with a focus on antibiotics, including glycopeptides (vancomycin), extended-
surgical management of injury that could not be definitively managed spectrum aminoglycosides (amikacin), and carbapenems (imipenem),
active against multidrug-resistant strains, were introduced for the
first time in an MSF hospital. However, in the absence of support
* Address correspondence to Richard A. Murphy, MD, Division of Infectious for optimized prescribing practices, broad-spectrum antibiotic use
Diseases, Harbor-UCLA Medical Center, 1000 W Carson St, Box 466, Torrance, CA increased, opportunities for parenteral to oral antibiotic transi-
90509.
E-mail address: rmurphy@labiomed.org (R.A. Murphy).
tions were missed, unnecessarily postsurgical antibiotic prophylaxis
Conflicts of Interest: None to report. was sometimes given, and as a result overall program costs escalated.
Additional Information: The protocols of the Médecins Sans Frontières (MSF)
Amman surgical project were approved within the framework of a formal agree-
ment between MSF, the Jordanian Red Crescent, and the Jordanian Ministry of Health. LOCAL SETTING
The present study involved the analysis of data collected for monitoring and eval-
uation and therefore satisfied the criteria for reports using routinely collected
programmatic data, set by the MSF independent Ethics Review Board in Geneva, The MSF surgical program typically admits 50 patients per month,
Switzerland. with most patients originating from Syria, Iraq, and Yemen.

0196-6553/© 2016 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajic.2016.03.036
1382 N. Bhalla et al. / American Journal of Infection Control 44 (2016) 1381-4

On average, monthly there are 140 surgeries, 120 inpatient admis- antibiotic use, review of current patients receiving parenteral an-
sions, and 800 outpatient consultations; there are approximately tibiotics with a focus on oral antibiotic transition, patient issues
220 patients in Amman under management at any given time. On affecting antibiotic management (eg, adverse drug events), and align-
project admission, all patients with suspected chronic osteomyeli- ing antibiotic strategy with planned surgical interventions. The AFP
tis (amounting to >50% of admitted patients) based on prevalent sinus recorded—for each stewardship interaction—the antibiotic recom-
tract discharge from bone or chronic nonunion undergo surgical ex- mendation, uptake recommendation, reason for rejection (if
ploration and debridement, at which time intraoperative cultures necessary), and if there was a need for adjudication by an outside
are routinely obtained. Among patients with infection at admis- specialist. For contentious or complex cases, an infectious dis-
sion, the most common pathogens were cephalosporin-resistant eases specialist was available (R.A.M.) for telemedicine consultation,
Enterobacteriaceae or methicillin-resistant Staphylococcus aureus.4 allowing some management controversies to be resolved by a spe-
An infection control program in the hospital includes an infection cialist outside of the hospital. The infectious diseases specialist also
control focal point, an infection control committee, and a commit- provided regular remote mentoring to the AFP. The activities of the
ted hospital management. Monitoring for nosocomial infections is ASP were recorded in a monthly report that was shared within the
active with a focus on surgical site infection. Cohort isolation is en- organization.
forced for patients with multidrug-resistance pathogens.
RELEVANT CHANGES
APPROACH
During the initial year of implementation, an average of 22 sur-
On October 1, 2013, MSF implemented an ASP within the frame- gical patients initiated antibiotic treatment monthly. The volume
work recommended by the Centers for Disease Control and of patients starting antibiotics as inpatients remained relatively con-
Prevention with the core elements of leadership, commitment, sistent throughout the first year of the ASP (Table 1). In the period
accountability, drug expertise, action, tracking, reporting, and from February-March, 52 patients initiated antibiotics; in April-
education.8 The main activity of the ASP team was to review new June 72 patients initiated antibiotics; in July-September 52 patients
hospital antibiotic prescriptions and make real-time recommen- initiated antibiotics; and in October-December 57 patients initi-
dations to optimize antibiotic choice, dose, duration, and route based ated antibiotics. Overall, all 233 patients requiring inpatient
on MSF treatment protocols and patient-level microbiology results. antibiotics in 2014 were reviewed during antibiotic rounds
An experienced, existing project physician was promoted to be the (Box 1).
antibiotic focal point (AFP) and led the program in collaboration with In the first year of implementation, a modification was recom-
a pharmacist, both under the supervision of the hospital manager. mended in the original antibiotic prescription in 106 of 233 patients
The aim of the AFP within the Amman project was to assure good (45%). Recommendations were accepted by surgical staff in 94 of
bug-drug match, to narrow the spectrum of parenteral antibiotics, 106 patients (88%), with the rate of acceptance increasing with each
to transition to oral antibiotics when appropriate, to consider quarter: in February-March, 18 of 23 changes (78%) were ac-
comorbidities and drug-drug interactions in decisions, and to stop cepted; in April-June, 18 of 22 changes (82%) were accepted; in
antibiotics when intraoperative cultures were sterile. July-September, 25 of 27 changes (93%) were accepted; and in
Although given protected time for reviewing prescriptions, October-December, 33 of 34 changes (97%) were accepted. Over the
liaising with surgical staff to assure implementation of recommen-
dations, and creating monthly reports, the AFP remained engaged
in clinical care as part of the overall hospital team. The pharma-
cist was as an active part of the ASP whose focus was appropriate Box 1 Lessons learned
drug dosing, review of potential drug-drug interactions, and—in
collaboration with nursing—assuring optimized antibiotic admin- • A simple antibiotic stewardship model that places one an-
istration, particularly for special patient groups. The pharmacist tibiotic focal point physician and one pharmacist at the
implemented tools for the nursing staff, including the develop- center of hospital prescribing was feasible and effective, re-
ment of tables outlining the appropriate administration of common ducing hospital antibiotic expenditures.
antibiotics, and provided regular in-service trainings. The pharma- • A nonspecialist physician can lead antibiotic stewardship in
cist also contributed knowledge of cost, current antibiotic inventory, contexts where specialists in infectious diseases are not
and could suggest the substitution of an equivalent drug when short- readily available.
ages demanded it. • Implementation of inpatient antibiotic stewardship led to
Weekly multidisciplinary antibiotic rounds provided the forum recognition of related needs, specifically to other clinical care
where the members of the ASP team could interact with each other quality improvements, including follow-up of patients re-
and with hospital stakeholders, including the infection control nurse, ceiving outpatient antibiotic enhanced therapy.
surgeons, and nursing staff. The agenda included current hospital

Table 1
Process indicators for implementation of antibiotic stewardship in a humanitarian surgical project, Amman, Jordan

Initiating Changes Changes Referred to infectious

antibiotic, suggested, accepted, diseases specialist,
Time period n n (%) n (%) Reason for rejection n (%)
February-March 2013 52 23 (44) 18 (78) Presence of internal hardware; patient allergies/previous reported intolerance 7 (13)
to suggested antibiotic; reason cannot be explained or classified
April-June 2013 72 22 (31) 18 (82) High clinical suspicion/elevated CRP; patient nearing end of planned therapy 10 (14)
(2); reason cannot be explained or classified (2)
July-September 2013 52 27 (52) 25 (93) High clinical suspicion/elevated CRP (2) 11 (21)
October-December 2013 57 34 (60) 33 (97) Patient allergies/previous reported intolerance to suggested antibiotic 8 (14)

CRP, C-reactive protein.

 N. Bhalla et al. / American Journal of Infection Control 44 (2016) 1381-4 1383

Fig 1. Monthly total antibiotic costs (U.S. dollars) after stewardship program implementation and use of selected antibiotics. Vertical arrow indicates timing of introduc-
tion of the antibiotic stewardship program.

entire period, 36 of 233 patients (15%) were referred for telemedicine from 78% to 97%, suggesting increasing acceptability of the ASP. The
consult with the infectious diseases specialist. approach, premised on real-time active review of antibiotic pre-
Despite a similar program volume (2013: N = 563; mean ad- scriptions and using existing physician and pharmacist human
missions, 106 per month; 2014: N = 533; mean admissions, 104 per resources, may be a useful model for initiating an ASP in hospitals
month), when compared with the year prior to ASP implementa- in LMICs with high levels of antibiotic utilization.
tion, project antibiotic use in 2014 declined considerably. Despite Several additional changes were associated with ASP introduc-
a similar program volume, when compared with the year prior to tion. We observed a shift in the previous practice of antibiotic initiation
implementation, project antibiotic cost in 2014 declined from ap- or modification by the surgical staff without clear justification and
proximately $252,077 (average, $21,006/month) to <$159,948 observed a significant decrease in the use of empirical antibiotics
($13,329/month). Most of the costs savings were achieved with a before culture reporting. An initial reluctance expressed by project
reduction in the inappropriate use of imipenem and the surgeons to share responsibility for antibiotic management with the
substitution of another—more narrow-spectrum—parenteral or oral stewardship team improved markedly over the first year as evi-
agent. There were no significant differences in basic patient char- denced by the improvement in the recommendation acceptance rate.
acteristics and case mixture between the 2 periods. This may have stemmed from the approach of the stewardship team
A major goal of the ASP was to reduce unnecessary use of costly toward transparent, collaborative decision-making and the intro-
parenteral antibiotics, including carbapenems, glycopeptides, and duction of a new clinic for follow-up of patients receiving long-
broad-spectrum aminoglycosides, when an alternative antibiotic was term antimicrobials where treatment failure and toxicity concerns
appropriate. In the period after introduction of the ASP, use of were addressed. This improvement in decision transparency and ad-
imipenem declined from a mean of 2,206 vials per month to 1,620 ditional outpatient oversight of surgical patients receiving oral
vials per month, use of vancomycin declined from a mean of antibiotics may also have strengthened the confidence of the sur-
545 vials per month to 438 vials per month, and use amikacin de- geons in the ASP. The ASP led to the recognition of related needs in
clined from a mean of 242 vials per month to 164 vials per month the project, including the need to train more junior physicians on
(Fig 1). appropriate antibiotic use and to make greater use of pharmacy ex-
With respect to the cost of the ASP itself, the ASP used existing pertise toward improving patient care in the hospital.
human resources. Although the monthly compensation of the phy- This study has limitations. The rollout of the ASP was not part
sician leader was increased by approximately 15%, the ASP did not of a research study, but it was a pragmatic response to the need to
require the addition of full-time positions. Mortality data for the improve the use of antibiotics within a humanitarian surgical hos-
Amman surgical project were reviewed to explore if the introduc- pital; some process and outcome data are not available. Further, it
tion of antibiotic stewardship was associated with a change in is not possible to fully associate the reduction in antibiotics costs
inpatient mortality. In the year prior to ASP implementation (October over the period of interest to the impact of the ASP alone because
1, 2012-September 30, 2013), inpatient mortality was observed in there were concomitant improvements in antibiotic prices, includ-
2 of 1,268 total admitted patients (0.2%), and in the year after, mor- ing for imipenem-cilastatin. However, the overall volume of broad-
tality occurred in 0 of 1,121 patients (0%). spectrum antibiotic use also declined, which would not be expected
from price shifts alone. Prospective studies will better define the
DISCUSSION/CONCLUSIONS impact of an ASP in LMIC hospitals.
The management of chronic, frequently multidrug-resistant
A simple ASP in a humanitarian surgical project proved to be fea- chronic osteomyelitis with surgery and long-term antibiotic therapy
sible, well-accepted, and effective in markedly reducing antibiotic is expensive. The Amman reconstructive surgery project is among
expenditures and reducing use of broad-spectrum antibiotics. Over the most costly in the MSF portfolio. As a result, the implementa-
time, adoption of the recommendations of a nonspecialist physician— tion of an ASP was seen as a way to improve the quality of care and
who was recognized locally as an expert—by surgical staff improved to optimize MSF resources in contexts where the cost of medical
1384 N. Bhalla et al. / American Journal of Infection Control 44 (2016) 1381-4

care is relatively higher than in traditional MSF settings. In very varied 2. Davey P, Brown E, Charani E, Fenelon L, Gould IM, Holmes A, et al. Interventions
to improve antibiotic prescribing practices for hospital inpatients. Cochrane
humanitarian environments, ASPs are being piloted in MSF hospi-
Database Syst Rev 2013;(4):CD003543.
tals in African contexts with flexibility to account for differences in 3. Del Arco A, Tortajada B, de la Torre J, Olalla J, Prada JL, Fernández F, et al. The impact
human resources available, variations in how antibiotics are over- of an antimicrobial stewardship programme on the use of antimicrobials and the
used or misused, and availability of microbiology laboratory support. evolution of drug resistance. Eur J Clin Microbiol Infect Dis 2015;34:247-51.
4. Boyles TH, Whitelaw A, Bamford C, Moodley M, Bonorchis K, Morris V, et al.
Antibiotic stewardship ward rounds and a dedicated prescription chart reduce
antibiotic consumption and pharmacy costs without affecting inpatient mortality
Acknowledgments or re-admission rates. PLoS ONE 2013;8:e79747.
5. Fakri RM, Al Ani AM, Rose AM, Alras MS, Daumas L, Baron E, et al. Reconstruction
of nonunion tibial fractures in war-wounded Iraqi civilians, 2006-2008: better
We thank all those who supported the initiation and the stew- late than never. J Orthop Trauma 2012;26:e76-82.
ardship team in MSF hospital in Amman, Jordan, for their support, 6. Murphy RA, Ronat JB, Fakhri RM, Herard P, Blackwell N, Abgrall S, et al. Multidrug-
particularly the nursing and surgical staff. resistant chronic osteomyelitis complicating war injury in Iraqi civilians. J Trauma
2011;71:252-4.
7. Teicher CL, Ronat JB, Fakhri RM, Basel M, Labar AS, Herard P, et al. Antimicrobial
drug-resistant bacteria isolated from Syrian war-injured patients, August
References 2011-March 2013. Emerg Infect Dis 2014;20:1949-51.
8. Pollack LA, Srinivasan A. Core elements of hospital antibiotic stewardship
1. Tamma PD, Cosgrove SE. Antimicrobial stewardship. Infect Dis Clin North Am programs from the Centers for Disease Control and Prevention. Clin Infect Dis
2011;25:245-60. 2014;59(Suppl):S97-100.

Neglected disease in mentally ill patients:

Major tuberculosis outbreak in a psychiatric hospital

Molecular detection of Clostridium difficile on inert

surfaces from a Costa Rican hospital during and
after an outbreak

Inappropriate Urinary Catheter Reinsertion in

Hospitalized Older Patients